{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:47:55Z","timestamp":1782841675340,"version":"3.54.5"},"reference-count":48,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Deep learning (DL) algorithms for digital breast tomosynthesis (DBT) have proliferated, demonstrating emerging potential in enhancing lesion detection and classification.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to compare the diagnostic performance of DL algorithms for DBT with that of radiologists of varying experience and assess the clinical impact of DL assistance.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n A systematic search of PubMed, Embase, Web of Science, and the Cochrane Library was conducted up to November 8, 2025. Included studies compared the performance of stand-alone DL algorithms for DBT, radiologist interpretation alone, and DL-assisted diagnosis. Study quality was assessed using the Prediction Model Risk of Bias Assessment Tool+Artificial Intelligence (PROBAST+AI). Performance metrics were pooled using bivariate random effects and generalized linear mixed models.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n A total of 13 studies with 38,565 patients were included in the final analysis. Stand-alone DL algorithms achieved a pooled sensitivity of 0.88 (95% CI 0.80-0.93), specificity of 0.74 (95% CI 0.59-0.85), and area under the receiver operating characteristic curve (AUC) of 0.89 (95% CI 0.86-0.92). While DL performance showed no statistically significant difference compared to all radiologists (AUC=0.89 vs 0.88;\n P<\/jats:italic>\n =.64) or senior radiologists (AUC=0.89 vs 0.90;\n P<\/jats:italic>\n =.48), DL demonstrated significantly superior sensitivity compared to junior radiologists (0.88 vs 0.76;\n P<\/jats:italic>\n =.03). Notably, DL assistance did not statistically improve diagnostic metrics for radiologists across any experience level. Meta-regression identified validation methods as a significant source of heterogeneity.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n DL algorithms for DBT exhibited strong diagnostic proficiency and showed higher sensitivity than junior radiologists, suggesting their potential utility as adjunctive tools to help reduce oversight in less experienced settings. However, given that DL assistance did not significantly elevate overall human performance, current models act primarily as supplementary aids rather than definitive clinical tools. Future prospective multimodal studies are warranted to validate these findings and optimize clinical integration.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/91659","type":"journal-article","created":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T16:55:40Z","timestamp":1778086540000},"page":"e91659-e91659","source":"Crossref","is-referenced-by-count":0,"title":["Deep Learning Algorithms Versus Radiologists in Digital Breast Tomosynthesis for Breast Cancer Detection: Systematic Review and Meta-Analysis"],"prefix":"10.2196","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-1665-3217","authenticated-orcid":false,"given":"Shewen","family":"Lyu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-6406-8147","authenticated-orcid":false,"given":"Zepeng","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-9962-9612","authenticated-orcid":false,"given":"Yujing","family":"Mu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2193-4841","authenticated-orcid":false,"given":"Luyao","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3043-8874","authenticated-orcid":false,"given":"Xiaohua","family":"Pei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,5,6]]},"reference":[{"issue":"6","key":"R1","doi-asserted-by":"publisher","first-page":"524","DOI":"10.3322\/caac.21754","article-title":"Breast cancer statistics, 2022","volume":"72","author":"Giaquinto","journal-title":"CA Cancer J Clin"},{"issue":"3","key":"R2","doi-asserted-by":"publisher","first-page":"229","DOI":"10.3322\/caac.21834","article-title":"Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries","volume":"74","author":"Bray","journal-title":"CA Cancer J Clin"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1148\/radiol.2019180760","article-title":"Digital breast tomosynthesis: concepts and clinical practice","volume":"292","author":"Chong","journal-title":"Radiology"},{"issue":"2","key":"R4","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1148\/rg.2021200101","article-title":"Digital breast tomosynthesis: update on technology, evidence, and clinical practice","volume":"41","author":"Gao","journal-title":"Radiographics"},{"key":"R5","doi-asserted-by":"publisher","DOI":"10.1016\/j.breast.2024.103767","article-title":"Digital breast tomosynthesis for breast cancer diagnosis in women with dense breasts and additional breast cancer risk factors: a systematic review","volume":"77","author":"Raichand","journal-title":"Breast"},{"issue":"6","key":"R6","doi-asserted-by":"publisher","DOI":"10.1002\/mco2.70247","article-title":"Deep learning in digital breast tomosynthesis: current status, challenges, and future trends","volume":"6","author":"Wang","journal-title":"MedComm (2020)"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102049","article-title":"Applying deep learning in digital breast tomosynthesis for automatic breast cancer detection: a review","volume":"71","author":"Bai","journal-title":"Med Image Anal"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2017.10.014","article-title":"Improving digital breast tomosynthesis reading time: a pilot multi-reader, multi-case study using concurrent Computer-Aided Detection (CAD)","volume":"97","author":"Balleyguier","journal-title":"Eur J Radiol"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1148\/radiol.211105","article-title":"Artificial intelligence for reducing workload in breast cancer screening with digital breast tomosynthesis","volume":"303","author":"Shoshan","journal-title":"Radiology"},{"issue":"2","key":"R10","doi-asserted-by":"publisher","first-page":"202","DOI":"10.1007\/s11547-023-01751-1","article-title":"The role of an artificial intelligence software in clinical senology: a mammography multi-reader study","volume":"129","author":"Bassi","journal-title":"Radiol Med"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.m2632","article-title":"Preferred reporting items for systematic review and meta-analysis of diagnostic test accuracy studies (PRISMA-DTA): explanation, elaboration, and checklist","volume":"370","author":"Salameh","journal-title":"BMJ"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1136\/bmj-2024-082505","article-title":"PROBAST+AI: an updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods","volume":"388","author":"Moons","journal-title":"BMJ"},{"issue":"7","key":"R13","doi-asserted-by":"publisher","first-page":"760","DOI":"10.1016\/j.jclinepi.2014.01.006","article-title":"Applying Grading of Recommendations Assessment, Development and Evaluation (GRADE) to diagnostic tests was challenging but doable","volume":"67","author":"Gopalakrishna","journal-title":"J Clin Epidemiol"},{"issue":"5","key":"R14","doi-asserted-by":"publisher","first-page":"621","DOI":"10.1177\/0272989X08319957","article-title":"Bivariate random effects meta-analysis of ROC curves","volume":"28","author":"Arends","journal-title":"Med Decis Making"},{"issue":"11","key":"R15","doi-asserted-by":"publisher","first-page":"1539","DOI":"10.1002\/sim.1186","article-title":"Quantifying heterogeneity in a meta-analysis","volume":"21","author":"Higgins","journal-title":"Stat Med"},{"issue":"9","key":"R16","doi-asserted-by":"publisher","first-page":"882","DOI":"10.1016\/j.jclinepi.2005.01.016","article-title":"The performance of tests of publication bias and other sample size effects in systematic reviews of diagnostic test accuracy was assessed","volume":"58","author":"Deeks","journal-title":"J Clin Epidemiol"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2025.112356","article-title":"Dynamic AI-assisted ipsilateral tissue matching for digital breast tomosynthesis","volume":"192","author":"Morrell","journal-title":"Eur J Radiol"},{"issue":"3","key":"R18","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1093\/jbi\/wbad015","article-title":"Use of artificial intelligence for digital breast tomosynthesis screening: a preliminary real-world experience","volume":"5","author":"Letter","journal-title":"J Breast Imaging"},{"issue":"1","key":"R19","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1007\/s12282-022-01396-4","article-title":"Artificial intelligence computer-aided detection enhances synthesized mammograms: comparison with original digital mammograms alone and in combination with tomosynthesis images in an experimental setting","volume":"30","author":"Uematsu","journal-title":"Breast Cancer"},{"issue":"6","key":"R20","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2021200299","article-title":"Artificial intelligence detection of missed cancers at digital mammography that were detected at digital breast tomosynthesis","volume":"3","author":"Dahlblom","journal-title":"Radiol Artif Intell"},{"issue":"1","key":"R21","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1007\/s11604-022-01327-5","article-title":"Comparisons between artificial intelligence computer-aided detection synthesized mammograms and digital mammograms when used alone and in combination with tomosynthesis images in a virtual screening setting","volume":"41","author":"Uematsu","journal-title":"Jpn J Radiol"},{"issue":"3","key":"R22","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1148\/radiol.212381","article-title":"Artificial intelligence evaluation of 122 969 mammography examinations from a population-based screening program","volume":"303","author":"Larsen","journal-title":"Radiology"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1093\/jbi\/wbae058","article-title":"External validation of a commercial artificial intelligence algorithm on a diverse population for detection of false negative breast cancers","volume":"7","author":"Plimpton","journal-title":"J Breast Imaging"},{"issue":"Suppl 2","key":"R24","doi-asserted-by":"publisher","DOI":"10.1117\/1.JMI.12.S2.S22005","article-title":"Workload reduction of digital breast tomosynthesis screening using artificial intelligence and synthetic mammography: a simulation study","volume":"12","author":"Dahlblom","journal-title":"J Med Imaging (Bellingham)"},{"key":"R25","doi-asserted-by":"publisher","DOI":"10.1016\/j.csbj.2025.06.008","article-title":"An explainable-by-design end-to-end AI framework based on prototypical part learning for lesion detection and classification in Digital Breast Tomosynthesis images","volume":"27","author":"Berti","journal-title":"Comput Struct Biotechnol J"},{"issue":"11","key":"R26","doi-asserted-by":"publisher","first-page":"8682","DOI":"10.1007\/s00330-021-07992-w","article-title":"Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation: a multi-reader multi-case study","volume":"31","author":"van Winkel","journal-title":"Eur Radiol"},{"issue":"4","key":"R27","doi-asserted-by":"publisher","DOI":"10.1148\/rycan.230149","article-title":"AI-enhanced mammography with digital breast tomosynthesis for breast cancer detection: clinical value and comparison with human performance","volume":"6","author":"Resch","journal-title":"Radiol Imaging Cancer"},{"issue":"1","key":"R28","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1148\/radiol.2021203555","article-title":"AI-based strategies to reduce workload in breast cancer screening with mammography and tomosynthesis: a retrospective evaluation","volume":"300","author":"Raya-Povedano","journal-title":"Radiology"},{"issue":"3","key":"R29","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1148\/radiol.2021204432","article-title":"Impact of artificial intelligence decision support using deep learning on breast cancer screening interpretation with single-view wide-angle digital breast tomosynthesis","volume":"300","author":"Pinto","journal-title":"Radiology"},{"issue":"3","key":"R30","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.230318","article-title":"Impact of AI for digital breast tomosynthesis on breast cancer detection and interpretation time","volume":"6","author":"Park","journal-title":"Radiol Artif Intell"},{"issue":"2","key":"R31","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.230137","article-title":"Impact of a categorical AI system for digital breast tomosynthesis on breast cancer interpretation by both general radiologists and breast imaging specialists","volume":"6","author":"Kim","journal-title":"Radiol Artif Intell"},{"issue":"6","key":"R32","doi-asserted-by":"publisher","first-page":"3958","DOI":"10.1007\/s00330-023-10426-4","article-title":"Impact of real-life use of artificial intelligence as support for human reading in a population-based breast cancer screening program with mammography and tomosynthesis","volume":"34","author":"El\u00edas-Cabot","journal-title":"Eur Radiol"},{"issue":"4","key":"R33","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2019180096","article-title":"Improving accuracy and efficiency with concurrent use of artificial intelligence for digital breast tomosynthesis","volume":"1","author":"Conant","journal-title":"Radiol Artif Intell"},{"issue":"6","key":"R34","doi-asserted-by":"publisher","DOI":"10.2214\/AJR.25.32889","article-title":"Commercial artificial intelligence versus radiologists: NPV and recall rate in large population-based digital mammography and tomosynthesis screening mammography cohorts","volume":"225","author":"Chen","journal-title":"AJR Am J Roentgenol"},{"issue":"5","key":"R35","doi-asserted-by":"publisher","first-page":"2518","DOI":"10.1007\/s00330-018-5886-0","article-title":"Decrease in interpretation time for both novice and experienced readers using a concurrent computer-aided detection system for digital breast tomosynthesis","volume":"29","author":"Chae","journal-title":"Eur Radiol"},{"issue":"5","key":"R36","doi-asserted-by":"publisher","first-page":"584","DOI":"10.1053\/j.semnuclmed.2022.02.003","article-title":"Application of deep learning in breast cancer imaging","volume":"52","author":"Balkenende","journal-title":"Semin Nucl Med"},{"issue":"8","key":"R37","doi-asserted-by":"publisher","first-page":"4474","DOI":"10.1007\/s00330-025-11406-6","article-title":"Deep learning-based breast cancer diagnosis in breast MRI: systematic review and meta-analysis","volume":"35","author":"Abdullah","journal-title":"Eur Radiol"},{"issue":"664","key":"R38","doi-asserted-by":"publisher","DOI":"10.1126\/scitranslmed.abo4802","article-title":"Improving breast cancer diagnostics with deep learning for MRI","volume":"14","author":"Witowski","journal-title":"Sci Transl Med"},{"issue":"1","key":"R39","doi-asserted-by":"publisher","DOI":"10.1186\/s13244-022-01259-8","article-title":"Deep learning based on ultrasound images assists breast lesion diagnosis in China: a multicenter diagnostic study","volume":"13","author":"Gu","journal-title":"Insights Imaging"},{"issue":"2","key":"R40","doi-asserted-by":"publisher","first-page":"244","DOI":"10.1038\/s41591-020-01174-9","article-title":"Robust breast cancer detection in mammography and digital breast tomosynthesis using an annotation-efficient deep learning approach","volume":"27","author":"Lotter","journal-title":"Nat Med"},{"issue":"1","key":"R41","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-022-00559-z","article-title":"Deep learning in image-based breast and cervical cancer detection: a systematic review and meta-analysis","volume":"5","author":"Xue","journal-title":"NPJ Digit Med"},{"issue":"1","key":"R42","doi-asserted-by":"publisher","DOI":"10.1186\/s13244-025-01983-x","article-title":"Mammography-based artificial intelligence for breast cancer detection, diagnosis, and BI-RADS categorization using multi-view and multi-level convolutional neural networks","volume":"16","author":"Tan","journal-title":"Insights Imaging"},{"issue":"13","key":"R43","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics15131631","article-title":"Deep learning with transfer learning on digital breast tomosynthesis: a radiomics-based model for predicting breast cancer risk","volume":"15","author":"Galati","journal-title":"Diagnostics (Basel)"},{"key":"R44","doi-asserted-by":"publisher","DOI":"10.1038\/s41698-024-00514-z","article-title":"Diagnostic performance of deep learning in ultrasound diagnosis of breast cancer: a systematic review","volume":"8","author":"Dan","journal-title":"NPJ Precis Oncol"},{"issue":"5","key":"R45","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.222639","article-title":"Standalone AI for breast cancer detection at screening digital mammography and digital breast tomosynthesis: a systematic review and meta-analysis","volume":"307","author":"Yoon","journal-title":"Radiology"},{"issue":"8056","key":"R46","doi-asserted-by":"publisher","first-page":"888","DOI":"10.1038\/s41586-025-08675-y","article-title":"Multimodal generative AI for medical image interpretation","volume":"639","author":"Rao","journal-title":"Nature"},{"issue":"3","key":"R47","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.240982","article-title":"Navigating the AI landscape in medical imaging: a critical analysis of technologies, implementation, and implications","volume":"315","author":"Sosna","journal-title":"Radiology"},{"key":"R48","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2023.111159","article-title":"A scoping review of interpretability and explainability concerning artificial intelligence methods in medical imaging","volume":"169","author":"Champendal","journal-title":"Eur J Radiol"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jmir.org\/2026\/1\/e91659","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:15:49Z","timestamp":1782839749000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2026\/1\/e91659"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,6]]},"references-count":48,"URL":"https:\/\/doi.org\/10.2196\/91659","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,6]]},"article-number":"v28i6e91659"}}