{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,28]],"date-time":"2026-10-28T15:05:47Z","timestamp":1793199947320,"version":"4.5.0"},"reference-count":20,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Colorectal cancer liver metastasis (CRLM) presents considerable challenges in both diagnosis and prognosis, as conventional approaches often are limited by subjectivity, variability, and limited efficiency. Recent advances in deep learning have shown great potential for automated extraction of pathological features, offering improved diagnostic accuracy and more reliable prognostic predictions.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and validate a multi-model ensemble deep learning framework (colon cancer liver metastasis network [CLM-Net]) for the automatic recognition and prognostic prediction of CRLM from pathological images, thereby enhancing diagnostic accuracy and clinical applicability.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n A total of 197 pathologically annotated CRLM cases were collected and integrated from publicly available datasets (Kaggle and The Cancer Imaging Archive) to construct high-quality training and independent test sets. CLM-Net was built upon base convolutional neural network architectures including VGG16, DeepLab-v3, and U-Net, incorporating multi-scale atrous convolutions, a squeeze-and-excitation attention mechanism, a conditional random field refinement module, and transfer learning strategies. The framework was comprehensively evaluated using 5-fold cross-validation and independent testing for classification, segmentation, and prognostic prediction tasks. For survival prediction, 1024-dimensional image feature vectors extracted by CLM-Net were analyzed using logistic regression and random forest classifiers, with Kaplan-Meier curves and log-rank tests used for survival analysis.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n CLM-Net demonstrated superior performance in pathological image recognition, achieving 94% accuracy, 92% recall, 93%\n F<\/jats:italic>\n 1<\/jats:sub>\n -score, and an area under the receiver operating characteristic curve of 0.96 on the independent test set, outperforming single models. For survival prediction, CLM-Net with multi-scale attention achieved an area under the receiver operating characteristic curve of 0.864. Kaplan-Meier analysis revealed its significantly stronger risk stratification ability compared with VGG16, U-Net, and DeepLab-v3. Precision-recall curves and heatmaps further confirmed the model\u2019s high robustness and generalizability in unseen samples. Clinical evaluation by pathologists indicated strong interpretability and diagnostic utility, with a concordance rate of 90%.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n The proposed CLM-Net framework, through the integration of diverse deep learning architectures and mechanisms, markedly improved the recognition and prognostic prediction of CRLM, demonstrating excellent generalization and clinical translation potential. Future studies will focus on multi-center validation and integration of multimodal features to further optimize its role in precision medicine.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/73311","type":"journal-article","created":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T18:00:21Z","timestamp":1775584821000},"page":"e73311-e73311","source":"Crossref","is-referenced-by-count":1,"title":["Automatic Recognition and Prognostic Prediction of Colorectal Liver Metastases Using a Multi-Scale Deep Learning Framework: Model Development and Validation Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-7343-294X","authenticated-orcid":false,"given":"Houwen","family":"Long","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-0455-0382","authenticated-orcid":false,"given":"Feng","family":"Ying","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-2958-7787","authenticated-orcid":false,"given":"Shi","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-6825-0210","authenticated-orcid":false,"given":"Long","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,4,7]]},"reference":[{"issue":"24","key":"R1","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.adf5464","article-title":"Single-cell and spatial transcriptome analysis reveals the cellular heterogeneity of liver metastatic colorectal cancer","volume":"9","author":"Wang","journal-title":"Sci Adv"},{"issue":"7","key":"R2","doi-asserted-by":"publisher","first-page":"1209","DOI":"10.1016\/j.cmet.2023.04.013","article-title":"Extracellular vesicles in fatty liver promote a metastatic tumor microenvironment","volume":"35","author":"Wang","journal-title":"Cell Metab"},{"issue":"9","key":"R3","doi-asserted-by":"publisher","first-page":"541","DOI":"10.1038\/s41568-021-00383-9","article-title":"The immunological and metabolic landscape in primary and metastatic liver cancer","volume":"21","author":"Li","journal-title":"Nat Rev Cancer"},{"issue":"15","key":"R4","doi-asserted-by":"publisher","DOI":"10.3390\/cancers14153707","article-title":"Deep neural network models for colon cancer screening","volume":"14","author":"Kavitha","journal-title":"Cancers (Basel)"},{"issue":"3","key":"R5","doi-asserted-by":"publisher","first-page":"266","DOI":"10.5114\/pg.2023.129494","article-title":"The role of deep learning in diagnosing colorectal cancer","volume":"18","author":"Bousis","journal-title":"Prz Gastroenterol"},{"issue":"23","key":"R6","doi-asserted-by":"publisher","DOI":"10.3390\/s22239250","article-title":"Colon cancer diagnosis based on machine learning and deep learning: modalities and analysis techniques","volume":"22","author":"Tharwat","journal-title":"Sensors (Basel)"},{"issue":"1","key":"R7","doi-asserted-by":"publisher","DOI":"10.1186\/s12967-022-03615-0","article-title":"Global research trends and foci of artificial intelligence-based tumor pathology: a scientometric study","volume":"20","author":"Shen","journal-title":"J Transl Med"},{"issue":"7","key":"R8","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13071314","article-title":"AI techniques of dermoscopy image analysis for the early detection of skin lesions based on combined CNN features","volume":"13","author":"Olayah","journal-title":"Diagnostics (Basel)"},{"issue":"13","key":"R9","doi-asserted-by":"publisher","DOI":"10.2174\/1573405619666230217100130","article-title":"Cancer detection based on medical image analysis with the help of machine learning and deep learning techniques: a systematic literature review","volume":"19","author":"Sood","journal-title":"CMIR"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.1016\/j.critrevonc.2024.104528","article-title":"Early cancer detection using deep learning and medical imaging: a survey","volume":"204","author":"Ahmad","journal-title":"Crit Rev Oncol Hematol"},{"issue":"22","key":"R11","doi-asserted-by":"publisher","first-page":"2199","DOI":"10.1001\/jama.2017.14585","article-title":"Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer","volume":"318","author":"Ehteshami Bejnordi","journal-title":"JAMA"},{"issue":"5","key":"R12","doi-asserted-by":"publisher","DOI":"10.3390\/biomimetics8050442","article-title":"Artificial intelligence in regenerative medicine: applications and implications","volume":"8","author":"Nosrati","journal-title":"Biomimetics (Basel)"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2021.798962","article-title":"Accelerating hyperparameter tuning in machine learning for alzheimer\u2019s disease with high performance computing","volume":"4","author":"Zhang","journal-title":"Front Artif Intell"},{"key":"R14","doi-asserted-by":"publisher","DOI":"10.3389\/fnins.2024.1363930","article-title":"Adaptive feature medical segmentation network: an adaptable deep learning paradigm for high-performance 3D brain lesion segmentation in medical imaging","volume":"18","author":"Zaman","journal-title":"Front Neurosci"},{"issue":"1","key":"R15","doi-asserted-by":"publisher","DOI":"10.1186\/s12880-024-01479-y","article-title":"Enhancing image-based diagnosis of gastrointestinal tract diseases through deep learning with EfficientNet and advanced data augmentation techniques","volume":"24","author":"Zubair Rahman","journal-title":"BMC Med Imaging"},{"issue":"2","key":"R16","doi-asserted-by":"publisher","first-page":"430","DOI":"10.1038\/s41591-022-02134-1","article-title":"Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer","volume":"29","author":"Foersch","journal-title":"Nat Med"},{"issue":"10221","key":"R17","doi-asserted-by":"publisher","first-page":"350","DOI":"10.1016\/S0140-6736(19)32998-8","article-title":"Deep learning for prediction of colorectal cancer outcome: a discovery and validation study","volume":"395","author":"Skrede","journal-title":"The Lancet"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-56820-w","article-title":"Color-CADx: a deep learning approach for colorectal cancer classification through triple convolutional neural networks and discrete cosine transform","volume":"14","author":"Sharkas","journal-title":"Sci Rep"},{"issue":"4","key":"R19","doi-asserted-by":"publisher","first-page":"04TR01","DOI":"10.1088\/1361-6560\/abcd17","article-title":"Interpretation and visualization techniques for deep learning models in medical imaging","volume":"66","author":"Huff","journal-title":"Phys Med Biol"},{"issue":"1","key":"R20","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1515\/ijb-2022-0113","article-title":"Survival analysis using deep learning with medical imaging","volume":"20","author":"Morrison","journal-title":"Int J Biostat"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e73311","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:03:53Z","timestamp":1782842633000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e73311"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,7]]},"references-count":20,"URL":"https:\/\/doi.org\/10.2196\/73311","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,7]]},"article-number":"v14i24e73311"}}