Pathological images play a crucial role in the diagnosis of critically ill cancer patients. Since cancer patients often seek medical assistance when their condition is severe, doctors face the urgent challenge of completing accurate diagnoses and developing surgical plans within a limited timeframe. The complexity and diversity of pathological images require a significant investment of time from specialized physicians for processing and analysis, which can lead to missing the optimal treatment window.
Purpose
Current medical decision support systems are challenged by the high computational complexity of deep learning models, which demand extensive data training, making it difficult to meet the real-time needs of emergency diagnostics.
Method
This study addresses the issue of emergency diagnosis for malignant bone tumors such as osteosarcoma by proposing a Lightened Boundary-enhanced Digital Pathological Image Recognition Strategy (LB-DPRS). This strategy optimizes the self-attention mechanism of the Transformer model and innovatively implements a boundary segmentation enhancement strategy, thereby improving the recognition accuracy of tissue backgrounds and nuclear boundaries. Additionally, this research introduces row-column attention methods to sparsify the attention matrix, reducing the computational burden of the model and enhancing recognition speed. Furthermore, the proposed complementary attention mechanism further assists convolutional layers in fully extracting detailed features from pathological images.
Results
The DSC value of LB-DPRS strategy reached 0.862, the IOU value reached 0.749, and the params was only 10.97 M.
Conclusion
Experimental results demonstrate that the LB-DPRS strategy significantly improves computational efficiency while maintaining prediction accuracy and enhancing model interpretability, providing powerful and efficient support for the emergency diagnosis of malignant bone tumors such as osteosarcoma.
ZhongJHuYZhangG, et al.An updated systematic review of radiomics in osteosarcoma: utilizing CLAIM to adapt the increasing trend of deep learning application in radiomics. Insights Imaging2022; 13: 38.
2.
SadikMOzlemKHuseyinM, et al.Attributes of cancer patients admitted to the emergency department in one year. World J Emerg Med2014; 5: 85–90.. PMID: 25215154; PMCID: PMC4129880.
3.
DaiTWuJ, A novel feature alignment-based cell nucleus identification scheme for medical decision-making systems,” 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Lisbon, Portugal, 2024, 3222–3225, https://doi.org/10.1109/BIBM62325.2024.10822119.
4.
XuY, et al.Twenty-year outcome of prevalence, incidence, mortality and survival rate in patients with malignant bone tumors.Int J Cancer2024; 154: 226–240.
5.
FanLSowmyaAMeijeringE, et al.Cancer survival prediction from whole slide images with self-supervised learning and slide consistency. IEEE Trans Med ImagingMay 2023; 42: 1401–1412.
6.
TomaszewskiJE. Overview of the role of artificial intelligence in pathology: the computer as a pathology digital assistant.Artif Intell Pathol. Elsevier2025: 343–368.
7.
LuoTZengJGouF. Continuous refinement-based digital pathology image assistance scheme in medical decision-making systems. IEEE J Biomed Health Inform2024; 28: 2091–2102.
8.
McGenityCClarkeELJenningsC, et al.Artificial intelligence in digital pathology: a systematic review and meta-analysis of diagnostic test accuracy. npj Digit Med2024; 7: 14.
9.
LiBLiuFLvB. Cytopathology image analysis method based on high- resolution medical representation learning in medical decision-making system. Complex Intell Syst2024; 10: 4253–4274.
10.
TangXLiuJ. Artificial intelligence multiprocessing scheme for pathology images based on transformer for nuclei segmentation. Complex Intell Syst2024; 10: 5831–5849.
11.
LiuJXiaoC. Research on artificial-intelligence-assisted medicine: a survey on medical artificial intelligence. Diagnostics2024; 14: 1472.
12.
CaiePDDimitriouNArandjelovićO. Precision medicine in digital pathology via image analysis and machine learning.Artif Intell Pathol. Elsevier2025: 233–257.
13.
LiLHeKZhuX. A pathology image segmentation framework based on deblurring and region proxy in medical decision-making system. Biomed Signal Process Control2024; 95: 106439.
14.
ZhangDY, et al.Implementation of digital pathology and artificial intelligence in routine pathology practice.Lab Invest2024; 104: 102111.
15.
ZhanXLongH. A semantic fidelity interpretable-assisted decision model for lung nodule classification. Int J Comput Assist Radiol Surg2024; 19: 625–633.
16.
HeKZhuJLiL. Two-stage coarse-to-fine method for pathological images in medical decision-making systems. IET Image Process2024; 18: 175–193.
17.
HeKLiWuJ. Asymptotic multilayer pooled transformer based strategy for medical assistance in developing countries. Comput Electr EngOctober 2024; 119: 109493.
18.
HagerPJungmannFHollandR, et al.Evaluation and mitigation of the limitations of large language models in clinical decision-making. Nat Med2024; 30: 2613–2622.
19.
PanYYeQ. Intelligent cell images segmentation system: based on SDN and moving transformer. Sci Rep2024; 14: 24834.
20.
ZuB, et al.SwinT-SRNet: swin transformer with image super-resolution reconstruction network for pollen images classification.Eng Appl Artif Intell2024; 133: 108041.
21.
ChenHMeiKZhouY, et al.Auxiliary diagnosis of breast cancer based on machine learning and hybrid strategy. IEEE Access2023; 11: 96374–96386.
22.
ChenHXiongWWuJ, et al.Decision-Making model based on ensemble method in auxiliary medical system for non-small cell lung cancer. IEEE Access2020; 8: 171903–171911.
23.
BhattiUATangHWuG, et al.Deep learning with graph convolutional networks: an overview and latest applications in computational intelligence. Int J Intell Syst2023; 2023: 8342104, 28 pages.
24.
MengY, et al.Transformer-based deep learning models for predicting permeability of porous media.Adv Water Res2023; 179: 104520.
25.
LiaoZWangJShiZ, et al.Revolutionary potential of ChatGPT in constructing intelligent clinical decision support systems. Ann Biomed Eng2024; 52: 125–129.
26.
LiW, et al.Artificial intelligence auxiliary diagnosis and treatment system for breast cancer in developing countries. J Xray Sci Technol2024; 32: 395–413.
27.
HeKZhuJLiL. FASNet: feature alignment-based method with digital pathology images in assisted diagnosis medical system. Heliyon2024; 10: e40350.
28.
HuangZ. Medical assisted-segmentation system based on global feature and stepwise feature integration for feature loss problem. Biomed Signal Process Control2024; 89: 105814.
29.
MayfieldJDBaileyKBorkowskiAA, et al.Pilot lightweight denoising algorithm for multiple sclerosis on spine MRI. J Digit Imaging2023; 36: 1877–1884.
30.
PfaffL, et al.Self-supervised MRI denoising: leveraging stein’s unbiased risk estimator and spatially resolved noise maps. Sci Rep2023; 13: 22629.
31.
ZhongX, et al.An intelligent MRI assisted diagnosis and treatment system for osteosar-coma based on super-resolution. Complex Intell Syst2024; 10: 1–20.
32.
YuM, et al.RIRGAN: an end-to-end lightweight multi-task learning method for brain MRI su-per-resolution and denoising. Comput Biol Med2023; 167: 107632.
33.
ZhangZ, et al.Gradient-Guided Network with Fourier Enhancement for Glioma Segmentation in Mul-timodal 3D MRI. J Biomed Health Inform2024; 28: 6778–6790.
34.
SongPLiJFanH, et al.TGDAUNet: transformer and GCNN based dual-branch attention UNet for medical image segmentation. Comput Biol Med2023; 167: 107583.
35.
FanXZhouJJiangX, et al.CSAP-UNet: convolution and self-attention paralleling network for medical image segmentation with edge enhancement. Comput Biol Med2024; 172: 108265.
36.
ZhangXLiuYGuoS, et al.EG-Unet: edge-guided cascaded networks for automated frontal brain segmentation in MR images. Comput Biol Med2023; 158: 106891.
37.
ElazabNGab AllahWElmogyM. Computer-aided diagnosis system for grading brain tumor using histopathology images based on color and texture features. BMC Med Imaging2024; 24: 177.
38.
LiX, et al.MH2AFormer: An Efficient Multiscale Hierarchical Hybrid Attention With a Transformer for Bladder Wall and Tumor Segmentation. IEEE J Biomed Health Inform2024; 28: 4772–4784.
39.
ZhangGHeZZhangY, et al.SC-Net: symmetrical conical network for colorectal pa-thology image segmentation. Comput Methods Programs Biomed2024; 248: 108119.
40.
OteKHashimotoFOnishiY, et al.List-Mode PET image reconstruction using deep image prior. IEEE Trans Med ImagingJune 2023; 42: 1822–1834.
41.
DaiTGuanPChenZ, et al.Opportunistic routing for mobile edge computing: a community detected and task priority aware approach. Comput Netw2025; 258: 111000.
42.
BahadirCDOmarMRosenthalJ, et al.Artificial intelligence applications in histopathology. Nat Rev Electr Eng2024; 1: 93–108.
43.
HanYHolsteGDingY, et al.Radiomics-Guided global-local transformer for weakly supervised pathology localization in chest X-rays. IEEE Trans Med ImagingMarch 2023; 42: 750–761.
44.
GuanPYuKWeiW, et al.Big data analytics on lung cancer diagnosis framework with deep learning. IEEE/ACM Trans Comput Biol BioinfJuly–Aug. 2024; 21: 757–768.
45.
WuFZhuangX. Minimizing estimated risks on unlabeled data: a new formulation for semi-supervised medical image segmentation. IEEE Trans Pattern Anal Mach Intell1 May 2023; 45: 6021–6036.
46.
FanLSowmyaAMeijeringE, et al.Cancer survival prediction from whole slide images with self-supervised learning and slide consistency. IEEE Trans Med ImagingMay 2023; 42: 1401–1412.
47.
SubramanianSGhafouriAScheufeleKM, et al.Ensemble in-version for brain tumor growth models with mass effect. IEEE Trans Med ImagingApril 2023; 42: 982–995.
48.
NiuYLingZZhuJ. Pathological image segmentation method based on multiscale and dual attention. Int J Intell Syst2024: Article ID 9987190, 19pages. https://doi.org/10.1155/int/9987190
49.
HeM, et al.Manual and semi-automated measurement and calculation of osteosarcoma treatment effect using whole slide image and qupath.Pediatr Dev Pathol2024; 27: 32–38.
50.
HeKWuJ. Imagesegmentation technology based on transformer in medical decision-making system. IET Image Process2023; 17: 3040–3054.
51.
DeepakKVBharanidharanR. A survey on deep learning and machine learning techniques over histopathology image based osteosarcoma detection.Multimed Tools Appl2024: 1–24.
52.
YuanTZengJ. A medically assisted model for precise segmentation of osteosarcoma nuclei on pathological images. IEEE J Biomed Health InformAUGUST. 2023; 27: 3982–3993.
53.
Wu J, Dai T and Guan P. FedAPT: joint adaptive parameter freezing and resource allocation for communication-efficient. IEEE Internet Things J 2024; 11: 19520–19536.
54.
Gou F and Wu J., et al.Optimization of edge server group collaboration architecture strategy in IoT smart cities application. Peer Peer Netw Appl2024; 17: 3110.
55.
ZhanXLiuJLongH, et al.An intelligent auxiliary framework for bone malignant tumor lesion segmentation in medical image analysis. Diagnostics2023; 13: 23.
56.
HeKQinY. A novel medical decision-making system based on multi-scale feature enhancement for small samples. Mathematics2023; 11: 2116.
57.
LvBLiuFLiY, et al.Artificial intelligence-aided diagnosis solution by enhancing the edge features of medical images. Diagnostics2023; 13: 1063.
58.
WeiHLvBLiuF, et al.A tumor MRI image segmentation framework based on class-correlation pattern aggregation in medical decision-making system. Mathematics2023; 11: 1187.
59.
XuYFanXYangY. Numerical solution of ruin probability of continuous time model based on optimal adaptive particle swarm optimization-triangular neural network algorithm. Soft Comput2023; 27: 14321–14335.
60.
HuangJ, et al.User trust and evaluation mechanism based on community classification calculation in opportunistic social networks. J Intell Fuzzy Syst2024; 46: 2091–2108.
61.
YuSYaoX. Advances on immunotherapy for osteosarcoma. Mol Cancer2024; 23: 92.
62.
ZhouZ, et al.Self-supervised tumor segmentation and prognosis prediction in osteosarcoma using multiparametric MRI and clinical characteristics.Comput Methods Programs Biomed2024; 244: 107974.
63.
RonnebergerO.FischerP.BroxT.U-net: Convolutional networks for biomedical image segmentation. In Proceedings of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention, Munich, Germany, 5–9 October 2015; Springer.
64.
OktayOSchlemperJLe FolgocL, et al.Attention u-net: Learning where to look for the pancreas. arXiv preprint arXiv:1804.03999 (2018).
65.
BadrinarayananVKendallACipollaR. Segnet: a deep convolutional encoder-decoder architec-ture for image segmentation. IEEE Trans Pattern Anal Mach Intell2017; 39: 2481–2495.
66.
GaoYZhouMMetaxasDN, Utnet: a hybrid transformer architecture for medical image seg-mentation, in Medical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th In-ternational Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part III 24. Springer, 2021, pp. 61–71.
67.
ChenSNiuJDengC, et al.CE-Net: a coordinate embedding network for mismatching removal. IEEE Access2021; 9: 147634–147648.
68.
ChenJLuYYuQ, et al.Transunet: Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.043062021.
69.
CaoHWangYChenJ, et al.Swin-unet: unet-like pure trans-former for medical image segmentation. in Computer vision–ECCV 2022 workshops: Tel Aviv, Israel, October 23–27, 2022, proceedings, part III. 2023; Springer, pp. 205–218.
70.
DiakogiannisFIWaldnerFCaccettaP, et al.ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data.ISPRS J Photogramm Remote Sens2020; 162: 94–114.
