Sage Journals: Discover world-class research

Abstract

Purpose

This narrative review synthesizes the latest evidence on TeleStroke implementation and proposes a structured protocol for developing an advanced TeleStroke network.

Methods

A structured search was conducted in PubMed, Scopus, Web of Science, and Institute of Electrical and Electronics Engineers (IEEE) Xplore (2000-2024) using “telemedicine” OR “TeleStroke” combined with “regulatory aspects,” “ethics,” “information architecture,” “artificial intelligence (AI),” and “training.” The review sections were consensually defined by four vascular neurologists and two digital health experts. An advanced network was consensually defined as one that enables synchronous communication, secure data storage, compliance with data protection laws, remote neuroimaging visualization, integration with prehospital emergency systems, and a validated AI tool. Studies were analyzed for best practices, regulatory considerations, and AI-driven innovations.

Results

We included 142 publications between 2000 and 2024 in the final review. Key components of a TeleStroke network include deploying neurologists trained in cerebrovascular care and telemedicine, adopting secure cloud-based technologies for imaging and data storage, and utilizing blockchain to enhance data security. Embedded AI tools, such as neuroimaging analysis and natural language processing (NLP), expedite diagnostics and decision-making. Georeferenced algorithms direct ambulances to appropriate stroke centers based on patient severity and available resources. Integrating telemedicine into ambulance systems enables real-time consultation, while automated documentation within electronic medical records (EMRs) ensures continuity of care. Adhering to data privacy laws and maintaining shared clinical responsibility are essential for complete implementation.

Conclusions

A multidimensional approach integrating technology, regulation, and clinical training is crucial for an advanced TeleStroke network. AI-driven solutions, mobile platforms, and cloud-based infrastructure enhance clinical outcomes and metrics.

Keywords

Telemedicine stroke protocol digital health

Background

The implementation of telemedicine solutions is a significant advancement in overcoming challenges related to access to specialized healthcare, particularly in geographically isolated areas or those with economic development limitations.^1–5 Restricted access to specialists, like neurologists, in these regions can result in delayed diagnoses, inadequate treatment, and consequently unfavorable clinical outcomes for critical health conditions.⁶ This scenario is particularly critical for patients with suspected stroke, whether ischemic or hemorrhagic, as the time to diagnosis and initiation of treatment is crucial for patient recovery and the minimization of functional limitations.

Stroke represents one of the most severe medical conditions, with a high risk of sequelae, in addition to having a profound social and economic impact. It became the leading cause of death in Brazil in 2021 and 2022, highlighting the urgent need for effective strategies for its management and treatment.^{7, 8} Rapid diagnosis and the implementation of appropriate therapies are essential to reduce mortality and improve functional outcomes for patients.^8–10

In this context, telemedicine emerges as a viable and efficient solution. Studies demonstrate that telemedicine in stroke cases is equivalent to an in-person evaluation by a neurologist, significantly increasing the indication and application of reperfusion therapies such as intravenous thrombolysis, which is extremely useful for the recovery of ischemic stroke patients.^11–13 The implementation of telemedicine services for stroke has been shown to improve patient functional outcomes, and it is becoming a cost-effective strategy with significant social impact^13–17 (Table 1). The adoption of this technology not only provides access to the knowledge and experience of neurology specialists regardless of the patient’s geographic location but also optimizes resources, reducing the pressure on overcrowded health centers and allowing a more equitable¹⁸ distribution of high-quality healthcare.^{12, 19, 20}

Table 1.

Summary of Telemedicine Service Objectives.

Objective	Description
Access improvement	Enhance access to early diagnosis and treatment of stroke.
Service integration	Integrate hospital and prehospital actors.
Continuous monitoring	Enable continuous monitoring and follow-up of stroke patients.
Equitable healthcare	Provides high-quality healthcare regardless of geographic location.

Methods

This narrative review aimed to synthesize the latest evidence on TeleStroke network implementation and propose a structured protocol for establishing an advanced TeleStroke network. The term advanced was consensually defined as a network that ensures:

Synchronous communication between healthcare professionals;

Secure storage and logging of sensitive data;

Compliance with data protection regulations;

Remote visualization and manipulation of neuroimaging;

Integration with prehospital emergency services; and

Deployment of validated artificial intelligence (AI) tools to improve functional outcomes and care metrics.

Literature Search Strategy

A structured search was conducted using PubMed, Scopus, Web of Science, and Institute of Electrical and Electronics Engineers (IEEE) Xplore databases to identify relevant publications from January 2000 to February 2025. The following search descriptors were used:

Primary terms: “telemedicine” OR “TeleStroke.”

Sub-descriptors (used alone or in combination): “regulatory aspects,” “ethics,” “information architecture,” “AI,” and “training.”

The selection of these terms was guided by the predefined review sections, ensuring alignment with key aspects of TeleStroke network implementation.

Inclusion criteria comprised peer-reviewed articles, guidelines, and official reports that addressed any aspect of TeleStroke implementation aligned with the predefined framework for advanced networks, which includes synchronous communication, data protection, remote imaging, AI tools, and integration with emergency services. Exclusion criteria included non-English articles and abstracts without full-text access. After removing duplicates, two authors independently screened titles and abstracts, followed by full-text reviews. Disagreements were resolved by consensus. Ultimately, 130 articles were included.

Consensus Process for Review Sections

The review sections were consensually defined by a multidisciplinary team comprising three vascular neurologists and two experts in digital health and computational sciences. The consensus aimed to address the most critical components for the successful deployment of an advanced TeleStroke system.

Data Extraction and Synthesis

The identified studies were screened for relevance and assessed for their contributions to the following core areas:

Regulatory and ethical aspects: Legal frameworks governing TeleStroke implementation.

Computational architecture: Network infrastructure, data storage, and interoperability solutions.

AI integration: AI-driven decision-support systems and automated imaging analysis.

Training requirements: Capacity building for neurologists and healthcare professionals using TeleStroke systems.

Given the narrative nature of this review, no formal meta-analysis was performed. Instead, findings were synthesized to provide best practices, technological strategies, and regulatory considerations. The proposed implementation framework was developed based on evidence from the literature and expert recommendations.

Initial Assessment, Objectives, and Planning

Definition of Objectives of an Advanced TeleStroke Service

First, an agreement with the stakeholders responsible for implementing the TeleStroke service must be established.^{21, 22} This agreement should clearly define the objectives of a care network^{6, 23–25} that ideally encompasses all stages of stroke care.

The objective for a TeleStroke service should be the implementation of a mobile telemedicine service in secondary and tertiary healthcare units, as well as a mobile prehospital service for the assessment,²⁶ transfer,²⁷ treatment, follow-up, and counter-referral of patients with suspected or confirmed cerebrovascular disease.^{20, 28} Additionally, the service should enhance access to early diagnosis and treatment of stroke, coordinating hospital and prehospital actors²⁹ (Table 1).

Needs Assessment

Staffing

Human resources for a stroke telemedicine service should include roles in coordination, management, information technology (IT), training, and regulation. These roles should be organized into tiers with quarterly goals for performance and engagement.^{19, 30} Each position is vital for efficiency and compliance, addressing activations, system issues, and technology implementation bottlenecks,^{31, 32} as detailed in Figure 1.

Figure 1.

Roles of a TeleStroke Service Staff.

Neurologists. Neurologists must be registered medical specialists, hired under defined shifts, ensuring uninterrupted service coverage.^{22, 33} Preferably trained in vascular neurology (VN),^{34, 35} they should be certified in the National Institutes of Health Stroke Scale (NIHSS) and proficient in telemedicine. Responsibilities include patient documentation, adherence to ethical and data protection standards, and prompt teleconsultation responses.^{36, 37} Competitive benefits and performance incentives are recommended.^{21, 22, 37} Compliance with National Medical Council regulations governs practice scope, teleconsultation responsibilities, and electronic record requirements.^{38, 39} Neurologists must respond to all teleconsultation requests within 3-10 min of being called.^{22, 34, 35, 38, 39}

Implementation Challenges

During the implementation of a TeleStroke network, several challenges may hinder service effectiveness. One critical barrier lies in the limitations of commercially available communication platforms, which often lack native integration with electronic health records (EHRs).^{40, 41} This impedes the documentation and sharing of clinical decisions among care team members and can fragment continuity of care. Additionally, many of these tools rely on generic or non-customizable alert systems, which may fail to adequately notify neurologists on call.^{22, 42} Ineffective alarm configurations or user interface issues can result in delayed responses or unavailability of specialists at critical moments, especially in high-demand scenarios.

Moreover, the need for reliable, high-speed internet connectivity remains a barrier in remote or underserved areas.²⁰ In many areas, TeleStroke coverage had to be limited to daytime hours because of inadequate 4G/5G coverage in ambulances or emergency units.^43–45 Staffing is another constraint: ensuring the availability of trained vascular neurologists 24/7 requires contractual adjustments, incentive structures, and shift optimization—factors often overlooked in the planning stages. Finally, resistance to workflow changes and digital tools from local teams can hinder adoption; in the early stages of the implementation framework, several spoke hospitals reported low compliance with teleconsult activation protocols due to insufficient training and unclear responsibilities.^{46, 47}

Implementation in Resource-constrained Settings

Satellite internet, particularly low-Earth orbit (LEO) constellations such as Starlink and OneWeb, has emerged as a feasible option to provide real-time connectivity in underserved or geographically isolated regions.⁴⁸ Pilot studies in Alaska and rural parts of Australia and Chile demonstrated that these services could support the transmission of neuroimaging data and facilitate synchronous video consultations with acceptable latency and bandwidth for acute stroke care.^{19, 36, 49} Although these technologies come with initial setup costs, models based on government subsidies, public–private partnerships, or pooled financing among regional hospitals have proven sustainable in several low- and middle-income countries (LMIC) settings.⁴⁴

Additionally, bandwidth-adaptive platforms can sustain core TeleStroke functionalities under poor connectivity conditions. Systems such as store-and-forward telemedicine allow asynchronous transmission of computed tomography (CT) scans and clinical data, which are uploaded automatically once a connection becomes available. In Kenya and India, for instance, hybrid offline-first telemedicine platforms have enabled structured stroke triage using pre-installed NIHSS and modified Rankin scale (mRS) scoring modules embedded in mobile apps, allowing paramedics and non-specialist physicians to initiate evaluation even without internet access.⁵⁰ Data compression algorithms and AI-enhanced Digital Imaging and Communications in Medicine (DICOM) uploaders—already validated in LMIC settings—help reduce file sizes without compromising diagnostic accuracy.⁵¹

Technical Resources

The proposed telemedicine mobile application⁵² features are summarized in Table 2.

Table 2.

Telemedicine Mobile Application Features.

Feature	Description
Mobile communication application with DICOM viewer	An instant messaging app supporting private/group conversations, linked to medical council registration, providing real-time alerts, a user-friendly interface, low bandwidth usage, and high-resolution video calls.
Embedded DICOM viewer	Integration of a DICOM viewer for viewing imaging exams like CT scans and MRIs directly on mobile devices.
High resolution and advanced features	Advanced features such as zoom, contrast adjustment, and the ability to view multiple imaging sequences with high image resolution.
Security and privacy	Utilizing blockchain for data protection, end-to-end encryption, and secure data storage. For mobile and in-hospital use, identification and real-time data transfer are ensured, with alternative storage for unlinked cases.
Development	Agile software development methodologies for iterative feature delivery and user feedback integration.
Data storage	Storage of patient records by name and identifier, complying with national laws and best practices for electronic health records.
Hosting	Robust, scalable hosting infrastructure that supports increasing users and data volumes, with web and app store accessibility.
Usability testing	Testing with healthcare professionals to ensure an intuitive and practical user interface.
Security testing	Conducting security audits and penetration tests to identify and address potential vulnerabilities.
Pilot launch	Pilot launch in a limited number of centers to assess functionality, effectiveness, and acceptance before a broader rollout.
Data packages and mobile internet bandwidth	Consideration of data usage legislation and telecom challenges to ensure mobile platform access without prohibitive costs.
Data reimbursement	Developing mechanisms for compensating healthcare professionals for data usage via reimbursement or subsidies.
Device provisioning	Providing mobile devices with publicly funded data packages to ensure universal access without data consumption concerns.
Partnerships with telecom providers	Partnerships with telecom operators for optimized bandwidth, subsidized/free access to the platform, and treating it as an essential service.

Note: CT, computerized tomography; DICOM, Digital Imaging and Communications in Medicine; MRI, magnetic resonance imaging.

The adoption of strategies to mitigate the impact of data consumption on healthcare professionals is crucial for the program’s success and sustainability, enabling greater participation and effective implementation in various settings.^{15, 31, 37, 53}

In addition, desktop-based platforms and mobile applications should be integrated. In particular, electronic medical records (EMRs) and imaging software should be integrated for collecting, registering, and viewing patients’ data, in a two-way traffic of information.^{31, 37, 54}

Embedded Software in CT and Magnetic Resonance Imaging (MRI) Suites

Spoke hospital units must integrate secure software within imaging suites to automatically transmit DICOM files to a centralized cloud server. The software must support patient identification and employ data encryption, ensuring privacy and adherence to data protection standards.^{51, 52, 55}

Cloud Server for Instant Access to Images and Text Interactions

A cloud server must securely store DICOM files,^{51, 55, 56} enabling instant access for neurologists during teleconsultations. Backup and redundancy policies⁵⁴ ensure data integrity, while a nationally located server adheres to local laws, facilitating regulatory compliance with standards like the General Data Protection Law (GDPL), and enhancing data privacy and security management.^57–59

Storage Technology

Blockchain technology ensures secure, immutable, and encrypted data storage, enhancing transparency and auditability while preserving patient confidentiality.^{58, 60–63} Data access will comply with health privacy regulations, restricted to authorized personnel, and guided by necessity and minimal sufficiency principles to ensure controlled and monitored access.⁶²

Compliance with the GDPL (or Local Equivalent)

All data storage and processing operations will strictly adhere to the principles established by the GDPL, including transparency, purpose, adequacy, necessity, security, prevention, non-discrimination, and accountability. This compliance not only reinforces the commitment to protecting patients’ data but also establishes a foundation of trust with the platform’s users.^{22, 57–60}

Application Selection Criteria

The proposed selection criteria for TeleStroke applications are represented^{12, 64–66} in Figure 2.

Figure 2.

Application of Selection Criteria.

Prehospital Protocols and Assessment

Healthcare professionals must be trained in advance on the use of telemedicine technology and stroke care protocols, including the rapid assessment of stroke patients using the NIHSS,⁶⁷ the mRS,⁶⁸ and prehospital validated screening tools⁶⁹ (Table 3). The use of the FAST-ED (face, arms, speech, time, eyes, and denial/neglect) scale^{70, 71} is recommended, as discussed in the working groups that are implementing the stroke care pathway for screening patients potentially with proximal vessel occlusion.

Table 3.

Prehospital Stroke Scales.

Scale	Evaluated Signs/Criteria	Description
Cincinnati Prehospital Stroke Scale (CPSS)	– Facial droop (ask the patient to show their teeth or smile) – Arm weakness (ask the patient to extend both arms forward with palms up for 10 s) – Abnormal speech (ask the patient to repeat a simple phrase)	The CPSS is one of the simplest and most widely used scales. The presence of any of these signs is a strong indicator of stroke.
Los Angeles Prehospital Stroke Screen (LAPSS)	– Exclusion criteria (history of seizures, recent trauma, and so on) – Age – History of mobility impairment – Capillary blood glucose – Facial, arm, hand grip, and speech evaluation	The LAPSS includes detailed criteria to help distinguish stroke from other conditions with similar symptoms.
Melbourne Ambulance Stroke Scale (MASS)	– Facial function – Arm strength – Speech – Vision assessment	The MASS is like the CPSS but includes an assessment of vision. It helps in quickly identifying stroke patients for appropriate treatment.
FAST-ED Scale	– Face: Assesses the presence of facial paresis – Normal: 0 points – Partial paresis: 1 point – Complete paresis: 2 points – Arms: Checks arm strength – Normal: 0 points – Drift: 1 point – Immediate fall: 2 points – Speech: Tests speech clarity – Normal: 0 points – Slurred/incomprehensible speech: 1 point – Complete aphasia: 2 points – Time: Importance of recognizing symptoms quickly – Eye: Assesses gaze deviation or vision loss – Normal: 0 points – Gaze deviation: 1 point – Denial/Neglect: Checks for the presence of neglect or denial of disability – Normal: 0 points – Presence of neglect: 1 point	The FAST-ED scale is an expansion of the well-known FAST scale (face, arms, speech, and time), adding elements to potentially identify patients with large vessel occlusions. The FAST-ED scale is particularly useful for emergency teams, including paramedics and EMTs, allowing them to quickly assess stroke severity and communicate critical information to the destination hospital. Patients with higher FAST-ED scores are more likely to benefit from advanced therapies such as mechanical thrombectomy, which is most effective when performed early.

Note: EMTs, emergency medical technicians; FAST-ED, face, arms, speech, time, eyes, and denial/neglect.

Recording teleconsultations in hospital and emergency contexts demands accuracy, security, and seamless integration with EMR.^{22, 58, 60} Ideally, telemedicine platforms should enable direct teleconsultation activation from the EMR via specific tools, supporting efficient communication.^{61, 72–75} Post-consultation, the EMR should be automatically updated with specialist notes, diagnoses, and recommendations, ensuring comprehensive documentation. Both requesting and responding physicians must electronically input and sign their contributions for accessibility and care continuity.^{54, 76}

When Integration Is Not Possible

If direct integration with the EMR is not feasible,^{61, 77} the telemedicine application must maintain its records. The steps include an independent management of secure databases for teleconsultation details,^{46, 47} ensuring patient access and legal compliance.^{50, 61, 63, 73, 78} Records must include patient identification⁷⁸ and comply with privacy regulations, employing encryption and access controls to protect sensitive information.^{59, 62}

Integration with a National Unified Health Record System

For countries with a national unified health record system,^{40, 79, 80} the telemedicine platform must consider integrating with the national system to enable automatic transfer of teleconsultation records,^{41, 81–83} ensuring consolidated access for healthcare professionals.⁷⁸ Additionally, it should grant patients full access to their teleconsultation data, fostering transparency and empowerment in healthcare management.

Ethical and Regulatory Aspects

A detailed record of the opinions provided by the teleconsultant, as well as the information received that supported those opinions, should be maintained.^{84, 85} The physician requesting the opinion or consultation from another physician may or may not be solely responsible for the treatment, as well as for the decisions and recommendations made to the patient.^{46, 86}

The teleconsultant is responsible for providing a quality opinion, based on the available information and in accordance with medical practice standards. They must ensure that their recommendations are grounded in clinical evidence and best practices.^{63, 87}

Although the requesting physician remains primarily responsible for the patient’s treatment, the teleconsultant shares responsibility, particularly regarding the accuracy and appropriateness of the opinion given.⁸⁶ In some jurisdictions, responsibility may be shared, where both physicians can be held accountable in case of an error or omission.⁸⁸

At the conclusion of the teleconsultation, the involved healthcare professional will write a report. This report must be inserted into the patient’s medical record and also into the shared/unified medical record if the patient has one.⁶⁰ Additionally, the report will be sent to the responsible physician and the requesting physician (if they are not the same) through a secure messaging system or equivalent platform in asynchronous cases.

Care Protocol

Healthcare professionals must first be trained in the use of telemedicine technology and stroke care protocols, including rapid assessment using prehospital screening scales and/or the NIHSS⁶⁷ via video or not, the mRS,⁶⁸ and prehospital screening scales. It is suggested to use the FAST-ED scale for screening patients potentially with proximal vessel occlusion and, therefore, potentially eligible for mechanical thrombectomy.^{25, 89, 90}

Screening scales can be embedded in a mobile application within the emergency mobile system, serving as a prerequisite to opening the case discussion.^{69, 91, 92} Only after reaching the minimum score on at least two scales is the communication screen with the Neurology on-call team activated.^{92, 93}

A diagram of the assistance and transfer of suspected stroke cases is illustrated in Figure 3.

Figure 3.

Note: CT, computerized tomography; CTA, computed tomography angiography; DICOM, Digital Imaging and Communications in Medicine; EMS, emergency medical services.

Activation Model

The initial stroke screening should adopt patient identification and documentation via secure platforms and determine the urgency of referral for treatment.^{71, 91} Ideally, identify the patient by civil registration or medical record from the unit initiating the request. For patients on public roads or in mobile services, national identification documents and the date of birth can be useful for patient identification. The person initiating the request should also include essential clinical data, such as triage scale score, NIHSS, and mRS, along with a summary of the main complaint and other relevant clinical information. In cases of direct audio and video activation, patient identification data must be completed.^{94, 95} After the call, both the requesting physician and the specialist must individually document the discussion and the proposed treatment plan. Voice-to-text platforms can be used for transcribing teleconsultations via audio or video, respecting privacy and data security standards.⁹⁶ Teleconsultation supports continuity of care, enabling alignment with local teams and optimizing dehospitalization^{42, 97} within the Stroke Care Line framework.^{14, 98, 99}

Georeferencing

Integrating georeferencing into a mobile teleconsultation platform optimizes stroke patient transport by aligning real-time patient data with hospital capacity,^{28, 100, 101} ensuring timely access to the most appropriate care.⁵²

The implementation model of this algorithm begins with the collection and integration of critical data about the hospital’s installed capacity, including the availability of essential resources such as CT scans, intensive care units (ICUs), and mechanical thrombectomy equipment. These data are provided by local managers and are kept up to date to reflect the real capacity of healthcare services. Thus, the mobile platform can access this information to assess which units are prepared to treat the type of stroke that the patient is presenting.^{44, 102}

The georeferencing algorithm processes this clinical data in combination with the ambulance’s location and the available hospital resources. The choice of hospital is based on proximity and the ability to provide the necessary care, considering the urgency and specificity of the required treatment.^{103, 104} The platform embedded in ambulances can also be integrated with the bed regulation systems of registered hospital units.

Once the decision is made, the mobile telemedicine platform automatically issues a notification to the selected hospital.^104–106 This notification includes detailed information about the patient, such as identification data and current clinical condition, as well as any triage scale results indicating the need for specialized treatment.

Additionally, the patient’s identification data and clinical condition are sent to the hospital’s regulatory center.^{63, 87, 107}

Benchmarks

A robust example of large-scale TeleStroke implementation is provided by the German Stroke Society, which developed detailed national recommendations for organizing teleconsultation services within a TeleStroke network.¹⁰⁸ These networks operate through structured partnerships between tertiary stroke centers and peripheral hospitals, ensuring 24/7 access to neurovascular expertise. The model includes well-defined processes for initial patient registration, bidirectional video consultation, neuroimaging co-assessment, and documented therapeutic decision-making. Each teleconsultation adheres to strict data protection standards and includes quality assurance mechanisms such as time-stamped process tracking and standardized reporting templates. Importantly, the system allows even semi-elective cerebrovascular consultations and the structured coordination of inter-hospital transfers, enhancing access to thrombectomy and specialized care.

In parallel, the TeleSpecialists^® network in the United States demonstrates the scalability of a private TeleStroke provider model. Operating across more than 200 hospitals, the system offers real-time neurologist access, integrates cloud-based imaging and documentation, and incorporates performance dashboards to monitor quality metrics.⁴² Evidence from this network shows significant reductions in door-to-needle times and increased rates of intravenous thrombolysis in previously underserved hospitals.¹

An additional instructive case comes from Spain, where a comprehensive national survey revealed that 12 of the 17 autonomous communities have implemented TeleStroke networks, covering over 22% of the country’s population.¹⁰⁹ These programs range from regional networks—such as in Andalusia and Catalonia, which integrate emergency medical services and vascular neurologists—to smaller provincial or hospital-based systems. Andalusia’s centralized model is particularly noteworthy: it allows 93% of the population to reach a vascular neurologist within 30 min via telemedicine.¹⁰⁹ Despite challenges such as suboptimal videoconferencing quality and limited registry access, these programs have increased the accessibility and equity of stroke care, especially in rural areas.

Another notable case is the South Australian TeleStroke Network,³⁶ which successfully implemented a regionally coordinated service covering rural and underserved areas. Through a hub-and-spoke model supported by trained stroke physicians and standardized workflows, the network demonstrated reduced stroke mortality and improved adherence to reperfusion therapy protocols.³⁶ The initiative emphasized digital infrastructure investment, clinical training, and mobile imaging integration, making acute stroke treatment feasible in geographically dispersed locations. Similarly, the Neurovascular Network for Stroke Patients (NEVAS) network in Bavaria (Germany),¹⁶ maintained robust clinical outcomes and time metrics throughout the COVID-19 pandemic by relying on telemedicine-supported stroke triage and management, even in emergency-declared periods.

In the United States, the Remote Evaluation of Acute Ischemic Stroke (REACH) TeleStroke program,¹² which connects rural hospitals with academic stroke centers, has shown sustained benefits in expanding access to reperfusion therapies and decreasing the rate of unnecessary patient transfers.

Embeddable AI Tools

In addition to neuroimaging visualization and manipulation tools, algorithms for processing radiology images can be embedded in TeleStroke platforms.¹¹⁰ These algorithms should be hosted on cloud servers to maintain the speed and efficiency of the TeleStroke platform.^110–112

These tools use deep learning algorithms trained to identify signs of large vessel occlusions (LVOs) in computed tomography angiography (CTA) images.^{113, 114} Additional neuroimaging processing capabilities include the detection of intracranial hemorrhage, subarachnoid hemorrhage, and automated scoring of collateral assessments and the Alberta Stroke Program Early CT Score (ASPECTS).^{37, 115, 116}

Natural language processing (NLP) tools can analyze large volumes of unstructured clinical data, such as medical notes, radiology reports, and EHRs. In cases of ischemic stroke, NLP can help identify patterns in symptoms and clinical signs within medical records that suggest an ischemic stroke.^{117, 118} This enables automated alerts that expedite response times for healthcare professionals. Additionally, these tools can be integrated into clinical decision-support systems to aid in interpreting complex data, such as imaging results or medical histories.^{117, 118}

Speech-to-text technologies offer the ability to transcribe verbal interactions between healthcare professionals and patients or among medical team members. These tools enable physicians to document cases and medical orders quickly and accurately, without manual typing.^{111, 119–121} This can streamline the care process and ensure that all essential information is recorded immediately. Furthermore, healthcare professionals can use voice commands to interact with EMRs or to activate specific stroke treatment protocols.¹²¹

The benefits of these tools include reducing diagnostic and treatment times.^{49, 122, 123} However, significant challenges remain, such as the need for AI model training on data specific to the local population, integration of these technologies with existing systems, and ensuring data privacy and security.^{119, 120, 124}

Moreover, AI has shown increasing relevance in addressing several implementation barriers in TeleStroke services. In the Catalonia Stroke Program (Spain), AI-based tools were integrated into regional imaging systems to automatically identify LVOs and prioritize patients for thrombectomy referral. This not only improved decision-making accuracy but also reduced treatment delays in high-volume centers.¹⁷ In Switzerland, the use of AI-assisted real-time video analysis in prehospital settings enabled paramedics to perform remote NIHSS scoring, facilitating early stroke identification and streamlining communication with stroke units.²⁴ Beyond acute triage, AI has also enhanced TeleStroke documentation and coordination. In a US pilot study,^{117, 118} NLP tools were employed to extract structured data—such as NIHSS, mRS, and thrombolysis decisions—from free-text emergency department notes, improving continuity of care and enabling real-time audits of stroke care metrics. In South Korea, a rural TeleStroke initiative¹⁰⁶ implemented AI-driven decision-support protocols to guide non-specialist physicians through acute stroke management steps, based on patient symptoms and uploaded imaging, resulting in faster therapeutic interventions and greater protocol adherence.¹⁰⁶

Training Proposal

The integration of TeleStroke training into VN fellowship programs is a strategic response to the increasing demand for TeleStroke. This can be a strategy for the sustainability of services; however, training initiatives are still scarce in many programs, especially among those in LMIC.¹²⁵

Pioneered by the McGovern Medical School, this model involves a formal TeleStroke rotation, allowing fellows to conduct remote consultations under structured supervision. Training encompasses orientation to telehealth platforms, shadowing experienced stroke attendings, and progressively independent patient evaluations. Didactics, such as case conferences, further support skill development.¹²⁶

Training in TeleStroke involves developing a comprehensive set of skills to ensure effective, efficient, and professional remote stroke care. The evaluation of fellows should focus on the following core competencies¹²⁷:

Technical skills: Proficiency in using telemedicine tools, such as Zoom, adjusting the field of view, and maneuvering robots (if applicable). Fellows should recognize and address connection issues, including knowing whom to contact for technical assistance.

History-taking: The ability to gather an efficient yet detailed patient history relevant to acute therapy decisions, including contraindications and recommendations to treatments like intravenous tissue plasminogen activator (IV tPA), treatment windows, vital signs, and laboratory results.

Imaging interpretation: Competence in remotely reviewing imaging to identify acute stroke findings, such as early ischemic changes, hyperdense signs, or hemorrhages, and to guide therapy decisions.

Handling complex cases: Knowledge of managing challenging clinical scenarios, such as recent myocardial infarction or previous strokes (ischemic or hemorrhagic).

Neurological examination: Skill in conducting quick, accurate stroke and neurologic assessments, and in communicating complex NIHSS components to staff.

Team communication with spoke staff: Ability to emphasize urgency during stroke codes, communicate plans clearly, and expedite treatment, such as calculating tPA dosing.

Team communication with spoke physicians: Proficiency in discussing findings and treatment plans.

Team communication with patients and families: Clear and concise communication of clinical assessments and plans, including obtaining consent.

Identifying endovascular candidates: Recognizing potential candidates for endovascular therapy and navigating transfer logistics, including Emergency Medical Treatment and Labor Act (EMTALA) rules and coordination with hubs.

Professionalism: Developing strong rapport with spoke physicians, nursing staff, patients, and families, ensuring a collaborative and respectful environment.

Regulatory frameworks, such as the Accreditation Council for Graduate Medical Education (ACGME) milestones, highlight the need for nuanced supervision protocols, particularly distinguishing between direct and indirect supervision in time-sensitive scenarios. The post-pandemic regulatory environment underscores the importance of institutional collaboration to adapt to shifting supervision and billing requirements.^{126, 127}

Emerging modalities like mobile TeleStroke, enabling prehospital consultations in ambulances, add a layer of complexity to training, requiring tailored approaches. Simulation-based training, virtual rotations, and inter-program collaborations are proposed to ensure equitable access to TeleStroke education across institutions, including those without established networks.^{126, 127}

A formal evaluation framework categorizes trainees’ development into five levels, focusing on their ability to manage technical systems, conduct efficient assessments, communicate effectively, and coordinate complex care.¹²⁶

1. Level 1

a. Conducts a complete and relevant history and examination.

b. Manages acute stroke patients remotely using telemedicine tools, addressing acute risk factors effectively.

2. Level 2

a. Performs efficient and complete assessments, identifying surgical or endovascular candidates and high-risk patients (e.g., intracranial hemorrhage).

b. Resolves technical issues or seeks appropriate assistance when needed.

3. Level 3

a. Integrates history and examination findings into treatment decisions.

b. Manages acute stroke cases without unnecessary delays, efficiently coordinating with spoke hospital teams for transfers or advanced care.

4. Level 4

a. Demonstrates proficiency in remote assessments and treatments with minimal delays.

b. Facilitates patient transfers for higher levels of care, ensuring compliance with regulations (e.g., EMTALA).

c. Communicates effectively with spoke physicians, staff, patients, and families, maintaining professionalism throughout.

5. Level 5

a. Serves as a role model for other trainees by demonstrating excellence in telemedicine consultations.

b. Efficiently manages acute stroke cases, triages post-tPA patients, and coordinates advanced care and transfers seamlessly.

Financing Mechanisms and Sustainability of TeleStroke Networks

Public funding remains central, especially in resource-limited settings and rural regions where the benefits of TeleStroke are potentially highest but adoption is lowest.⁴³ Government programs may subsidize infrastructure costs—including telecommunication equipment and connectivity—while academic centers often serve as hubs, absorbing initial technology investments.⁴³ Partnerships with telecommunications providers have also emerged as viable models to expand broadband access and minimize fixed connectivity costs. In some models, costs are defrayed by pooling smaller hospitals into centralized service contracts, coordinated through regional networks or resource centers that aggregate demand and streamline credentialing and technical support.^{128, 129}

Long-term sustainability requires ongoing strategies beyond the initial implementation phase. One critical pillar is integrating TeleStroke into national reimbursement frameworks, including procedure-based payments and bundled-care models. Despite growing evidence of cost-effectiveness, current reimbursement systems often fail to capture the operational expenses of TeleStroke at the originating site.^{129, 130} To offset this, hospital systems may explore direct billing for remote neurologist consultations and reallocation of avoided transfer costs. Sustaining clinician engagement is equally essential; structured training programs, digital literacy development, and the incorporation of TeleStroke workflows into clinical routines increase utilization and acceptance over time. Embedding TeleStroke practices within broader telemedicine strategies also promotes scalability, enabling health systems to leverage shared platforms, reduce redundancy, and achieve financial and operational resilience.^{43, 128}

Conclusion

The use of appropriate information and communication technologies allows specialized medical treatment for people living in distant locations or with few professionals. TeleStroke allows to expand the actions of professionals and community health agents, integrating them into the services offered by hospitals and reference centers.

Our proposed protocol differs from existing TeleStroke models by offering a comprehensive and integrated framework that explicitly addresses the architecture and security of digital infrastructures, a topic often underrepresented in previous medical literature. In addition to highlighting traditional elements such as specialist availability and teleconsultation workflows, our review emphasizes critical innovations in software structure, including embedded DICOM viewers, blockchain-based secure storage, and compliance with national data protection regulations and EHR assessments. Furthermore, we reviewed advanced technologies such as AI for neuroimaging triage and NLP, .369solutions that ensure interoperability with EHR and national health data repositories. This multidimensional approach ensures that the proposed protocol is technically robust and suitable for broad implementation in diverse health systems.

Authors Contribution

João Brainer Clares de Andrade: Design, supervision, and data collection; Daniela Laranja G Rodrigues, Claudia Galindo Novoa, and Gisele Sampaio Silva: Supervision, data collection, and final review. Ivan Pisa, Sophia Oliveira Querobin, Marialdo Augusto Cordeiro Souza, Milena Roberta Guimarães Brianti, and Octavio Marques Pontes-Neto: Data collection and final review.

Footnotes

Declaration of Conflicting Interests

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

Ethical Approval

Ethical approval was obtained from the relevant ethics committee or Institutional Review Board (IRB).

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors Andrade, Laranja, Pontes-Neto, Martins, Novoa, and Silva received research grant from Ministry of Health (Brazil), and Pisa received research grant from National Council for Scientific and Technological Development (Brazil).

Informed Consent

The participant has provided informed consent for the submission of the article to the journal.

ORCID iDs

João Brainer Clares de Andrade

Milena Roberta Guimarães Brianti

References

Jagolino-Cole

, Zha

, Sangha

, Song

, Majersik

JJ.

Careers in telestroke: Toward a virtual and tangible purpose. Stroke . 2023;54(5):E220–E223. doi:10.1161/STROKEAHA.122.039149

Regenhardt

, Awad

, Kraft

, . Characterizing reasons for stroke thrombectomy ineligibility among potential candidates transferred in a hub-and-spoke network. Stroke Vasc Interv Neurol . 2022;2(5). doi:10.1161/svin.121.000282

Mohamed

, Elsherif

, Legere

, Fatima

, Shuaib

, Saqqur

Is telestroke more effective than conventional treatment for acute ischemic stroke? A systematic review and meta-analysis of patient outcomes and thrombolysis rates. Int J Stroke . 2024;19(3):280–292. doi:10.1177/17474930231206066

Garcia-Esperon

, Bladin

, Kleinig

, . The acute telestroke model of care in Australia: A potential roadmap for other emergency medical services? Med J Aust . 2022;216(10): 498–500. doi:10.5694/mja2.51519

Olatunji

, Kokori

, Isarinade

, . Revolutionizing stroke care in Africa: A mini review of the transformative potential of mobile stroke units. Medicine (United States) . 2023;102(44): E35899. doi:10.1097/MD.0000000000035899

Richard

, Mehrotra

, Schwamm

, . Improving population access to stroke expertise via telestroke: Hospitals to target and the potential clinical benefit. J Am Heart Assoc . 2022;11(8). doi:10.1161/JAHA.122.025559

Mosconi

, Paciaroni

Treatments in ischemic stroke: Current and future. Eur Neurol . 2022;85(5):349–366. doi:10.1159/000525822

Nguyen

, Klein

, Berberich

, . Late window imaging selection for endovascular therapy of large vessel occlusion stroke: An international survey. Stroke Vasc Interv Neurol . 2023;3(1). doi:10.1161/svin.122.000595

Yan

, Viswanathan

, Estrada

, . Abstract TMP48: Effects of telestroke on endovascular therapy timeliness and outcomes. Stroke . 2022;53(Suppl_1). doi:10.1161/str.53.suppl_1.tmp48

10.

Kimmel

, Walder

, Ovalle

, . Stroke survivors’ telemedicine experiences during the COVID-19 pandemic: A phenomenological investigation. Digit Health . 2024;10. doi:10.1177/20552076241281464

11.

Clausen

, Greve

, Müller

, Kristoffersen

, Schytz

HW.

Telemedicine in headache care: A systematic review. Cephalalgia . 2022;42(13):1397–1408. doi:10.1177/03331024221111554

12.

Demaerschalk

, Aguilar

, Ingall

, . Stroke telemedicine for Arizona Rural Residents, the legacy telestroke study. Telemed Rep . 2022;3(1):67–78. doi:10.1089/tmr.2022.0002

13.

, Fawcett

AP.

Outcomes in patients with acute stroke treated at a comprehensive stroke center using telemedicine versus in-person assessments. J Telemed Telecare . 2023. doi:10.1177/1357633X231169115

14.

Zha

, Trevino

, Ankrom

, . Inpatient teleneurology follow-up has comparable outcomes to in-person neurology follow-up. Neurol Clin Pract . 2022;12(6):E181–E188. doi:10.1212/CPJ.0000000000200096

15.

De La Torre-Diéz

, López-Coronado

, Vaca

, Aguado

, De Castro

Cost-utility and cost-effectiveness studies of telemedicine, electronic, and mobile health systems in the literature: A systematic review. Telemed e-Health . 2015;21(2): 81–85. doi:10.1089/tmj.2014.0053

16.

Masouris

, Kellert

, Müller

, . Performance and clinical outcomes in telestroke remain robust during the COVID-pandemic: Insight into the NEVAS network. J Neurol . 2024;271(9):6045–6055. doi:10.1007/s00415-024-12578-9

17.

López Seguí

, Franch Parella

, Gironès García

, . A cost-minimization analysis of a medical record-based, store and forward and provider-to-provider telemedicine compared to usual care in Catalonia: More agile and efficient, especially for users. Int J Environ Res Public Health . 2020;17(6):2008. doi:10.3390/ijerph17062008

18.

Wishart

, Hamouda

, Elsaady

, Aslam

, Perala

, Khan

The adoption and impact of telemedicine on health equity: A narrative review from the Jamaican context. Cureus . 2024. doi:10.7759/cureus.69650

19.

Hendrickx

, Kuznia

, Maneval

Use of telestroke to improve access to care for rural patients with stroke symptoms. Crit Care Nurse . 2023;43(5):49–56. doi:10.4037/ccn2023505

20.

Arumuganathan

, Adcock

, Espinosa

, Findley

Analysis of telestroke usage in rural critical access emergency departments. Telemed e-Health . 2023;29(12):1828–1833. doi:10.1089/tmj.2022.0408

21.

Boss

, Heaton

, Gulati

Abstract WMP30: How to beat the metrics with telestroke: A community hospital primary stroke center experience. Stroke . 2022;53(Suppl_1). doi:10.1161/str.53.suppl_1.wmp30

22.

Demaerschalk

, Berg

, Chong

, . American telemedicine association: Telestroke guidelines. Telemed e-Health . 2017;23(5):376–378. doi:10.1089/tmj.2017.0006

23.

Kraft

, Regenhardt

, Awad

, . Spoke-administered thrombolysis improves large-vessel occlusion early recanalization: The real-world experience of a large academic hub-and-spoke telestroke network. Stroke Vasc Interv Neurol . 2023;3(1). doi:10.1161/svin.122.000427

24.

Thilemann

, Traenka

, Schaub

, . Real-time video analysis allows the identification of large vessel occlusion in patients with suspected stroke: Feasibility trial of a “telestroke” pathway in Northwestern Switzerland. Front Neurol . 2023;14. doi:10.3389/fneur.2023.1232401

25.

Poongkunran

, Ulep

, Stuntz

, . Diagnostic accuracy of telestroke consultation: A Louisiana based tele-network experience. Front Neurol . 2023;14. doi:10.3389/fneur.2023.1141059

26.

Ebinger

, Fiebach

, Audebert

HJ.

Mobile computed tomography: Prehospital diagnosis and treatment of stroke. Curr Opin Neurol . 2015;28(1):4–9. doi:10.1097/WCO.0000000000000165

27.

Pervez

, Silva

, Masrur

, . Remote supervision of IV-tPA for acute ischemic stroke by telemedicine or telephone before transfer to a regional stroke center is feasible and safe. Stroke . 2010;41(1). doi:10.1161/STROKEAHA.109.560169

28.

Bluma

, Bowser

, Abelt

, Bennett

, Landreman

Abstract WP32: Addressing disparate clinical stroke resources with a telestroke taskforce. Stroke . 2023;54(Suppl_1). doi:10.1161/str.54.suppl_1.wp32

29.

McCormick

, Estrada

, Whitney

, . Teleneurology comprehensive inpatient consultations expedite access to care and decreases hospital length of stay. Neurohospitalist . 2021;11(3): 229–234. doi:10.1177/19418744211000951

30.

Barragán-Prieto

, Pérez-Sánchez

, Moniche

, . Express improvement of acute stroke care accessibility in large regions using a centralized telestroke network. Eur Stroke J . 2022;7(3):259–266. doi:10.1177/23969873221101282

31.

Vishnu

, Bhatia

, Khurana

, . Smartphone-based telestroke vs “stroke physician” led acute stroke management (SMART INDIA): A protocol for a cluster-randomized trial. Ann Indian Acad Neurol . 2022;25(3):422. doi:10.4103/aian.aian_1052_21

32.

Yadav

, Zurmehly

, Mion

, . Abstract TMP40: Standardized interdisciplinary team handoff communication improves thrombectomy door-to-groin time in acute ischemic stroke evaluated by telestroke. Stroke . 2022;53(Suppl_1). doi:10.1161/str.53.suppl_1.tmp40

33.

Mantese

, da Silva Aquino

, Figueira

, Rodrigues

, Basso

, Da Rosa

PR.

Telemedicine as support for primary care referrals to neurologists: Decision-making between different specialists when guiding the case over the phone. Arq Neuropsiquiatr . 2021;79(4):299–304. doi:10.1590/0004-282X-ANP-2020-0137

34.

Rodríguez Cruz

, Pérez Sánchez

, Cuello

, . Workload of on-call emergency room neurologists in a Spanish tertiary care centre. A one-year prospective study. Neurología . 2014;29:193–199. www.elsevier.es/neurologia

35.

Hoang

, Nuwer

MR.

Changes in emergency department coverage for the neurologist. Neurol Clin Pract . 2013;3:334–340. www.neurology.org/cp

36.

Goh

, Hillier

, Kelly

, . Implementation of the South Australian Regional Telestroke service is associated with improved care quality and lower stroke mortality: A retrospective cohort study. Aust J Rural Health . 2023;31(5):878–885. doi:10.1111/ajr.13012

37.

Gutiérrez-Zúñiga

, Uclés

, Sánchez-Manso

, . The use of a smartphone application to improve stroke code workflow metrics: A pilot study in a comprehensive stroke centre. Digit Health . 2022;8. doi:10.1177/20552076221137252

38.

, Parker

, Jagolino

, . Telemedicine can replace the neurologist on a mobile stroke unit. Stroke . 2017;48(2):493–496. doi:10.1161/STROKEAHA.116.015363

39.

Roeder

, Leira

EC.

The role of the vascular neurologist in optimizing stroke care. Neurol Clin . 2024;42(3):739–752. doi:10.1016/j.ncl.2024.03.007

40.

Kataria

, Ravindran

Electronic health records: A critical appraisal of strengths and limitations. J R Coll Physicians Edinb . 2020;50(3):262–268. doi:10.4997/jrcpe.2020.309

41.

Janett

, Yeracaris

PP.

Electronic medical records in the American health system: Challenges and lessons learned. Cien Saude Colet . 2020;25(4):1293–1304. doi:10.1590/1413-81232020254.28922019

42.

Pestka

, Boes

, Ramezani

, . Implementing telestroke in the inpatient setting: Identifying factors for success. Stroke . 2024;55(6):1517–1524. doi:10.1161/STROKEAHA.123.046024

43.

Alishahi Tabriz

, Turner

, Williams

, Babu

, North

, Shea

CM.

Association of financial factors and telemedicine adoption for heart attack and stroke care among rural and urban hospitals: A longitudinal study. Telemed e-Health . 2022;28(6):781–788. doi:10.1089/tmj.2021.0341

44.

Tunkl

, Agarwal

, Ramage

, . Telemedicine networks for acute stroke: An analysis of global coverage, gaps, and opportunities. Int J Stroke . 2024. doi:10.1177/17474930241298450

45.

International Telecommunication Union (ITU). Population Coverage, by Mobile Network Technology (SDG 9.c.1)—5G Subset, ITU DataHub [dataset]. 2024.

46.

Jallal

, Serhier

, Berrami

, Bennani Othmani

Current state and prospects of telemedicine in Morocco: Analysis of challenges, initiatives, and regulatory framework. Cureus . 2023. doi:10.7759/cureus.50963

47.

Arora

, Huda

, Verma

, Khetan

, Sangwan

RK.

Challenges, barriers, and facilitators in telemedicine implementation in India: A scoping review. Cureus . 2024. doi:10.7759/cureus.67388

48.

Furlepa

, Tenderenda

, Kozłowski

, Marczak

, Wierzba

, Śliwczyński

Recommendations for the development of telemedicine in Poland based on the analysis of barriers and selected telemedicine solutions. Int J Environ Res Public Health . 2022;19(3):1221. doi:10.3390/ijerph19031221

49.

Irfan

, Yaqoob

The digital lifeline: Telemedicine and artificial intelligence synergy as a catalyst for healthcare equity in Pakistan. Cureus . 2024. doi:10.7759/cureus.54017

50.

Haskew

, Rø

, Saito

, . Implementation of a cloud-based electronic medical record for maternal and child health in rural Kenya. Int J Med Inform . 2015;84(5):349–354. doi:10.1016/j.ijmedinf.2015.01.005

51.

Salazar

, Useche

, Bermúdez

, . Accuracy and reliability of the recommendation for IV thrombolysis in acute ischemic stroke based on interpretation of head CT on a smartphone or a laptop. Am J Roentgenol . 2020;214(4):877–884. doi:10.2214/AJR.19.21896

52.

Martins

SCO

, Weiss

, Almeida

, . Validation of a smartphone application in the evaluation and treatment of acute stroke in a comprehensive stroke center. Stroke . 2020;51(1):240–246. doi:10.1161/STROKEAHA.119.026727

53.

Greenberg

, Javeed

, Zhang

, . Current and future applications of mobile health technology for evaluating spine surgery patients: A review. J Neurosurg Spine . 2023;38(5):617–626. doi:10.3171/2022.12.SPINE221302

54.

Baker

, Stanley

Telemedicine technology: A review of services, equipment, and other aspects. Curr Allergy Asthma Rep . 2018;18(11):60. doi:10.1007/s11882-018-0814-6

55.

Shkirkova

, Akam

, Huang

, . Feasibility and utility of an integrated medical imaging and informatics smartphone system for management of acute stroke. Int J Stroke . 2017;12(9):953–960. doi:10.1177/1747493017694386

56.

John

, Stock

, Cerejo

, . Brain imaging using mobile CT: Current status and future prospects. J Neuroimaging . 2016;26(1):5–15. doi:10.1111/jon.12319

57.

Clark

, Capuzzi

, Harrison

Telemedicine: Medical, legal and ethical perspectives. Med Sci Monit . 2010;16(12): RA261–RA272.

58.

Ansarian

, Baharlouei

Applications and challenges of telemedicine: Privacy-preservation as a case study. Arch Iran Med . 2023;26(11):654–661. doi:10.34172/aim.2023.96

59.

Andreadis

, Muellers

, Lin

, . Navigating privacy and security in telemedicine for primary care. Am J Manag Care . 2024;30(Spec. No. 6):SP459–SP463. doi:10.37765/ajmc.2024.89553

60.

Mishkin

, Zabinski

, Holt

, Appelbaum

PS.

Ensuring privacy in telemedicine: Ethical and clinical challenges. J Telemed Telecare . 2023;29(3):217–221. doi:10.1177/1357633X221134952

61.

Hardin

WD.

Clinical information systems. Semin Pediatr Surg . 2000;9(1):35–39. doi:10.1016/S1055-8586(00)70006-0

62.

Lee

, Kung

, Udayasankaran

, . An architecture and management platform for blockchain-based personal health record exchange: Development and usability study. J Med Internet Res . 2020;22(6):e16748. doi:10.2196/16748

63.

Houtchens

, Allen

, Clemmer

, Lindberg

, Pedersen

Telemedicine protocols and standards: Development and implementation. J Med Syst . 1995;19(2):93–119. doi:10.1007/BF02257060

64.

Worthmann

, Winzer

, Schuppner

, Gumbinger

, Barlinn

Telestroke networks for area-wide access to endovascular stroke treatment. Neurol Res Pract . 2023;5(1). doi:10.1186/s42466-023-00237-9

65.

Mazzon

, Delfino

, Mirelis

, . Breaking down barriers: Easter Island’s first telestroke thrombolysis experience and case report. J Stroke Cerebrovasc Dis . 2022;31(6). doi:10.1016/j.jstrokecerebrovasdis.2022.106412

66.

Jun-O’Connell

, Sivakumar

, Henninger

, . Outcomes of telestroke inter-hospital transfers among intervention and non-intervention patients. J Clin Med Res . 2023;15(6):292–299. doi:10.14740/jocmr4945

67.

de Andrade

JBC

, de Paula Pacheco

, Camilo

, . An algorithm for the National Institute of Health Stroke Scale assessment: A multicenter, two-arm and cluster randomized study. J Stroke Cerebrovasc Dis . 2024;33(7):107723. doi:10.1016/j.jstrokecerebrovasdis.2024.107723

68.

C de Andrade

, Quinn

, Carbonera

, . An automated flowchart for the Modified Rankin Scale assessment: A multicenter inter-rater agreement analysis. Int J Stroke . 2024;19(7):789–797. doi:10.1177/17474930241246157

69.

Walter

, Hooo RJ

Van

, Lorenzano

, Saban

Validation of a cloud-based tele-stroke system reliability in determining National Institutes of Health Stroke Scale scores for acute ischemic stroke screening in the emergency department. Front Neurol . 2022;13:973165.

70.

Lima

, Silva

, Furie

, . Field assessment stroke triage for emergency destination. Stroke . 2016;47(8):1997–2002. doi:10.1161/STROKEAHA.116.013301

71.

Nogueira

, Silva

, Lima

, . The FAST-ED app: A smartphone platform for the field triage of patients with stroke. Stroke . 2017;48(5):1278–1284. doi:10.1161/STROKEAHA.116.016026

72.

Ford

, Reuber

Comparisons of communication in medical face-to-face and teleconsultations: A systematic review and narrative synthesis. Health Commun . 2024;39(5):1012–1026. doi:10.1080/10410236.2023.2201733

73.

Apathy

, Zabala

, Gomes

, Spaar

, Krevat

, Ratwani

RM.

Telemedicine and in-person visit modality mix and electronic health record use in primary care. JAMA Netw Open . 2024;7(4):e248060. doi:10.1001/jamanetworkopen.2024.8060

74.

O’Shea

, Vaughan Sarrazin

, Nassar

, . Using electronic medical record notes to measure ICU telemedicine utilization. J Am Med Inform Assoc . 2017;24(5):969–974. doi:10.1093/jamia/ocx029

75.

Zvárová

EFMI symposium on “electronic health record healthcare registers and telemedicine”. Int J Med Inform . 2006;75(3-4):183–184. doi:10.1016/j.ijmedinf.2005.07.021

76.

Banchev

, Goldmann

, Marquardt

, . Impact of telemedicine tools on record keeping and compliance in haemophilia care. Hamostaseologie . 2019;39(04):347–354. doi:10.1055/s-0038-1676128

77.

Jiang

, Ross

, Bai

Unveiling the adoption and barriers of telemedicine in US hospitals: A comprehensive analysis (2017-2022). J Gen Intern Med . 2024;39(13):2438–2445. doi:10.1007/s11606-024-08853-0

78.

Doarn

, Merrell

RC.

A national strategy for telemedicine and e-health. Telemed e-Health . 2007;13(3):243–244. doi:10.1089/tmj.2007.9982

79.

Logan

, Faeldon

, Kent

, Ong

, Marsden

A scoping review of stroke services within the Philippines. BMC Health Serv Res . 2024;24(1):1006. doi:10.1186/s12913-024-11334-z

80.

Birinci

Ş.

A digital opportunity for patients to manage their health: Turkey national personal health record system (The e-Nabız). Balkan Med J . 2023;40(3):215–221. doi:10.4274/balkanmedj.galenos.2023.2023-2-77

81.

Nahvijou

, Esmaeeli

, Kalaghchi

, Sheikhtaheri

, Zendehdel

Using electronic health record system to establish a national patient’s registry: Lessons learned from the cancer registry in Iran. Int J Med Inform . 2023;180:105245. doi:10.1016/j.ijmedinf.2023.105245

82.

Eisman

, Chen

, Wu

, . Learning health system linchpins: Information exchange and a common data model. J Am Med Inform Assoc . 2024. doi:10.1093/jamia/ocae277

83.

Celuppi

, Mohr

ETB

, Felisberto

, . Dez anos do Prontuário Eletrônico do Cidadão e-SUS APS: em busca de um Sistema Único de Saúde eletrônico. Rev Saude Publica . 2024;58(1):23. doi:10.11606/s1518-8787.2024058005770

84.

Bonica

, Johns

, Jadvar

Telehealth and telemedicine: Regulatory and medicolegal landscape. Clin Nucl Med . 2024;49(7):644–647. doi:10.1097/RLU.0000000000005254

85.

Ricci

, Caraffa

, Gibelli

Telemedicine as a strategic tool to enhance the effectiveness of care processes: Technological and regulatory evolution over the past two decades. Healthcare (Switzerland) . 2023;11(5). doi:10.3390/healthcare11050734

86.

Al-Alawy

, Moonesar

IA.

Perspective: Telehealth—Beyond legislation and regulation. SAGE Open Med . 2023;11. doi:10.1177/20503121221143223

87.

Downes

, Mervin

, Byrnes

, Scuffham

PA.

Telemedicine for general practice: A systematic review protocol. Syst Rev . 2015;4(1). doi:10.1186/s13643-015-0115-2

88.

Fields

BG.

Regulatory, legal, and ethical considerations of telemedicine. Sleep Med Clin . 2020;15(3):409–416. doi:10.1016/j.jsmc.2020.06.004

89.

Nair

, Khan

, Stang

, . Thrombolysis in stroke mimics: Comprehensive stroke centers vs telestroke sites. Can J Neurol Sci . 2023;50(6):838–844. doi:10.1017/cjn.2022.325

90.

Baratloo

, Rahimpour

, Abushouk

, Safari

, Lee

, Abdalvand

Effects of telestroke on thrombolysis times and outcomes: A meta-analysis. Prehosp Emerg Care . 2018;22(4): 472–484. doi:10.1080/10903127.2017.1408728

91.

Khor

, Bown

, Barrett

, Counsell

, MacLeod

, Reid

Pre-hospital notification is associated with improved stroke thrombolysis timing. J R Coll Physicians Edinb . 2015;45(3): 190–195. doi:10.4997/jrcpe.2015.303

92.

McCartan

, Lee

, Bejleri

, Murphy

, Hickey

, Williams

The impact of telemedicine enabled pre-hospital triage in acute stroke—A protocol for a mixed methods systematic review. HRB Open Res . 2022;5:32. doi:10.12688/hrbopenres.13514.1

93.

Kim

, Groombridge

, Romero

, Clare

, Fitzgerald

MC.

Decision support capabilities of telemedicine in emergency prehospital care: Systematic review. J Med Internet Res . 2020;22(12). doi:10.2196/18959

94.

Eyupoglu

, Altug

, Sener

, . Effect of teleconsultation on the application of thrombolytic therapy in stroke patients in the emergency department. Ir J Med Sci . 2024;193(2):1019–1024. doi:10.1007/s11845-023-03497-1

95.

Candefjord

, Andersson Hagiwara

, Sjöqvist

, . Video support for prehospital stroke consultation: Implications for system design and clinical implementation from prehospital simulations. BMC Med Inform Decis Mak . 2024;24(1):146. doi:10.1186/s12911-024-02539-7

96.

Thorup

, Bundgaard

, Pedersen

PU.

Transformation of health professional/patient caring relationships through information and communication technologies used in telemedicine: A scoping review protocol. JBI Database System Rev Implement Rep . 2019;17(4):470–478. doi:10.11124/JBISRIR-2017-003661

97.

Dawson

, Anderson

, Binns

, . Strategy-training post-stroke via tele-rehabilitation: A pilot randomized controlled trial. Disabil Rehabil . 2024;46(1):67–76. doi:10.1080/09638288.2022.2154397

98.

Alabyad

, Lemuel-Clarke

, Antwan

, . Telemedicine impact on post-stroke outpatient follow-up in an academic healthcare network during the COVID-19 pandemic. J Stroke Cerebrovasc Dis . 2023;32(8):107213. doi:10.1016/j.jstrokecerebrovasdis.2023.107213

99.

Wong

AKC

, Kwok

VWY

, Wong

FKY

, . Improving post-acute stroke follow-up care by adopting telecare consultations in a nurse-led clinic: Study protocol of a hybrid implementation-effectiveness trial. J Adv Nurs . 2024;80(3):1222–1231. doi:10.1111/jan.15960

100.

Alhajala

, Hendricks-Jones

, Shawver

, . Expansion of telestroke coverage in community hospitals: Unifying stroke care and reducing transfer rate. Ann Neurol . 2024;95(3):576–582. doi:10.1002/ana.26839

101.

Zachrison

, Richard

, Schwamm

, . Abstract WP48: How to optimize population access to acute stroke expertise. Stroke . 2022;53(Suppl_1). doi:10.1161/str.53.suppl_1.wp48

102.

Kurunawai

, Chen

, Willcour

, . Implementation of an optimised tele-medicine platform for stroke in South Australia improves patient care. Front Neurol . 2024;15. doi:10.3389/fneur.2024.1428198

103.

Rogers

, Madathil

, Agnisarman

, . A systematic review of the implementation challenges of telemedicine systems in ambulances. Telemed e-Health . 2017;23(9):707–717. doi:10.1089/tmj.2016.0248

104.

Mazya

, Berglund

, Ahmed

, . Implementation of a prehospital stroke triage system using symptom severity and teleconsultation in the Stockholm Stroke Triage Study. JAMA Neurol . 2020;77(6):691. doi:10.1001/jamaneurol.2020.0319

105.

Valenzuela Espinoza

, Van Hooff

, De Smedt

, . Development and pilot testing of 24/7 in-ambulance telemedicine for acute stroke: Prehospital stroke study at the Universitair Ziekenhuis Brussel-Project. Cerebrovasc Dis . 2016;42(1-2):15–22. doi:10.1159/000444175

106.

Espiritu

, San Jose

MCZ.

A call for a stroke referral network between primary care and stroke-ready hospitals in the Philippines. Neurologist . 2021;26(6):253–260. doi:10.1097/NRL.0000000000000357

107.

Jun

, Yang

, Kim

, . Telemedicine protocols for the management of patients with acute spontaneous intracerebral hemorrhage in rural and medically underserved areas in Gangwon state: Recommendations for doctors with less expertise at local emergency rooms. J Korean Neurosurg Soc . 2023. doi:10.3340/jkns.2023.0199

108.

Wiestler

, Zickler

, Erdur

, . Recommendations for the organization of the teleconsultation service in a telestroke network. Neurol Res Pract . 2024;6(1):24. doi:10.1186/s42466-024-00318-3

109.

Barragán-Prieto

, Pérez-Sánchez

, Castellanos

, González

, Montaner

Estudio de la situación actual del Teleictus en España. Neurología . 2025;40(2):182–190. doi:10.1016/j.nrl.2023.05.003

110.

Rezaei

, Khouzani

, . Integrating artificial intelligence into telemedicine: Revolutionizing healthcare delivery . Kindle.

111.

Kuziemsky

, Maeder

, John

, . Role of artificial intelligence within the telehealth domain. Yearb Med Inform . 2019;28(1):35–40. doi:10.1055/s-0039-1677897

112.

Roßkopf

, Meder

Healthcare 4.0—Medicine in transition. Herz . 2024. doi:10.1007/s00059-024-05267-w

113.

Iqbal

, Cortés Jaimes

, Makineni

, . Reimagining healthcare: Unleashing the power of artificial intelligence in medicine. Cureus . 2023. doi:10.7759/cureus.44658

114.

, Li

, Wei

, Li

GY.

Innovation and challenges of artificial intelligence technology in personalized healthcare. Sci Rep . 2024;14(1). doi:10.1038/s41598-024-70073-7

115.

Arora

, Gaekwad

, Evans

, . Diagnostic utility of computed tomography perfusion in the telestroke setting. Stroke . 2022;53(9):2917–2925. doi:10.1161/STROKEAHA.122.038798

116.

Salazar

, Useche

, Granja

, . Reliability and accuracy of individual Alberta Stroke Program Early CT Score regions using a medical and a smartphone reading system in a telestroke network. J Telemed Telecare . 2021;27(7):436–443. doi:10.1177/1357633X19881863

117.

Pool

, Indulska

, Sadiq

Large language models and generative AI in telehealth: A responsible use lens. J Am Med Inform Assoc . 2024. doi:10.1093/jamia/ocae035

118.

Moura

, Jones

, Sheikh

, . Implications of large language models for quality and efficiency of neurologic care: Emerging issues in neurology. Neurology . 2024;102(11): e209497. doi:10.1212/WNL.0000000000209497

119.

Amjad

, Kordel

, Fernandes

A review on innovation in healthcare sector (telehealth) through artificial intelligence. Sustainability (Switzerland) . 2023;15(8). doi:10.3390/su15086655

120.

Setya Rintyarna

, Insantuan

, Widiawati

, Yuridian Purwoko

, Muhammadiyah Jember

, Haekto-Dinas Kesehatan Kabupaten Timor Tengah Utara

Telehealth in remote areas: A new artificial intelligence-based model. Int J Sci Soc . 2023;5. http://ijsoc.goacademica.com

121.

El Naamani

, Musmar

, Gupta

, . The artificial intelligence revolution in stroke care: A decade of scientific evidence in review. World Neurosurg . 2024;184:15–22. doi:10.1016/j.wneu.2024.01.012

122.

Hulsen

Applications of the metaverse in medicine and healthcare. Adv Lab Med . 2024;5(2):159–165. doi:10.1515/almed-2023-0124

123.

Bit-Avragim

, Bousquet

, Cantù

, Omboni

, Ravot

, Tunnah

The evolving reality of digital health. Digit Health . 2024;10. doi:10.1177/20552076241277646

124.

Pencina

, McCall

, Economou-Zavlanos

NJ.

A federated registration system for artificial intelligence in health. JAMA . 2024;332(10):789. doi:10.1001/jama.2024.14026

125.

Mendes

, Poppe

, Lima

, . Comprehensive portrait of stroke fellowship training in Brazil: A national survey study. J Stroke Cerebrovasc Dis . 2024;33(6):107697. doi:10.1016/j.jstrokecerebrovasdis.2024.107697

126.

Guzik

, Jagolino-Cole

, Mijalski Sells

, . Telestroke training: Considerations for expansion of vascular neurology program requirements. Stroke . 2025;56(1):209–218. doi:10.1161/STROKEAHA.124.047826

127.

Jagolino

, Jia

, Gildersleeve

, . A call for formal telemedicine training during stroke fellowship. Neurology . 2016;86(19):1827–1833. doi:10.1212/WNL.0000000000002568

128.

Zachrison

, Richard

, Mehrotra

Paying for telemedicine in smaller rural hospitals: Extending the technology to those who benefit most. JAMA Health Forum . 2021;2(8): E211570. doi:10.1001/jamahealthforum.2021.1570

129.

Naqvi

, Cohen

, Kim

, . Inequities in telemedicine use among patients with stroke and cerebrovascular diseases. Neurol Clin Pract . 2023;13(2). doi:10.1212/cpj.0000000000200148

130.

Bagot

, Moloczij

, Barclay-Moss

, Vu

, Bladin

, Cadilhac

DA.

Sustainable implementation of innovative, technology-based health care practices: A qualitative case study from stroke telemedicine. J Telemed Telecare . 2020;26(1-2): 79–91. doi:10.1177/1357633X18792380

A Narrative Review and a Proposed Protocol for Implementing an Advanced TeleStroke Service

Abstract

Purpose

Methods

Results

Conclusions

Keywords

Background

Summary of Telemedicine Service Objectives.

Methods

Literature Search Strategy

Consensus Process for Review Sections

Data Extraction and Synthesis

Initial Assessment, Objectives, and Planning

Definition of Objectives of an Advanced TeleStroke Service

Needs Assessment

Staffing

Roles of a TeleStroke Service Staff.

Implementation Challenges

Implementation in Resource-constrained Settings

Technical Resources

Telemedicine Mobile Application Features.

Embedded Software in CT and Magnetic Resonance Imaging (MRI) Suites

Cloud Server for Instant Access to Images and Text Interactions

Storage Technology

Compliance with the GDPL (or Local Equivalent)

Application Selection Criteria

Application of Selection Criteria.

Prehospital Protocols and Assessment

Prehospital Stroke Scales.

When Integration Is Not Possible

Integration with a National Unified Health Record System

Ethical and Regulatory Aspects

Care Protocol

Activation Model

Georeferencing

Benchmarks

Embeddable AI Tools

Training Proposal

Financing Mechanisms and Sustainability of TeleStroke Networks

Conclusion

Authors Contribution

Footnotes

Declaration of Conflicting Interests

Ethical Approval

Funding

Informed Consent

ORCID iDs

References