Sage Journals: Discover world-class research

Abstract

The complexity of cancer has long challenged the medical community, driving the need for improved early detection and treatment. Artificial intelligence (AI) has profoundly impacted oncology research in recent decades, resulting in innovative diagnostic and therapeutic approaches. This review synthesizes the critical applications of AI in oncology, focusing on 4 key areas: medical imaging, digital pathology, robotic surgery, and drug discovery. We highlight the role of AI in cancer diagnosis and treatment by reviewing key studies and machine learning methods, and we address the field’s current technical and ethical challenges. AI models have significantly enhanced the accuracy of medical imaging by efficiently detecting lesions and disease sites, leading to earlier and more precise diagnoses. In digital pathology, AI tools aid in risk prediction and facilitate the examination of extensive tissue sample sets for patterns and markers, simplifying the pathologists’ tasks. AI-powered robotic surgery provides different levels of automation, leading to precise and minimally invasive procedures that not only improve surgical outcomes but also lower readmission rates, hospital stays, and infection risks. Moreover, AI expedites the process of discovering cancer therapies by identifying potential lead compounds, predicting drug reactions, and repurposing current medications. In the past decade, several AI-developed drugs have successfully entered clinical trials. These significant advancements underscore the expanding role of AI in shaping the future of cancer diagnosis and treatment. Although standardization, transparency, and equitable implementation must be addressed, AI brings hope for more personalized and effective therapies.

Keywords

artificial intelligence cancer diagnosis treatment precision medicine

Graphical Abstract

This review examines the role of AI in cancer care, focusing on its applications in medical imaging, digital pathology, robotic surgery, and drug discovery.

Introduction to AI in the Field of Medicine

The integration of artificial intelligence (AI) in medicine marks a significant shift toward more accurate, personalized, and efficient healthcare practices, especially in cancer care.¹ AI refers to machines that perform tasks traditionally requiring human intelligence, such as understanding language, recognizing patterns, solving problems, learning from data, and improving performance over time. Key AI techniques include machine learning (ML) and deep learning (DL). ML uses algorithms to identify patterns and make predictions, while DL, a subset of ML, employs multi-layered neural networks inspired by the human brain, such as convolutional neural networks (CNNs). Other essential ML techniques include support vector machines (SVMs), decision trees, and K-means clustering algorithms.²

In healthcare, AI is applied in diagnostic algorithms, treatment recommendation systems, patient monitoring, and care management tools.³ AI can process and analyze vast amounts of healthcare data more efficiently than humans, leading to improved diagnostic accuracy, therapeutic interventions, and patient outcomes.⁴

AI-Powered Precision Medicine: Advancing Personalized Treatment

Precision medicine categorizes patients in clinical trials utilizing personal data.⁵ The objective is to enhance efficacy and safety outcomes, ultimately increasing the likelihood of clinical success and drug approval.⁶ This approach acknowledges that individuals have unique genetic, molecular, and clinical characteristics influencing their treatment responses. Precision medicine aims to enhance treatment outcomes and minimize side effects by tailoring therapies to these individual differences.⁷

AI is vital in precision medicine, as seen in the genomic profiling of tumors that assists in making targeted therapy decisions for patients with breast or lung cancer.⁸ AI-driven tools can also facilitate the rapid and accurate interpretation of genomic data, providing real-time recommendations for personalized treatments.⁹

AI has been instrumental in genetic analysis by identifying transcription start sites, modeling regulatory elements, and accurately predicting gene expression from genotype data. These developments are crucial in understanding the relationship between genomic variations, disease presentation, treatment efficacy, and prognosis.^10,11 In medulloblastoma, AI analysis of numerous exomes has enabled the administration of precise and optimal treatments for pediatric patients.¹² AI-driven precision medicine holds the potential to change cancer care by offering more accurate diagnoses, predicting disease risks before symptoms manifest, and designing tailored treatment plans that prioritize safety and efficiency.¹³ Integrating AI into clinical practice is expected to become more widespread as research advances, leading to even more significant advancements in cancer care.¹⁴ Figure 1 illustrates the process of precision oncology.¹⁵

Figure 1.

The precision oncology process. Where AI plays a crucial role in each stage.¹⁶ Firstly, AI will analyze patient data, including molecular tests and medical images.¹⁷ Secondly, AI algorithms scrutinize vast amounts of molecular data to identify genetic mutations and tumor biomarkers.¹⁸ Thirdly, AI stores and uncovers new insights about personalized data.¹⁹ Fourthly, AI algorithms create predictive calculators to determine disease risk and recommend the most effective treatment targets and practices.²⁰ Finally, AI can access medical records to aid in monitoring patients.²¹

The Historical Background of AI in Cancer Research

The exploration of AI in cancer research began in the 1970s, with early attempts at computer-aided diagnosis and the development of expert systems. These early AI programs primarily concentrated on diagnosing blood infections, which ultimately laid the groundwork for future medical applications.²² Over the years, advancements in computational power, complex algorithms, and the availability of extensive biomedical datasets have propelled AI from essential pattern recognition to sophisticated ML and DL.²³ These models can identify subtle diagnostic signals in imaging, genomic, and clinical data, significantly advancing cancer diagnosis, prognosis, and treatment planning.²⁴

This study aims to review recent advancements and provide future directions for AI-driven oncology, focusing on 4 major fields: medical imaging, pathology, surgery, and drug discovery.

Methodology

Protocol Registration and PRISMA Adherence

This article comprehensively analyzes AI applications in oncology using a structured methodology incorporating PRISMA guidelines. This approach minimizes bias in literature selection and ensures rigorous identification of trends, gaps, and innovations in AI and oncology. To enhance transparency and reproducibility, this review has been prospectively registered on PROSPERO (CRD 42023384772). The methodology strictly followed the PRISMA 2020 checklist and its 2021 extension for search reporting, with detailed documentation of search strategies, study selection, and data synthesis.

Search Strategy

We searched 7 databases: Wiley, Taylor & Francis, Scopus, ScienceDirect, Sage, Google Scholar, and Springer. Additionally, citation tracking was performed in Google Scholar. The search covered January 2013 to March 2024 to capture AI advancements over the past decade. We combined keywords using the following Boolean operators:

AI/ML terms: “artificial intelligence,” “machine learning,” “deep learning,” “neural networks,” “predictive analytics”

Oncology terms: “oncology,” “cancer,” “tumor,” “malignancy,” “carcinoma”

Application terms: “diagnosis,” “prognosis,” “treatment planning,” “radiomics,” “genomics”

We expanded the search methodology to include gray literature and clinical trial registries, such as ClinicalTrials.gov and WHO ICTRP, following best review practices. We employed the PICO framework to refine search terms, focusing on population (cancer patients), intervention (AI applications), comparison (traditional methods), and outcomes (diagnostic accuracy and treatment efficacy).

Inclusion and Exclusion Criteria

We removed 85 duplicate and incomplete texts from an initial pool of 275 documents before screening, leaving 192 records for title and abstract screening. Inclusion criteria prioritized studies focused on AI applications in oncology, including diagnosis, treatment, and drug discovery. Only studies with full-text availability in English were considered. Exclusion criteria removed non-oncology studies (eg, AI in cardiology), non-peer-reviewed articles, conference abstracts, research lacking sufficient technical or clinical detail, and non-human studies. Duplicates were identified and removed using EndNote’s automated deduplication, followed by manual verification. Studies published before 2013 were excluded to focus on recent advances. This rigorous process ensured a relevant and high-quality dataset.

Study Screening and Selection

Two independent reviewers conducted title and abstract screening and full-text review using Rayyan’s AI-assisted platform. Disagreements were resolved by consensus. Inter-rater reliability was calculated using Cohen’s kappa coefficient (κ = 0.82), indicating strong agreement. After screening, 31 records were excluded for irrelevance at the title and abstract stage, leaving 185 full-text articles for eligibility assessment. After a full-text review, 14 additional studies were excluded for irrelevance or insufficient methodological detail, resulting in a final set of 171 studies included in this review. Figure 2 presents the PRISMA 2020 flow diagram illustrating the detailed study selection process.

Figure 2.

Overview of the review process. PRISMA 2020 flow diagram detailing the identification, screening, eligibility assessment, and inclusion of studies, with reasons for exclusions at each stage.

Data Extraction and Synthesis

We used a standardized template to extract data from the 171 included studies, focusing on the following areas:

AI methodologies (e.g., model architectures, datasets, validation metrics such as AUC, sensitivity, and specificity)

Clinical applications (e.g., diagnostic accuracy, treatment outcomes, workflow improvements)

Ethical considerations (e.g., bias mitigation strategies, interpretability, regulatory compliance)

The collected data were organized by themes to find common patterns, like how deep learning techniques are mainly used in radiology, and to point out areas that need improvement, such as the lack of AI tools being used in actual clinical practice. A Supplemental Table summarizing key references and their insights is provided in Table 1 and Figure 3a to enhance transparency and reproducibility.

Table 1.

Key References on the Impact of AI in Healthcare and Oncology.

Focus area	Key insights	References
Overview of AI in medicine	The article highlights the diverse applications of AI in diagnostics and treatment, which improve accuracy and efficiency.	Ciardiello et al¹³
AI in healthcare	The studies discuss AI’s transformative impact on clinical decision-making processes and healthcare delivery.	Yu et al², Di Nicolantonio et al¹¹
AI in cancer research	The articles provide a systematic analysis of top-cited articles, emphasizing AI’s role in advancing oncology research.	Malik et al¹, David et al²⁵
Generative AI in healthcare	The studies explore the opportunities and challenges of generative AI technologies in healthcare settings.	Gambardella et al¹⁰, Liu et al²⁶
Revolutionizing healthcare with AI	The studies examine how AI drives transformation in clinical practice, enhancing patient care and operational workflows.	Zhang and Kamel Boulos⁴, Ciardiello et al¹³
Precision medicine in cancer research	The articles review AI’s contribution to personalized oncology, enabling targeted therapies based on patient-specific data.	Gambardella et al¹⁰, Rusert et al¹²
Personalized medicine in cancer therapy	The studies discuss recent advancements in cancer treatment facilitated by AI-powered precision medicine approaches.	Malik et al¹, Di Nicolantonio et al¹¹
AI applications in oncology	The articles focus on integrating AI technologies to establish and advance precision medicine in oncology.	Alowais et al⁵, Rahman et al²⁷

Figure 3.

(a) Conceptual roadmap outlining this review’s framework and thematic structure. (b) A visual map created by VOS Viewer shows groups of related research and trends, emphasizing connections between different fields and new areas in AI applications in oncology.

Quality Assessment

Each study was evaluated twice using a modified QUADAS-2 tool to assess the following aspects:

Technical domain: reproducibility, dataset diversity, external validation

Clinical domain: impact on patient outcomes, feasibility for adoption

Ethical domain: reporting of biases, data privacy measures, FDA/CE compliance

To acknowledge evolving methodologies in AI-oncology research, we retained low-quality studies (eg, small sample sizes or unreproducible methods), but flagged them in the analysis.

Network Analysis

The synthesis approach combined elements of systematic mapping and narrative synthesis. Each section brought together results from 20 to 30 important studies that were carefully examined for their role in combining different fields in AI and cancer. VOS Viewer (version 1.6.2) software was employed for bibliometric analysis and visualization of research trends, identifying clusters of related studies. Narrative synthesis was used to understand and combine the findings in context, highlighting studies that show connections between different fields (Figure 3b).

Early Detection and Diagnosis

AI has substantially impacted the early detection and diagnosis of cancer, offering promising avenues for future research and patient care.²⁸ Timely cancer diagnosis is crucial, as survival rates are higher in the earlier stages when the cancer is smaller and has not spread throughout the body.²⁹ AI can analyze various patient assessments to provide more accurate information about survival prognosis and predict disease progression in cancer patients. Healthcare providers often use a predictive model known as the Breast Cancer Risk Assessment Tool (the Gail model) to estimate a woman’s risk of developing breast cancer. This model calculates the risk of developing breast cancer over the next 5 years as well as throughout her lifetime, up to the age of 90.³⁰

The Gail model is widely used for estimating a woman’s risk of developing breast cancer. It takes into account 7 critical risk factors, including age, age at the onset of menstruation, age at the birth of the first child (or if the individual has not given birth), family history of breast cancer (in a mother, sister, or daughter), the number of past breast biopsies, the number of breast biopsies indicating atypical hyperplasia, and race/ethnicity.³¹ By incorporating these factors, the Gail model can provide a more accurate assessment of a woman’s risk of developing breast cancer over the next 5 years as well as throughout her lifetime, up to the age of 90.

Another example is a study by Koo et al that developed a novel model for individuals with metastatic or recurrent gastric cancer. This model incorporates 6 clinicopathological factors: high alkaline phosphatase level, poor Eastern Cooperative Oncology Group performance score, peritoneal metastasis, bone metastasis, high neutrophil-lymphocyte ratio, and low albumin level.³²

AI has made significant advancements in developing predictive models for disease progression, recurrence, and patient survival. These models utilize extensive clinical and molecular data, which include a patient’s genetic makeup, disease characteristics, and anticipated response to therapy.³³ These models can identify subtle diagnostics. By incorporating this information, the models provide more accurate risk assessments, thereby informing clinical decision-making and enhancing patient counseling.³⁴ Furthermore, Predictive models can significantly impact treatment outcomes by facilitating personalized and precise treatment planning.³⁵ By incorporating a patient’s genetic makeup, disease characteristics, and predicted response to therapy, these models assist oncologists in tailoring treatment regimens to suit individual patients, potentially resulting in improved treatment outcomes and reduced side effects.³⁶

AI-powered wearable devices and remote monitoring tools enhance patient care by continuously tracking vital signs and symptoms, enabling early intervention in case of adverse events or disease progression.³⁷

Medical Imaging

Medical imaging plays a crucial role in this complex and specific process.³⁸ However, DL algorithms, particularly Convolutional Neural Networks (CNNs), are highly accurate at identifying details in medical images like mammograms, ultrasounds, radiology, CT scans, PET scans, and MRIs to find tumors early.³⁹ Also, transformer neural networks have recently replaced CNNs in image processing due to their improved reliability and efficiency in computer vision tasks.⁴⁰

AI analysis in medical imaging encompasses image classification, segmentation, detection, and localization. It offers details about the probability of cancer and identifies disease sites, such as lesions and metastatic tumors. Some AI models have demonstrated accuracy surpassing that of human experts. For instance, an AI model developed by McKinney et al has significantly reduced false positives and negatives in breast cancer predictions in the USA and the UK. This AI outperformed average radiologists, decreasing their workload by 80%.⁴¹

In another study, a deep learning model trained on a large dataset achieved an accuracy of 82.9% in classifying images from digital single-operator cholangioscopy (DSOC), with a sensitivity of 83.5% and specificity of 82.4%.⁴² Table 2 demonstrates other selected AI models in medical imaging.

Table 2.

Selected AI Models in Medical Imaging From 2022 to 2024.

Imaging technique	Method	Description	Clinical status	Year	References
MRI	Deep Learning (DL)	Segmentation and classification of 18 distinct intracranial tumor types using T1- and T2-weighted images, including T2 contrast MRI sequences.	Research only; multi-center validation.	2022	Gao et al⁴³
	Convolutional Neural Network (CNN)	The prediction of tumor grading is combined with segmentation techniques that utilize whole 3D structural preoperative MRI data.	Research only; single-center validation.	2023	van der Voort et al⁴⁴
	Deep Learning (DL)	Real-time prostate contouring on MRI images from a 1.5 T MR-Linac system to guide radiation beam delivery.	Research only; single-center validation.	2024	Nachbar et al⁴⁵
CT scan	Deep Learning (DL)	“Sybil” estimates future lung cancer risk from a single low-dose CT (LDCT) scan.	FDA Accelerated Approval Program; investigational use only.	2023	Mikhael et al⁴⁶
CT scan	Convolutional Neural Network (CNN)	Detection of pancreatic ductal adenocarcinoma (PDA) on diagnostic CT scans and identification of visually occult PDA on pre-diagnostic CT scans.	Research only; multi-center validation.	2023	Korfiatis et al⁴⁷
PET Scan	Random Forest	Prediction of invisible intraprostatic lesions on 68 Ga-PSMA-11 PET in patients with primary prostate cancer.	Research only; multi-center validation.	2022	Yi et al⁴⁸
PET Scan	Deep Learning (DL)	“Masked-LMCTrans” algorithm reduces noise and enhances detail in low-dose PET images.	Research only; single-center validation.	2023	Wang et al⁴⁹
Endoscopy images	Convolutional Neural Network (CNN)	AI system for identification of esophageal cancer using endoscopy images.	Research only; multi-center prospective validation.	2022	Knabe et al⁵⁰
Endoscopy images	Deep Learning (DL)	Real-time AI application using endoscopic ultrasonography images to differentiate gastrointestinal stromal tumors from leiomyomas.	Clinically validated (no regulatory approval); prospective trial completed.	2024	Dong et al⁵¹
Ultrasound	Deep Convolutional Neural Network (DCNN)	EDL-BC, an ensemble deep learning model designed to identify high-risk lesions in breast cancer ultrasound images.	Research only; multi-center validation.	2023	Liao et al⁵²
Ultrasound	Convolutional Neural Network (CNN)	Novel grid-based deep feature generator leveraging multiple pre-trained CNNs for breast cancer detection and classification.	Research only; single-center validation.	2022	Liu et al⁵³

In 2024, Brancaccio et al assessed the effectiveness of AI in diagnosing skin cancer. Their review highlighted that while AI has been extensively used to analyze whole-body photographs for identifying changes in lesions, its integration into consumer-oriented smartphone applications and support for non-experts remain uncertain. Likewise, the role of AI in primary care is a topic of ongoing debate. General practitioners acknowledge the potential benefits of AI but emphasize the need for clear definitions concerning its specific applications.⁵⁴

Ha et al pointed out a significant drawback in using AI for analyzing mammograms, based on their study of more than 5700 women without symptoms who have dense breast tissue. The findings indicate that AI-enhanced mammography offers increased specificity, resulting in fewer false positives and a reduced need for follow-up imaging. However, combining mammography and supplemental ultrasound has proven to be a more effective strategy for detecting early-stage, node-negative cancers. Notably, ultrasound detected a significantly higher number of cancers, including cases that the AI system missed. These results suggest that for women with dense breasts, a combined screening approach utilizing both mammography and supplemental ultrasound may be more effective than relying solely on mammography augmented with AI.⁵⁵

Furthermore, AI models can become unstable due to changes or alterations. It has been reported that minor perturbations, such as slight patient movements, can lead to significant errors in AI results and incorrect diagnoses. Such problems are common during medical imaging, so networks require frequent retraining for each subsampling combination. As a result, this frequent retraining makes the process impractical for a wide range of applications.⁵⁶

Innovations in Digital Pathology

Digital pathology revolutionizes cancer diagnostics by converting traditional histopathology, immunohistochemistry, and cytology slides into high-resolution digital images using whole-slide scanners. By leveraging AI tools on these images, we can automate time-consuming tasks traditionally performed by pathologists. This automation not only enhances efficiency but also enables quicker and more reliable diagnoses, allowing pathologists to focus more on high-level decision-making tasks.⁵⁷

In 2021, the FDA authorized the first AI-based software to detect prostate cancer (Paige Prostate).⁵⁸ The Paige Prostate AI digital diagnostic tool uses data from the digital slide archive at New York’s Memorial Sloan Kettering Cancer Center (MSKCC) to classify whole-slide images of prostate biopsies as either “Suspicious” or “Not Suspicious” for prostatic adenocarcinoma.⁵⁹

A critical aspect of pathology is Biomarker detection. But traditional methods often face limitations in sensitivity, accuracy, and speed, especially when dealing with complex disease processes.⁶⁰

Niehues et al demonstrated that DL can effectively predict multiple biomarkers from histopathology images.⁶¹ They evaluated 6 leading DL architectures and found that numerous instance learning (MIL) models consistently outperformed traditional approaches. However, while the predictions for MSI and BRAF mutations were clinically promising, the forecasts for KRAS, NRAS, and PIKCA were not sufficiently reliable for clinical use.⁶²

AI in pathology enhances the precision of quantitative assessments and allows for the geographical contextualization of data through spatial algorithms.⁶³ Adding spatial metrics to immunohistochemistry (IHC) methods can make finding important biomarkers more useful in a clinical setting.⁶⁴ For example, a study showed that adding spatial information to IHC using multiplex IHC and immunofluorescence (IF) was better at predicting how patients would respond to immune checkpoint inhibitors than just using gene expression profiling (GEP) or regular IHC alone.⁶⁵ However, some common factors can affect AI results, such as color variations and errors in tissue preparation, including blur, tissue folds, and tissue tears. Also, foreign objects not included in the training set may be forcibly classified into predefined groups.⁶⁶ Therefore, human supervision is crucial for guaranteeing AI results. For example, Bodén et al showed that while digital image analysis reduced errors compared to visual methods, human-in-the-loop corrections were essential for addressing significant weaknesses, such as misclassification due to faint staining or incorrect separation of tumor and stroma regions.⁶⁷ Furthermore, the unavailability of data prevents the replication of many pathology AI models, despite their satisfactory performance. As Wagner et al reported, around 70% of pathology AI studies fail to provide publicly accessible data for independent validation.⁶⁸

Overall, despite challenges, computational pathology can enhance oncology through multiple approaches.⁶⁹ Table 3 shows other selected models developed for different AI applications in pathology.

Table 3.

Selected models were developed for different AI applications in pathology from 2020 to 2024.

Pathology field	Cancer type	Model input	Goal	Clinical status	Year	Reference
Risk prediction of cancer recurrence	Prostate	3D tissue volumes, including open-top light-sheet microscopy and microcomputed tomography	Recurrence risk prediction by capturing complex 3D morphological tissue features	Research use only	2024	Song et al⁷⁰
	Breast	Digital slides of tissue samples, with optional clinical data	Tumor likelihood assessment and recurrence score prediction	Research use only	2023	Howard et al⁷¹
	Colorectal	Digital images of hematoxylin and eosin (H&E)-stained tissue sections	Prediction of cancer recurrence likelihood after curative resection for stage I-III colorectal cancer	Research use only	2023	Nakanishi et al⁷²
Tumor metastasis detection	Pan-origin	Whole-slide images (WSIs)	Detection of lymph node cancer metastasis	Research use only	2024	Pan et al⁷³
Tumor metastasis detection	Bladder, Prostate, Breast	Whole-slide images	Detection of micrometastases in lymph nodes	Research use only	2023	Wu et al⁷⁴
Mitosis detection	Breast	H&E-stained biopsy slices	Mitotic cell detection	Research use only	2023	Shihabuddin and K⁷⁵
Mitosis detection	Uterus	Microscopic images of uterine leiomyosarcoma	Mitotic cell detection	Research use only	2022	Zehra et al⁷⁶
Biomarker detection	Prostate	Multiplex fluorescence immunohistochemistry	Quantification of Ki-67 via cell distance analysis using an algorithm	Clinically validated	2023	Blessin et al⁷⁷
	Thyroid	Papanicolaou-stained thyroid fine-needle aspiration cytology and ThinPrep cytological slides	Prediction of BRAF immunostaining results in thyroidectomy specimens	Research use only	2023	Wang et al⁷⁸
	Breast	Core biopsy samples from invasive breast cancer tumors	Quantification of Ki-67, estrogen receptor, and progesterone receptor	Clinically validated	2022	Abele et al⁷⁹
Imaging	Gastric	H&E-stained pathology slides	Tumor detection	Research use only	2021	Huang et al⁸⁰
Imaging	Prostate	Native, non-trained RGB WSIs (red, green, blue channels) of prostate core biopsy	Computational generation of H&E-stained images	Research use only	2020	Rana et al⁸¹

Robotic Surgery

AI has recently entered surgery, enhancing imaging, navigation, and robotics capabilities. This integration is poised to revolutionize surgical practices by linking real-time sensing with robotic control, facilitating varying degrees of autonomy in surgical procedures.⁸²

History of Robotic Surgery

Robotic systems in surgery have been a transformative leap in medical technology, with their utilization dating back over 20 years.⁸³ Integrating advanced hardware and AI has allowed these systems to assist in increasingly complex procedures, enhancing precision and improving patient outcomes.⁸⁴ Surgical robots like da Vinci have significantly advanced over traditional surgical methods. These systems provide surgeons greater control, flexibility, and access to intricate anatomical structures through robotic arms equipped with specialized surgical instruments.⁸⁵ The da Vinci robot, in particular, achieved a historic milestone in 1997 as the first surgical robot to receive FDA approval and has been deployed in over 8000 facilities, facilitating more than 13 million procedures.⁸⁶ Further studies have optimized this robot. For example, Eslamian et al developed an algorithm that autonomously positions the camera on the da Vinci surgical robot to optimize surgical visualization and determine the correct zoom level.⁸⁷

Medtronic’s latest robotic system, launched in 2021, is called the HUGO™ RAS System. This modular option is an alternative to the current market leader in surgical robotics, the da Vinci system.⁸⁸ The HUGO™ RAS System comprises an “open” surgical console with a high-definition 3D passive display, a system tower, and 4 arm carts. Each robotic arm is independent and features 6 different joints for flexibility.⁸⁹ Moreover, the cart column allows for the vertical placement of the arms by raising or lowering them.⁹⁰ Trocars attach these arms, while specialized motors known as instrument drive units control the instruments. This system provides adaptable and versatile arm movements due to its multiple joints.⁹¹ Figure 4 illustrates the historical development of surgical robots, showcasing their contribution to automated surgical tasks and intraoperative safety measures.^92
-94

Figure 4.

The timeline of surgical robots’ development and evolution.

AI Roles in Robotic Surgery

AI techniques for robotic and autonomous systems can be categorized into 4 groups: perception, localization and mapping, surgical workflow recognition, and human-robot interaction.⁹⁵

In perception, distinguishing between native tissue and non-native devices is a critical challenge in implementing AI. De Backer et al used deep learning networks to identify instruments during robot-assisted kidney transplantation, achieving a Dice score of 97.10%.⁹⁶ Ping et al developed a method for instrument detection in surgical endoscopy using a modified CNN and the You Only Look Once (YOLO) v3 algorithm, achieving a sensitivity of 93.02% for surgical instrument detection and 87.05% for tooltip detection.⁹⁷

As for providing information on native tissue, Doria et al developed stiffness models through a wearable fabric to give haptic feedback to tissue palpation of intrauterine leiomyomas during robotic surgery.⁹⁸

Accurate localization and mapping are essential in surgical oncology to ensure that tumors are wholly removed while preserving as much healthy tissue as possible. Marsden et al presented AI models using fluorescence imaging to generate an accurate heatmap of probable cancer location and guide cancer excision in oral and oropharyngeal surgery.⁹⁹ To identify safe dissection planes in tissues where they naturally separate, Kumazu et al utilized a sensitive deep-learning model that automatically segments loose connective tissue fibers during robot-assisted gastrectomy.¹⁰⁰

Studies have focused on higher levels of physical task automation in human-robot cooperation.

For instance, Saeidi et al created a CNN with a U-Net algorithm to find where the bowel is and completely automate laparoscopic bowel anastomoses, rather than just 1-step suturing.¹⁰¹

To automatically eliminate the smoke produced by electrocautery devices used in dissection and ligation, Wang et al proposed a convolutional neural network (CNN) combined with a Swim transformer. This innovative approach swiftly generates smoke-free surgical footage.¹⁰²

Surgical workflow recognition (SWR) provides surgeons with real-time feedback during operations to help prevent errors and assist in surgical training. Kitaguchi et al developed an AI model using a dataset of 300 laparoscopic colorectal resection videos. The model accurately classified 9 different surgical phases, achieving an average accuracy of 81%, and demonstrated an accuracy of 83.2% in distinguishing between dissection and exposure actions.¹⁰³

Sahu et al used a different method to reach SWR. They developed an RNN to recognize surgical phases by distinguishing the tools in videos.¹⁰⁴

To get through the time-consuming labeling process, Shi et al proposed an active learning system for gallbladder surgery.¹⁰⁵

Patient Recovery Statistics

Robotic surgery has been associated with numerous patient benefits, including reduced chances of readmission, shorter hospital stays, fewer minor surgical scars, a lower risk of infections, and less pain during recovery.¹⁰⁶ A study published in JAMA found that robotic surgery reduces the chance of readmission by half (52%) and revealed a four-fold (77%) reduction in the prevalence of blood clots.¹⁰⁷ Also, Patients’ physical activity, stamina, and quality of life improved significantly.¹⁰⁷ These findings demonstrate the benefits of robot-assisted surgery, indicating a safer surgery with fewer complications and a quicker return to daily activities.¹⁰⁸

While AI assistants have made significant strides in surgery, they can sometimes introduce time-consuming complexities. For instance, advancements in telesurgery offer promising capabilities; however, they often suffer from latency issues unless connected to high-speed, reliable networks. Latency exceeding 200 to 300 ms can compromise precision and safety, leading to increased fatigue for surgeons and a higher likelihood of errors.¹⁰⁹

Some studies have also reported low accuracies of AI in surgery. For example, in postoperative outcome prediction, an AI model predicted urinary continence after robot-assisted radical prostatectomy with a mean absolute error that ranged from 85.9 to 134.7 days.¹¹⁰ A Random Forest model detected only 35.7% of adverse events intraoperatively, despite 99.6% specificity.¹¹¹ Similarly, an AI model trained for autonomous tissue retraction tasks had a 20% error rate in detecting tissue flaps, and performance dropped further due to variability in tissue positioning.¹¹²

A systematic review conducted in 2021 analyzed 35 studies on the application of AI in robot-assisted surgery. It revealed that none of the models had been validated using datasets from multiple medical centers. Furthermore, models trained on small datasets, such as JIGSAWS and MICCAI EndoVis, demonstrated poor generalizability to real-world surgical situations and were limited in the tasks they could perform.¹¹³ These findings show how AI may struggle with complex, dynamic, and unpredictable surgical environments.

Overall, the integration of AI in surgery has significantly impacted the field, offering enhanced precision and improved patient recovery statistics.¹¹⁴ However, unlike screening and diagnosis, the implementation of AI in surgery is relatively new and requires further approval and standardization to ensure patient safety, optimal outcomes, and widespread adoption.¹¹⁵

AI in Anti-Cancer Drug Discovery

In the realm of treatment, the development of effective medications is a critical cornerstone.¹¹⁶ However, crafting new and triumphant anti-cancer drugs from the ground up remains a strenuous, costly, and time-intensive endeavor demanding intricate multidisciplinary collaborations spanning medicinal chemistry, computational chemistry, biology, pharmacology, and clinical research.¹¹⁷ The statistics paint a formidable picture, indicating that ushering a new drug from inception to clinical application can span over 10 to 17 years, with costs ballooning to almost $2.8 billion. Moreover, a mere 10% of compounds tested in clinical trials successfully navigate the arduous path to market availability.¹¹⁸ Figure 5 illustrates the steps of drug discovery.^119,120

Figure 5.

The typical steps of drug discovery.

Despite these challenges, AI has emerged as a formidable and optimistic technology in pursuing swifter, more cost-effective, and productive designs for anti-cancer drugs.¹²¹ Table 4 presents a compilation of anti-cancer drugs developed using AI and subsequently progressed to clinical studies.

Table 4.

Anti-Cancer Drugs Designed by AI Have Reached Clinical Studies.

Name	Generic name	Company	Therapeutic area	Target/mechanism of action	Development phase	Reference
REC-2282	AR-42	Recursion	Neurofibromatosis Type 2	Histone Deacetylase (HDAC) Inhibitor	Phase 2/3	Welling et al¹²²
RLY-4008	Lirafugratinib	Relay therapeutics	Solid tumors	FGFR (Fibroblast Growth Factor Receptor) Inhibitor	Phase 2	Schram et al¹²³
BPM31510	BPM31510	Berg	Solid tumors	Protein Bcl-2 Modulator	Phase 2	Subbiah et al¹²⁴
EXS-21546	EXS-21546	Exscientia	Solid tumors	A2a Receptor (A2aR) Antagonist	Phase 1	Vladimer et al¹²⁵
ISM3091	XL309	Insilico Medicine	BRCA-mutated tumors	USP1 (Ubiquitin-Specific Protease 1) Inhibitor	Phase 1	Philippidis¹²⁶
PHI-101	PHI101	Pharos iBio	Ovarian and breast cancer	FLT3 (Fms-like Tyrosine Kinase 3) Inhibitor	Phase 1	Yoon et al¹²⁷

Many different algorithms are used in drug discovery.¹²⁸ Some of the frequent algorithms in drug design are Support Vector Machines (SVM), Random Forest (RF), and Convolutional Neural Networks (CNN).¹²⁹ In Drug screening, Machine Learning (ML), Meta-heuristics, and similarity are mainly used.¹³⁰ In Chemical synthesis, SVMs, RFs, artificial neural networks (ANNs), CNNs, recurrent neural networks (RNNs), and graph neural networks (GNNs) are common.¹³¹ While in drug repurposing ML, deep learning (DL), network transmission, matrix completion, and matrix factorization are frequently used.¹³² Figure 6 demonstrates the role of AI in drug discovery.^133
-135

Figure 6.

The role of AI in different steps of drug discovery.

Anti-Cancer Drug Target Identification

Identifying interactions between drugs and their targets is a crucial first step in the design of effective anti-cancer medications.²⁷ The potency of these interactions is commonly quantified through binding affinity constants, which encompass metrics like dissociation constant (Kd), inhibition constant (Ki), and half-maximal inhibitory concentration (IC50).¹³⁶ Given the laborious and costly nature of experimentally pinpointing these interactions, the computational forecast of drug-target interactions holds significant appeal.¹³⁷ Precise and efficient predictions can significantly bolster drug development efforts and expedite the discovery of promising lead compounds or hits.¹³⁸

AI can integrate multiple datasets to select the best targets. For instance, Tong et al integrated clinical data with gene expression profiles and protein-protein interaction networks to predict candidate drug targets in liver cancer. They achieved an AUC of 0.88 using a one-class support vector machine.¹³⁹ Madhukar et al developed a Bayesian-based machine learning method (BANDIT), which integrated 6 data types on more than 2000 molecules to achieve approximately 90% target prediction accuracy.¹⁴⁰

On the Way to Hit and Lead Compounds

Following the identification of therapeutic targets crucial for anti-cancer drug development, the next step involves screening for hit compounds, molecules possessing initial activity against specific targets or modes of action.⁴¹ Computer-aided discovery of hit compounds primarily relies on high-throughput screening techniques.¹⁴¹ This screening process can be carried out through 2 main approaches: structure-based screening and ligand-based screening.¹⁴² Recent studies have demonstrated the efficacy of fragment-based screening methods in hit compound discovery.¹⁴³ While high-throughput screening has proven successful in numerous research and development endeavors, screening millions of compounds has encountered efficiency limitations and significant costs.¹⁴⁴ However, with the rise of GPU technology, enhanced computational capabilities, and swift advancements in AI, many virtual screening tools for hit compounds have emerged, enriching the arsenal of tools available for drug design.¹⁴⁵

ML models, specifically DL models like CNNs, have been developed to predict binding affinity and even generate 3D structures of protein-ligand complexes. These models are trained on massive datasets of known protein-ligand interactions and structures.

In their 2022 study, Arul Murugan et al conducted a comparative analysis of various computational screening methods for assessing ligand-protein interactions. Their findings identified 4 standout models: DEELIG, BgN-score, PerSPECT-ML, and TNet-BP, which recorded impressive Pearson correlation coefficients (Rp) of .89, .86, .84, and .83, respectively. Notably, DEELIG emerged as the most effective model, utilizing a comprehensive array of features that characterize both the protein binding pocket and the properties of the ligand, enabling it to predict binding affinity accurately.¹⁴⁶

Hit compounds must evolve into lead compounds to achieve refined and optimized properties suitable for preclinical testing. Consequently, it is essential to thoroughly investigate their properties to explore various synthetics and attain maximum optimization. This investigation utilizes QSAR (Quantitative Structure-Activity Relationship) models, which predict the biological activity of molecules based on their chemical structures. Additionally, QSPR (Quantitative Structure-Property Relationship) models are employed to forecast physicochemical properties, such as solubility and permeability, which play a critical role in drug development.¹⁴⁷

In 2023, Turon et al introduced ZariaChem, a groundbreaking tool designed to evaluate Quantitative Structure-Activity Relationships (QSAR) and Quantitative Structure-Property Relationships (QSPR) in low-resource environments. Utilizing a fully automated, machine-learning-driven system, they pioneered a virtual screening cascade aimed at drug discovery for malaria and tuberculosis, marking a significant advancement in this critical area of research.¹⁴⁸

Prediction of Anti-Cancer Drug Reactions

The relationship between drug reactions and their ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties is crucial in determining drug sensitivity and toxicity. Understanding these interactions is essential for optimizing therapeutic efficacy and minimizing adverse effects.¹⁴⁹ Precise drug reaction prediction can significantly enhance clinical trial efficacy, predict adverse drug reactions, and improve patient outcomes.¹⁵⁰ AI methods in drug design have increased due to advancements in AI technologies.²⁵ Past research has highlighted various ADMET characteristics such as Caco-2 permeability, carcinogenicity, blood-brain barrier permeability, and plasma protein binding.¹⁵¹ For instance, in predicting drug plasma protein binding, Mulpuru and Mishra developed a forecasting model for the portion of unbound drugs in human plasma by employing a chemical fingerprint alongside an accessible AutoML framework.¹⁵² In 2023, Swanson et al introduced ADMET-AI, the fastest web-based platform for ADMET predictions. This new tool uses a cutting-edge graph neural network, Chemprop-RDKit, to provide tailored predictions by matching input molecules with approved drugs. The advanced algorithms of this tool set a new standard in the field by streamlining the drug development process and enhancing the accuracy of assessments related to absorption, distribution, metabolism, excretion, and toxicity.¹⁵³

Anti-Cancer Drug Repurposing

The effective identification of novel applications for approved or well-established clinical drugs plays a crucial role in drug discovery, a process commonly referred to as drug repositioning.¹⁵⁴ Presently, one of the primary approaches for drug repurposing involves predicting drug-target relationships.¹⁵⁵ For example, to enable personalized drug repositioning leveraging genomic data, Cheng et al introduced a genome-wide localization system network algorithm (GPSnet).¹⁵⁶ This technique utilizes patient-specific DNA and RNA sequencing data concerning specific targets to delineate disease modules for repurposing drugs.¹⁵⁷ Their research verified that Ouabain’s established arrhythmia and heart failure medication effectively targets the HIF1α/LEO1-mediated cellular metabolic pathways in lung adenocarcinomas, hinting at potential anti-tumor properties.^156,157

Liu et al proposed the IDMCSS method to improve disease module identification by refining protein-protein interaction (PPI) networks. They incorporate potential interactions and remove incorrect ones based on similarities between disease-associated proteins and their neighbors to improve detection accuracy.²⁶

Forecasting interactions between drugs and diseases is crucial for disease-focused drug repurposing.¹⁵⁸ For instance, Jarada et al introduced an innovative deep-learning framework, SNF-NN, designed to anticipate novel drug-disease interactions by utilizing data on drug-related similarities, disease-related similarities, and existing drug-disease interactions.¹⁵⁹

Conclusion

Over the past decade, AI has profoundly transformed the field of oncology, enhancing early detection, precision diagnostics, robotic-assisted surgery, and expediting drug discovery. This review illustrates how AI systems have outperformed traditional methods in critical areas such as tumor classification and biomarker analysis. However, the integration of AI into clinical practice is not without challenges. Problems such as dataset biases, insufficient prospective validation, regulatory obstacles, and cost barriers require immediate attention and resolution.

Discussion

AI technologies in oncology are poised to bring significant advancements, including predictive analytics, targeted delivery, and image analysis. As shown in Table 5, AI-driven tools have demonstrated considerable potential in enhancing diagnostic accuracy, streamlining workflows, and personalizing treatment plans. However, several challenges must be addressed to ensure their reliable implementation in clinical settings.

Table 5.

A comparative Analysis of AI and Traditional Methods in Cancer Care Was Compiled From the Studies Reviewed in This Paper.

Category	AI Performance	Conventional Performance	Advantages of AI	Limitations	Reference
Medical imaging	94.5% sensitivity in breast cancer screening	88.3% accuracy by radiologists	6.2% higher cancer detection rate	Requires adherence to high-quality imaging protocols	McKinney et al⁴¹
Medical imaging	97.4% sensitivity in dense breast tissue (AI + mammography)	92.1% sensitivity with supplemental ultrasound	5.3% improved sensitivity	Limited availability to centers equipped with AI infrastructure	Ha et al⁵⁵
Medical imaging	92% pancreatic cancer detection on CT scans	70% detection at late clinical diagnosis	Enables 22% earlier cancer detection	Risk of false positives, especially in chronic pancreatitis cases	Korfiatis et al⁴⁷
Digital pathology	99.3% prostate cancer detection	96.6% accuracy by pathologists	2.7% reduction in missed diagnoses	Requires extensive expert-annotated training datasets	Perincheri et al⁵⁹
Digital pathology	98% concordance in HER2 scoring	Manual immunohistochemistry (IHC) with high variability	Provides standardized and reproducible results	Dependent on consistent stain quality	Frey et al⁶⁴
Digital pathology	95% accuracy in mitosis detection	78% accuracy via manual counting	17% improvement in grading consistency	Challenges in identifying borderline cases	Shihabuddin and Sabeena Beevi ⁷⁵
Digital pathology	92% sensitivity in lymph node metastasis detection	85% accuracy through manual review	7% increase in sensitivity	Requires access to high-resolution whole-slide imaging	Wu et al⁷⁴
Digital pathology	93% accuracy in Barrett’s carcinoma staging	88% accuracy in endoscopist assessment	5% more precise tumor staging	Necessitates correlation with endoscopic biopsy results	Knabe et al⁵⁰
Robotic surgery	An 85% 5-y survival rate with TORS for oropharyngeal cancer	78% survival with conventional surgery	7% improvement in long-term survival	Availability is limited to high-volume surgical centers	Nguyen et al¹⁰⁷
Robotic surgery	AUC 0.92 for continence prediction post-prostatectomy	AUC 0.76 using traditional nomograms	16% enhanced risk stratification	Requires validation across multiple clinical centers	Hung et al¹¹⁰
Robotic surgery	92% accuracy in tumor margin assessment during resection	85% accuracy via visual inspection	Reduced risk of cancer recurrence	Dependent on the integration of real-time imaging technologies	Marsden et al⁹⁹
Drug discovery	40% hit rate in virtual compound screening	1-5% hit rate in traditional screening methods	8 to 40-fold increase in screening efficiency	Performance depends heavily on chemical library diversity	Turon et al¹⁴⁸

Present AI models often rely on limited datasets and have primarily undergone retrospective trials for testing. To establish their reliability, these models require further multicenter validation in prospective studies. Although patient outcome assessments are critical for validation in areas like robotic-assisted surgery, there has been a notable lack of research focusing on these assessments, particularly in cancer-related contexts.¹¹³

The development of AI models must prioritize the use of diverse and representative datasets. For instance, many convolutional neural networks (CNNs) that demonstrate high accuracy in detecting skin lesions are trained on datasets where only 5% to 10% of the participants are Black. This results in a model accuracy for Black individuals that is roughly half that for White individuals, highlighting the critical need for diverse datasets to ensure equitable health outcomes.¹⁶⁰

The integration of large and diverse datasets is crucial to addressing complex challenges in healthcare. This necessitates collaboration between technology companies and healthcare providers. However, the reliance on medical and personal information underscores the need for robust frameworks to enable secure and anonymous data sharing. One promising approach is federated learning, which processes data locally at each institution. By sharing only model updates instead of raw data, this method effectively preserves privacy and enhances security.¹⁶¹

Future statistical studies should optimize workflows to minimize bias, including determining the necessary number of annotations or samples required for robust model training. Additionally, adherence to established guidelines for model reporting, such as the STARD-AI or TRIPOD-AI frameworks, is crucial for ensuring reproducibility and comparability across studies.^162,163

A significant concern in AI-driven medicine is the’ “black box” nature of many systems. This term refers to models that lack transparency in their input features and algorithms, rendering them difficult to interpret. Consequently, some clinicians hesitate to utilize AI tools, while others may blindly follow AI recommendations without fully understanding the underlying logic, potentially leading to inappropriate treatments.

For instance, an experimental study involving 28 pathology experts found that integrating AI improved overall diagnostic accuracy. However, it also resulted in a 7% bias rate, where initially accurate diagnoses were inaccurately altered following AI intervention. Notably, time pressure further amplified the reliance on incorrect AI outputs, underscoring significant risks associated with deskilling and over-dependence on technology.¹⁶⁴

A recent advancement in addressing the black box problem is the emergence of Explainable AI (XAI). This approach aims to make complex AI logic more interpretable for humans. By employing techniques such as data visualization and model simplification, XAI fosters greater trust and reproducibility in AI models. These developments are critical for enhancing transparency in AI systems, ensuring that users can understand and rely on AI-driven decisions.¹⁶⁵

Finally, the financial implications of integrating AI into medicine raise significant concerns. For example, hospital-wide AI implementation can exceed $36 billion annually,¹⁶⁶ creating challenges for resource-limited settings. Expenses of AI adoption may encompass expensive infrastructure, such as high-performance computing and cloud storage,¹⁶⁷ maintenance, and cybersecurity measures to protect sensitive genomic and imaging data,¹⁶⁸ clinician training to interpret AI outputs¹⁶⁹ and the cost of rigorous multi-center trials.¹⁷⁰ Consequently, comprehensive cost-benefit analyses and pilot studies are urgently needed to substantiate the case for the widespread adoption of AI in healthcare.

Footnotes

Abbreviations

AI – Artificial Intelligence

ANN – Artificial Neural Network

AUC – Area Under the Curve

BRAF – B-Raf Proto-Oncogene (biomarker)

CNN – Convolutional Neural Network

CT – Computed Tomography

DL – Deep Learning

DSOC – Digital Single-Operator Cholangioscopy

EDL-BC – Ensemble Deep Learning for Breast Cancer

FDA – Food and Drug Administration

FGFR – Fibroblast Growth Factor Receptor

GEP – Gene Expression Profiling

GPSnet – Genome-wide Positioning System network

HDAC – Histone Deacetylase

HER2 – Human Epidermal Growth Factor Receptor 2

H&E – Hematoxylin and Eosin (staining)

IHC – Immunohistochemistry

Ki-67 – Proliferation Marker Protein

LDCT – Low-Dose Computed Tomography

ML – Machine Learning

MRI – Magnetic Resonance Imaging

MSKCC – Memorial Sloan Kettering Cancer Center

PET – Positron Emission Tomography

QSAR – Quantitative Structure-Activity Relationship

QSPR – Quantitative Structure-Property Relationship

RAS – Robot-Assisted Surgery

RF – Random Forest

RNN – Recurrent Neural Network

SVM – Support Vector Machine

TORS – Trans Oral Robotic Surgery

USP1 – Ubiquitin-Specific Peptidase 1

WSI – Whole-Slide Image

Author Note

The study was conducted without financial support or relationships that could influence its outcomes. It was designed, executed, and analyzed independently, ensuring objectivity and integrity in the findings. The authors affirm that their work was carried out solely in their personal and academic capacities, free from the involvement of any governmental or sanctioned entities. They adhered to publishing ethics and compliance with relevant sanctions laws throughout the research and publication process.

ORCID iDs

Niki Najar Najafi

Helia Hajihassani

Maryam Azimzadeh Irani

Author Contributions

M.A.I. designed the presented idea, supervised the project, and edited the manuscript. N.N.N. and H.H. wrote the manuscript. Additionally, N.N.N. provided all the figures and plots. All authors have read and approved the final version of the manuscript and have full access to all of the data in this study. They take complete responsibility for the integrity and accuracy of the data analysis.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data supporting the findings of this study are available from the corresponding authors upon reasonable request.

References

Malik

Pathania

Rathaur

VK.

Overview of artificial intelligence in medicine. J Fam Med Prim Care. 2019;8(7):2328-2331.

Beam

Kohane

IS.

Artificial intelligence in healthcare. Nat Biomed Eng. 2018;2(10):719-731.

Musa

Afolabi

Zamit

, et al. Artificial intelligence and machine learning in cancer research: a systematic and thematic analysis of the top 100 cited articles indexed in Scopus database. Cancer Control. 2022;29:10732748221095946.

Zhang

Kamel Boulos

MN.

Generative AI in medicine and healthcare: promises, opportunities, and challenges. Future Internet. 2023;15(9):286.

Alowais

Alghamdi

Alsuhebany

, et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ. 2023;23(1):689.

Papadopoulou

Lytras

Advancing precision medicine in medical education: integrated, precise and data-driven smart solutions. Appl Res. 2023;2(6):e202200131.

Galasso

Precision medicine for whom? public health outputs from “Genomics England” and “All of Us” to make up for upstream and downstream exclusion. Am J Bioeth. 2024;24(3):71-85.

Bhinder

Gilvary

Madhukar

Elemento

Artificial intelligence in cancer research and precision medicine. Cancer Discov. 2021;11(4):900-915.

Yang

S-R

Schultheis

Mandelker

Ladanyi

Büttner

Precision medicine in non-small cell lung cancer: current applications and future directions. Semin Cancer Biol. 2022;84:184-198.

10.

Gambardella

Tarazona

Cejalvo

, et al. Personalized medicine: recent progress in cancer therapy. Cancers. 2020;12(4):1009.

11.

Di Nicolantonio

Vitiello

Marsoni

, et al. Precision oncology in metastatic colorectal cancer - from biology to medicine. Nat Rev Clin Oncol. 2021;18(8):506-525.

12.

Rusert

Juarez

Brabetz

, et al. Functional precision medicine identifies new therapeutic candidates for medulloblastoma. Cancer Res. 2020;80(23):5393-5407.

13.

Ciardiello

Martini

Napolitano

Tabernero

Cervantes

Clinical management of metastatic colorectal cancer in the era of precision medicine. CA Cancer J Clin. 2022;72(4):372-401.

14.

Uche-Anya

Anyane-Yeboa

Berzin

Ghassemi

, May FP. Artificial intelligence in gastroenterology and hepatology: how to advance clinical practice while ensuring health equity. Gut. 2022;71(9):1909-1915.

15.

Mateo

Steuten

Aftimos

, et al. Delivering precision oncology to patients with cancer. Nat Med. 2022;28(4):658-665.

16.

Chua

Gaziel-Yablowitz

Korach

, et al. Artificial intelligence in oncology: Path to implementation. Cancer Med. 2021;10(12):4138-4149.

17.

Chen

Lin

Sun

Artificial intelligence for assisting cancer diagnosis and treatment in the era of precision medicine. Cancer Commun. 2021;41(11):1100-1115.

18.

Dlamini

Francies

Hull

Marima

Artificial intelligence (AI) and big data in cancer and precision oncology. Comput Struct Biotechnol J. 2020;18:2300-2311.

19.

Hamamoto

Suvarna

Yamada

, et al. Application of artificial intelligence technology in oncology: towards the establishment of precision medicine. Cancers. 2020;12(12):3532.

20.

Kim

Bahl

Assessing risk of breast cancer: a review of risk prediction models. J Breast Imag. 2021;3(2):144-155.

21.

Elemento

Leslie

Lundin

Tourassi

Artificial intelligence in cancer research, diagnosis, and therapy. Nat Rev Cancer. 2021;21(12):747-752.

22.

Basu

Sinha

Ong

Basu

Artificial intelligence: how is it changing medical sciences and its future?

Indian J Dermatol. 2020;65(5):365-370.

23.

Luchini

Pea

Scarpa

Artificial intelligence in oncology: current applications and future perspectives. Br J Cancer. 2022;126(1):4-9.

24.

Vobugari

Raja

Sethi

Gandhi

Raja

Surani

SR.

Advancements in oncology with artificial intelligence - a review article. Cancers. 2022;14(5):1349.

25.

David

Thakkar

Mercado

Engkvist

Molecular representations in AI-driven drug discovery: a review and practical guide. J Cheminform. 2020;12(1):56.

26.

Liu

Zhu

Qiu

Locally adjust networks based on connectivity and semantic similarities for disease module detection. Front Genet. 2021;12:726596.

27.

Rahman

Islam

Rahman

, et al. Emerging promise of computational techniques in anti-cancer research: at a glance. Bioengineering. 2022;9(8):335.

28.

Haug

Drazen

JM.

Artificial intelligence and machine learning in clinical medicine, 2023. N Engl J Med. 2023;388(13):1201-1208.

29.

Koh

D-M

Papanikolaou

Bick

, et al. Artificial intelligence and machine learning in cancer imaging. Commun Med. 2022;2(1):133.

30.

Stevanato

Pedroso

Dell Agnolo

, et al. Use and applicability of the Gail model to calculate breast cancer risk: a scoping review. Asian Pac J Cancer Prev. 2022;23(4):1117-1123.

31.

Solikhah

Nurdjannah

Assessment of the risk of developing breast cancer using the gail model in Asian females: a systematic review. Heliyon. 2020;6(4):e03794.

32.

Koo

Ryu

Lee

Moon

Kang

YK.

New prognostic model for patients with advanced gastric cancer: fluoropyrimidine/platinum doublet for first-line chemotherapy. World J Gastroenterol. 2021;27(48):8357-8369.

33.

Briganti

Le Moine

Artificial intelligence in medicine: today and tomorrow. Front Med. 2020;7:509744.

34.

Abdulaal

Patel

Charani

Denny

Mughal

Moore

Prognostic modeling of COVID-19 using artificial intelligence in the United Kingdom: model development and validation. J Med Internet Res. 2020;22(8):e20259.

35.

Hernandez-Boussard

Macklin

Greenspan

, et al. Digital twins for predictive oncology will be a paradigm shift for precision cancer care. Nat Med. 2021;27(12):2065-2066.

36.

Musa

Zamit

Okeke

Artificial intelligence and machine learning in oncology: historical overview of documents indexed in the Web of Science database. Eurasian J Med Oncol. 2021;5(3):239-248.

37.

Huang

Yang

Fong

Zhao

Artificial intelligence in cancer diagnosis and prognosis: opportunities and challenges. Cancer Lett. 2020;471:61-71.

38.

Svoboda

Artificial intelligence is improving the detection of lung cancer. Nature. 2020;587(7834):S20-S22.

39.

Haleem

Javaid

Khan

IH.

Current status and applications of Artificial Intelligence (AI) in medical field: an overview. Curr Med Res Pract. 2019;9(6):231-237.

40.

Tang

Hou

, et al. SiGra: single-cell spatial elucidation through an image-augmented graph transformer. Nat Commun. 2023;14(1):5618.

41.

McKinney

Sieniek

Godbole

, et al. International evaluation of an AI system for breast cancer screening. Nature. 2020;577(7788):89-94.

42.

Mendes

Domingues

Aidos

Garcia

Matela

AI in breast cancer imaging: a survey of different applications. J Imaging. 2022;8(9):228.

43.

Gao

Shan

Guo

, et al. Development and validation of a deep learning model for brain tumor diagnosis and classification using magnetic resonance imaging. JAMA Netw Open. 2022;5(8):e2225608.

44.

van der Voort

Incekara

Wijnenga

MMJ

, et al. Combined molecular subtyping, grading, and segmentation of glioma using multi-task deep learning. Neuro-oncology. 2023;25(2):279-289.

45.

Nachbar

Lo Russo

Gani

, et al. Automatic AI-based contouring of prostate MRI for online adaptive radiotherapy. Z Med Phys. 2024;34(2):197-207.

46.

Mikhael

Wohlwend

Yala

, et al. Sybil: A validated deep learning model to predict future lung cancer risk from a single low-dose chest computed tomography. J Clin Oncol. 2023;41(12):2191-2200.

47.

Korfiatis

Suman

Patnam

, et al. Automated artificial intelligence model trained on a large data set can detect pancreas cancer on diagnostic computed tomography scans as well as visually occult preinvasive cancer on prediagnostic computed tomography scans. Gastroenterology. 2023;165(6):1533-1546.e4.

48.

Lin

, et al. Machine learning-based prediction of invisible intraprostatic prostate cancer lesions on (68) Ga-PSMA-11 PET/CT in patients with primary prostate cancer. Eur J Nucl Med Mol Imaging. 2022;49(5):1523-1534.

49.

Wang

Sheybani

, et al. AI transformers for radiation dose reduction in serial whole-body PET scans. Radiol Artif Intell. 2023;5(3):e220246.

50.

Knabe

Welsch

Blasberg

, et al. Artificial intelligence-assisted staging in Barrett’s carcinoma. Endoscopy. 2022;54(12):1191-1197.

51.

Dong

Zhao

Zheng

, et al. Efficacy of real-time artificial intelligence-aid endoscopic ultrasonography diagnostic system in discriminating gastrointestinal stromal tumors and leiomyomas: a multicenter diagnostic study. EClinicalMedicine. 2024;73:102656.

52.

Liao

Gui

, et al. Artificial intelligence-assisted ultrasound image analysis to discriminate early breast cancer in Chinese population: a retrospective, multicentre, cohort study. EClinicalMedicine. 2023;60:102001.

53.

Liu

Cui

Luo

, et al. Artificial intelligence-based breast cancer diagnosis using ultrasound images and grid-based deep feature generator. Int J Gen Med. 2022;15:2271-2282.

54.

Brancaccio

Balato

Malvehy

Puig

Argenziano

Kittler

Artificial intelligence in skin cancer diagnosis: a reality check. J Investig Dermatol. 2024;144(3):492-499.

55.

Jang

Youn

, et al. Screening outcomes of mammography with AI in dense breasts: a comparative study with supplemental screening US. Radiology. 2024;312(1):e233391.

56.

Antun

Renna

Poon

Adcock

Hansen

AC.

On instabilities of deep learning in image reconstruction and the potential costs of AI. Proc Natl Acad Sci USA. 2020;117(48):30088-30095.

57.

Betmouni

Diagnostic digital pathology implementation: learning from the digital health experience. Digit Health. 2021;7:20552076211020240.

58.

Shafi

Parwani

AV.

Artificial intelligence in diagnostic pathology. Diagn Pathol. 2023;18(1):109.

59.

Perincheri

Levi

Celli

, et al. An independent assessment of an artificial intelligence system for prostate cancer detection shows strong diagnostic accuracy. Mod Pathol. 2021;34(8):1588-1595.

60.

Xiao

Zhang

, et al. High-throughput proteomics and AI for cancer biomarker discovery. Adv Drug Deliv Rev. 2021;176:113844.

61.

Niehues

Quirke

West

et al. Generalizable biomarker prediction from cancer pathology slides with self-supervised deep learning: a retrospective multi-centric study. Cell Rep Med. 2023;4(4):100980.

62.

Zhang

Tang

, et al. Novel research and future prospects of artificial intelligence in cancer diagnosis and treatment. J Hematol Oncol. 2023;16(1):114.

63.

Rai Dastidar

Ethirajan

. Whole slide imaging system using deep learning-based automated focusing. Biomed Opt Express. 2020;11(1):480-491.

64.

Frey

Mamilos

Minin

, et al. AI-Based HER2-low IHC Scoring in Breast Cancer Across Multiple Sites, Clones, and Scanners. American Society of Clinical Oncology; 2023.

65.

Stein

Rimm

, et al. Comparison of biomarker modalities for predicting response to PD-1/PD-L1 checkpoint blockade: a systematic review and meta-analysis. JAMA Oncol. 2019;5(8):1195-1204.

66.

Komura

Ishikawa

Machine learning methods for histopathological image analysis. Comput Struct Biotechnol J. 2018;16:34-42.

67.

Bodén

ACS

Molin

Garvin

West

Lundström

Treanor

. The human-in-the-loop: an evaluation of pathologists’ interaction with artificial intelligence in clinical practice. Histopathology. 2021;79(2):210-218.

68.

Wagner

Matek

Boushehri

, et al. Built to last? Reproducibility and reusability of deep learning algorithms in computational pathology. Mod Pathol. 2024;37(1):100350.

69.

Acs

Rantalainen

Hartman

Artificial intelligence as the next step towards precision pathology. J Intern Med. 2020;288(1):62-81.

70.

Song

Williams

Williamson

DFK

, et al. Weakly supervised AI for efficient analysis of 3D pathology samples. ArXiv. 2023.

71.

Howard

Dolezal

Kochanny

, et al. Integration of clinical features and deep learning on pathology for the prediction of breast cancer recurrence assays and risk of recurrence. NPJ Breast Cancer. 2023;9(1):25.

72.

Nakanishi

Morooka

Omori

, et al. Artificial intelligence-based prediction of recurrence after curative resection for colorectal cancer from digital pathological images. Ann Surg Oncol. 2023;30(6):3506-3514.

73.

Pan

Dai

Wang

, et al. Clinically applicable pan-origin cancer detection for lymph nodes via artificial intelligence-based pathology. Pathobiology. 2024;91(5):345-358.

74.

Hong

, et al. Artificial intelligence-based model for lymph node metastases detection on whole slide images in bladder cancer: a retrospective, multicentre, diagnostic study. Lancet Oncol. 2023;24(4):360-370.

75.

Shihabuddin

Sabeena Beevi

Multi CNN based automatic detection of mitotic nuclei in breast histopathological images. Comput Biol Med. 2023;158:106815.

76.

Zehra

Anjum

Mahmood

, et al. A novel deep learning-based mitosis recognition approach and dataset for uterine leiomyosarcoma histopathology. Cancers. 2022;14(15):3785.

77.

Blessin

Yang

Mandelkow

, et al. Automated Ki-67 labeling index assessment in prostate cancer using artificial intelligence and multiplex fluorescence immunohistochemistry. J Pathol. 2023;260(1):5-16.

78.

Wang

Muzakky

Lee

Lin

Chao

TK.

Annotation-free deep learning-based prediction of thyroid molecular cancer biomarker BRAF (V600E) from cytological slides. Int J Mol Sci. 2023;24(3):2521.

79.

Abele

Tiemann

Krech

, et al. Noninferiority of artificial intelligence-assisted analysis of Ki-67 and estrogen/progesterone receptor in breast cancer routine diagnostics. Mod Pathol. 2023;36(3):100033.

80.

Huang

Tian

Zhan

, et al. Accurate diagnosis and prognosis prediction of gastric cancer using deep learning on digital pathological images: a retrospective multicentre study. EBioMedicine. 2021;73:103631.

81.

Rana

Lowe

Lithgow

, et al. Use of deep learning to develop and analyze computational hematoxylin and eosin staining of prostate core biopsy images for tumor diagnosis. JAMA Netw Open. 2020;3(5):e205111.

82.

Zhou

X-Y

Guo

Shen

Yang

G-Z.

Application of artificial intelligence in surgery. Front Med. 2020;14(4):417-430.

83.

Reddy

Gharde

Tayade

Patil

Reddy

Surya

Advancements in robotic surgery: a comprehensive overview of current utilizations and Upcoming Frontiers. Cureus. 2023;15(12):e50415.

84.

Biswas

Sikander

Kulkarni

Recent advances in robot-assisted surgical systems. Biomed Eng Adv. 2023;6:100109.

85.

Fiacchini

Vianini

Dallan

Bruschini

Is the Da Vinci Xi system a real improvement for oncologic transoral robotic surgery? A systematic review of the literature. J Robot Surg. 2021;15(1):1-12.

86.

D’Ettorre

Mariani

Stilli

, et al. Accelerating surgical robotics research: a review of 10 years with the da Vinci research kit. IEEE Robot Autom Mag. 2021;28(4):56-78.

87.

Eslamian

Reisner

Pandya

AK.

Development and evaluation of an autonomous camera control algorithm on the da Vinci surgical system. Int J Med Robot Comput Assist Surg. 2020;16(2):e2036.

88.

Ngu

JC-Y

Lin

CC-W

Sia

CJ-Y

Teo

N-Z

. A narrative review of the Medtronic Hugo RAS and technical comparison with the intuitive da Vinci robotic surgical system. J Robot Surg. 2024;18(1):99.

89.

Cacciatore

Costantini

Tedesco

, et al. Robotic Medtronic Hugo™ RAS System is now reality: introduction to a new simulation platform for training residents. Sensors. 2023;23(17):7348.

90.

Gueli Alletti

Chiantera

Arcuri

, et al. Introducing the new surgical robot hugo™ RAS: system description and docking settings for gynecological surgery. Front Oncol. 2022;12:898060.

91.

Balestrazzi

Paciotti

Piro

, et al. Comparative analysis of robot-assisted simple prostatectomy: the HUGO™ RAS system versus the DaVinci^® Xi system. Prostate Cancer Prostatic Dis. 2024;27(1):122-128.

92.

Tamaki

Rocco

Ozer

The future of robotic surgery in otolaryngology - head and neck surgery. Oral Oncol. 2020;101:104510.

93.

Hoeppner

Robotic cancer surgery. Cancers. 2021;13(19):4931.

94.

Gharagozloo

Tempesta

Meyer

Nguyen

Gruessner

Redan

History of Robotic Surgery. Springer; 2021.

95.

Knudsen

Ghaffar

Hung

AJ.

Clinical applications of artificial intelligence in robotic surgery. J Robot Surg. 2024;18(1):102.

96.

De Backer

Van Praet

Simoens

, et al. Improving augmented reality through deep learning: real-time instrument delineation in robotic renal surgery. Eur Urol. 2023;84(1):86-91.

97.

Ping

Wang

Yao

, et al. Application and evaluation of surgical tool and tool tip recognition based on convolutional neural network in multiple endoscopic surgical scenarios. Surg Endosc. 2023;37(9):7376-7384.

98.

Doria

Fani

Giannini

Simoncini

Bianchi

Enhancing the localization of uterine leiomyomas through cutaneous softness rendering for robot-assisted surgical palpation applications. IEEE Trans Haptics. 2021;14(3):503-512.

99.

Marsden

Weyers

Bec

, et al. Intraoperative margin assessment in oral and oropharyngeal cancer using label-free fluorescence lifetime imaging and machine learning. IEEE Trans Biomed Eng. 2021;68(3):857-868.

100.

Kumazu

Kobayashi

Kitamura

, et al. Automated segmentation by deep learning of loose connective tissue fibers to define safe dissection planes in robot-assisted gastrectomy. Sci Rep. 2021;11(1):21198.

101.

Saeidi

Opfermann

Kam

, et al. Autonomous robotic laparoscopic surgery for intestinal anastomosis. Sci Robot. 2022;7(62):eabj2908.

102.

Wang

Sun

Surgical smoke removal via residual Swin transformer network. Int J Comput Assist Radiol Surg. 2023;18(8):1417-1427.

103.

Kitaguchi

Takeshita

Matsuzaki

, et al. Automated laparoscopic colorectal surgery workflow recognition using artificial intelligence: experimental research. Int J Surg. 2020;79:88-94.

104.

Sahu

Szengel

Mukhopadhyay

Zachow

Surgical phase recognition by learning phase transitions. Curr Directions Biomed Eng. 2020;6:20200037.

105.

Shi

Jin

Dou

Heng

P-A.

LRTD: long-range temporal dependency based active learning for surgical workflow recognition. Int J Comput Assist Radiol Surg. 2020;15:1573-1584.

106.

Zhang

Weng

Lund

Applications of explainable artificial intelligence in diagnosis and surgery. Diagnostics. 2022;12(2):237.

107.

Nguyen

Luu

Mallen-St Clair

, et al. Comparison of survival after transoral robotic surgery vs nonrobotic surgery in patients with early-stage oropharyngeal squamous cell carcinoma. JAMA Oncol. 2020;6(10):1555-1562.

108.

Andras

Mazzone

van Leeuwen

FWB

, et al. Artificial intelligence and robotics: a combination that is changing the operating room. World J Urol. 2020;38:2359-2366.

109.

Mennella

Maniscalco

De Pietro

Esposito

Ethical and regulatory challenges of AI technologies in healthcare: a narrative review. Heliyon. 2024;10(4):e26297.

110.

Hung

Chen

Ghodoussipour

, et al. A deep-learning model using automated performance metrics and clinical features to predict urinary continence recovery after robot-assisted radical prostatectomy. BJU Int. 2019;124(3):487-495.

111.

Bhandari

Nallabasannagari

Reddiboina

, et al. Predicting intra-operative and postoperative consequential events using machine-learning techniques in patients undergoing robot-assisted partial nephrectomy: a Vattikuti Collective Quality Initiative database study. BJU Int. 2020;126(3):350-358.

112.

Attanasio

Scaglioni

Leonetti

, et al. Autonomous tissue retraction in robotic assisted minimally invasive surgery – a feasibility study. IEEE Robot Autom Lett. 2020;5(4):6528-6535.

113.

Moglia

Georgiou

Satava

Cuschieri

A systematic review on artificial intelligence in robot-assisted surgery. Int J Surg. 2021;95:106151.

114.

Gómez Ruiz

Lainez Escribano

Fernandez

Cristobal Poch

Santarrufina Martinez

Robotic surgery for colorectal cancer. Ann Gastroenterol Surg. 2020;4(6):646-651.

115.

Park

Lee

Park

Kim

Choi

G-S.

Robot surgery shows similar long-term oncologic outcomes as laparoscopic surgery for mid/lower rectal cancer but is beneficial to ypT3/4 after preoperative chemoradiation. Dis Colon Rectum. 2021;64(7):812-821.

116.

Chen

Garmire

Calvisi

Chua

M-S

Kelley

Chen

Harnessing big ‘omics’ data and AI for drug discovery in hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol. 2020;17(4):238-251.

117.

Kumar

Saha

A review on artificial intelligence and machine learning to improve cancer management and drug discovery. Int J Res Appl Sci Biotechnol. 2022;9(3):149-156.

118.

Pandiyan

Wang

A comprehensive review on recent approaches for cancer drug discovery associated with artificial intelligence. Comput Biol Med. 2022;150:106140.

119.

Berdigaliyev

Aljofan

An overview of drug discovery and development. Future Med Chem. 2020;12(10):939-947.

120.

Boike

Henning

Nomura

DK.

Advances in covalent drug discovery. Nat Rev Drug Discov. 2022;21(12):881-898.

121.

You

Lai

Pan

, et al. Artificial intelligence in cancer target identification and drug discovery. Signal Transduct Target Ther. 2022;7(1):156.

122.

Welling

Collier

Burns

, et al. Early phase clinical studies of AR-42, a histone deacetylase inhibitor, for neurofibromatosis type 2-associated vestibular schwannomas and meningiomas. Laryngoscope Investig Otolaryngol. 2021;6(5):1008-1019.

123.

Schram

Subbiah

Liao

, et al. Abstract IA006: Clinical Activity of Lirafugratinib (RLY-4008), a Highly Selective FGFR2 Inhibitor, in Patients With Advanced FGFR2-Altered Solid Tumors: The ReFocus Study. American Association for Cancer Research (AACR); 2023:IA006.

124.

Subbiah

Nastke

Miller

, et al. 1551P validation of response and survival biomarkers in a phase II trial of BPM 31510-IV in advanced refractory pancreatic ductal adenocarcinoma (PDAC). Ann Oncol. 2020;31:S949.

125.

Vladimer

Alt

Sehlke

, et al. 23P enriching for response: patient selection criteria for A2AR inhibition by EXS-21546 through ex vivo modelling in primary patient material. Oncol Technol. 2022;16:100128.

126.

Philippidis

StockWatch: Insilico CEO Breaks Down Exelixis Deal: Alex Zhavoronkov discusses USP1’s appeal to AI drug developer pipeline updates. GEN Edge. 2023;5(1):644-651.

127.

Yoon

S-S

Shin

D-Y

Lee

J-H

, et al. A potent small molecule inhibitor of FLT3, PHI-101 overcomes resistance in acute myeloid leukemia: efficacy and PK/PD profile in Phase 1 first in human study. Blood. 2022;140(Supplement 1):3365-3366.

128.

Sadybekov

Katritch

Computational approaches streamlining drug discovery. Nature. 2023;616(7958):673-685.

129.

Luu

Fiorini

Combes

Quemeneur

Bonneville

Bousquet

PJ.

Challenges of artificial intelligence in precision oncology: public-private partnerships including national health agencies as an asset to make it happen. Ann Oncol. 2024;35(2):154-158.

130.

Oliveira

Silva

Maia

Silva

Taranto

Virtual screening algorithms in drug discovery: a review focused on machine and deep learning methods. Drugs Drug Candidates. 2023;2(2):311-334.

131.

Jia

Pei

Wang

, et al. The roles of computer-aided drug synthesis in drug development. Green Synthesis Catalysis. 2022;3(1):11-24.

132.

Cai

Chu

, et al. Machine learning for drug repositioning: recent advances and challenges. Curr Res Chem Biol. 2023;3:100042.

133.

Jiménez-Luna

Grisoni

Weskamp

Schneider

Artificial intelligence in drug discovery: recent advances and future perspectives. Expert Opin Drug Discov. 2021;16(9):949-959.

134.

Shaker

Ahmad

Lee

Jung

In silico methods and tools for drug discovery. Comput Biol Med. 2021;137:104851.

135.

Patel

Shukla

Huang

Ussery

Wang

Machine learning methods in drug discovery. Molecules. 2020;25(22):5277.

136.

Pun

Ozerov

Zhavoronkov

AI-powered therapeutic target discovery. Trends Pharmacol Sci. 2023;44:561-572.

137.

Sharma

Singh

Chauhan

, et al. Role of artificial intelligence in drug discovery and target identification in cancer. Curr Drug Deliv. 2024;21(6):870-886.

138.

Wang

Sim

Kim

Y-S

Chang

Y-T.

Tools for target identification and validation. Curr Opin Chem Biol. 2004;8(4):371-377.

139.

Tong

Zhou

Wang

Identifying potential drug targets in hepatocellular carcinoma based on network analysis and one-class support vector machine. Sci Rep. 2019;9(1):10442.

140.

Madhukar

Khade

Huang

, et al. A Bayesian machine learning approach for drug target identification using diverse data types. Nat Commun. 2019;10(1):5221.

141.

Leibig

Brehmer

Bunk

Byng

Pinker

Umutlu

Combining the strengths of radiologists and AI for breast cancer screening: a retrospective analysis. Lancet Digit Health. 2022;4(7):e507-e519.

142.

Freeman

Geppert

Stinton

, et al. Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy. BMJ. 2021;374:n1872.

143.

Houssami

Kirkpatrick-Jones

Noguchi

Lee

CI.

Artificial Intelligence (AI) for the early detection of breast cancer: a scoping review to assess AI’s potential in breast screening practice. Expert Rev Med Devices. 2019;16(5):351-362.

144.

Dustler

Evaluating AI in breast cancer screening: a complex task. Lancet Digit Health. 2020;2(3):e106-e107.

145.

Goyal

Mann

Gandhi

, et al. Scope of artificial intelligence in screening and diagnosis of colorectal cancer. J Clin Med. 2020;9(10):3313.

146.

Arul Murugan

Ruba Priya

Narahari Sastry

Markidis

. Artificial intelligence in virtual screening: models versus experiments. Drug Discov Today. 2022;27(7):1913-1923.

147.

Muratov

Bajorath

Sheridan

, et al. QSAR without borders. Chem Soc Rev. 2020;49(11):3525-3564.

148.

Turon

Hlozek

Woodland

Kumar

Chibale

Duran-Frigola

First fully-automated AI/ML virtual screening cascade implemented at a drug discovery centre in Africa. Nat Commun. 2023;14(1):5736.

149.

Daoud

Borah

Deb

, et al. ADMET profiling in drug discovery and development: perspectives of in silico, in vitro and integrated approaches. Curr Drug Metab. 2021;22(7):503-522.

150.

Yang

CC.

Explainable artificial intelligence for predictive modeling in healthcare. J Heal Inform Res. 2022;6(2):228-239.

151.

Selvaraj

Chandra

Singh

SK.

Artificial intelligence and machine learning approaches for drug design: challenges and opportunities for the pharmaceutical industries. Mol Divers. 2022;26:1893-1913.

152.

Mulpuru

Mishra

In silico prediction of fraction unbound in human plasma from chemical fingerprint using automated machine learning. ACS Omega. 2021;6(10):6791-6797.

153.

Swanson

Walther

Leitz

, et al. ADMET-AI: a machine learning ADMET platform for evaluation of large-scale chemical libraries. bioRxiv. 2023. doi:10.1101/2023.12.28.573531

154.

Turanli

Altay

Borén

, et al. Systems biology based drug repositioning for development of cancer therapy. Semin Cancer Biol. 2021;68:47-58.

155.

Kori

Arga

Mardinoglu

Turanli

Repositioning of anti-inflammatory drugs for the treatment of cervical cancer sub-types. Front Pharmacol. 2022;13:884548.

156.

Cheng

Liu

, et al. A genome-wide positioning systems network algorithm for in silico drug repurposing. Nat Commun. 2019;10(1):3476.

157.

Wang

Yang

Chen

Wang

DeepDRK: a deep learning framework for drug repurposing through kernel-based multi-omics integration. Brief Bioinform. 2021;22(5):bbab048.

158.

Kim

Jung

Y-S

Park

J-H

Kim

S-J

Cho

Y-R.

Drug-disease association prediction using heterogeneous networks for computational drug repositioning. Biomolecules. 2022;12(10):1497.

159.

Jarada

Rokne

Alhajj

SNF-NN: computational method to predict drug-disease interactions using similarity network fusion and neural networks. BMC Bioinformatics. 2021;22:1-20.

160.

Kamulegeya

Bwanika

Okello

, et al. Using artificial intelligence on dermatology conditions in Uganda: a case for diversity in training data sets for machine learning. Afr Health Sci. 2023;23(2):753-763.

161.

Rieke

Hancox

, et al. The future of digital health with federated learning. NPJ Digit Med. 2020;3:119.

162.

Sounderajah

Ashrafian

Aggarwal

, et al. Developing specific reporting guidelines for diagnostic accuracy studies assessing AI interventions: the STARD-AI Steering Group. Nat Med. 2020;26(6):807-808.

163.

Collins

Moons

KGM

Dhiman

, et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ. 2024;385:q902. doi:10.1136/bmj.q902

164.

Rosbach

Ganz

Ammeling

Riener

Aubreville

. Automation Bias in AI-Assisted Medical Decision-Making Under Time Pressure in Computational Pathology. Springer; 2025:129-134.

165.

Amann

Blasimme

Vayena

Frey

Madai

VI.

Explainability for artificial intelligence in healthcare: a multidisciplinary perspective. BMC Med Inform Decis Mak. 2020;20(1):310.

166.

Keris

MP.

Artificial intelligence in medicine creates real risk management and litigation issues. J Healthc Risk Manag. 2020;40(2):21-26.

167.

Ganai

Pandow

BA.

Understanding financial impact of machine and deep learning in healthcare: an analysis. In: Raton

ed. Applications of Machine Learning and Deep Learning on Biological Data. Auerbach Publications; 2023;41-56.

168.

Marques

Almeida

Pereira

The medicine revolution through artificial intelligence: ethical challenges of machine learning algorithms in decision-making. Cureus. 2024;16(9):e69405.

169.

Jiang

, et al. Opportunities and challenges of artificial intelligence in the medical field: current application, emerging problems, and problem-solving strategies. J Int Med Res. 2021;49(3):3000605211000157.

170.

Islam

MM.

Exploring the impact of artificial intelligence in healthcare. J Artif Intell Gen Sci. 2024;2(1):228-238.

The Impact of Artificial Intelligence on Cancer Diagnosis and Treatment: A Review

Abstract

Keywords

Graphical Abstract

Introduction to AI in the Field of Medicine

AI-Powered Precision Medicine: Advancing Personalized Treatment

The Historical Background of AI in Cancer Research

Methodology

Protocol Registration and PRISMA Adherence

Search Strategy

Inclusion and Exclusion Criteria

Study Screening and Selection

Data Extraction and Synthesis

Quality Assessment

Network Analysis

Early Detection and Diagnosis

Medical Imaging

Innovations in Digital Pathology

Robotic Surgery

History of Robotic Surgery

AI Roles in Robotic Surgery

Patient Recovery Statistics

AI in Anti-Cancer Drug Discovery

Anti-Cancer Drug Target Identification

On the Way to Hit and Lead Compounds

Prediction of Anti-Cancer Drug Reactions

Anti-Cancer Drug Repurposing

Conclusion

Discussion

Footnotes

Abbreviations

Author Note

ORCID iDs

Author Contributions

Funding

Declaration of Conflicting Interests

Data Availability Statement

References