Publicado
Artificial intelligence algorithms versus conventional radiological interpretation in breast cancer screening and classification: Systematic review and meta-analysis
Algoritmos de inteligencia artificial versus interpretación radiológica convencional en tamizaje y clasificación del cáncer de mama: revisión sistemática con metaanálisis
DOI:
https://doi.org/10.15446/revfacmed.v73.114297Palabras clave:
Breast Neoplasms, Primary Health Care , Diagnosis, Computer-Assisted, Screening, Systematic Review, Meta-analysis (en)Neoplasias de la Mama, Inteligencia Artificial, Diagnóstico por Computador, Tamizaje, Revisión Sistemática, Metaanálisis (es)
Descargas
Introduction: Population-based breast cancer screening programs relying on direct interpretation of mammograms and ultrasounds by radiologists have limitations such as high workload, interobserver variability, false positives, and a shortage of specialists. Artificial intelligence (AI), through deep neural networks and other algorithms, could optimize early detection and characterization of breast lesions.
Objective: To evaluate the efficacy, in terms of operational characteristics (sensitivity, specificity, and area under the curve [AUC]), of AI algorithms in the interpretation of imaging studies (mammography, ultrasound, and magnetic resonance imaging) for (1) the detection of breast cancer in screening programs, (2) the classification of breast lesions (benign vs. malignant and BI-RADS categories), and (3) the stratification of molecular subtypes of breast cancer (e.g., triple-negative) compared to conventional interpretation by a radiologist (i.e., without AI assistance).
Materials and methods: Systematic review with meta-analysis of observational studies and clinical trials published up to May 29, 2024, in Medline, Embase, and LILACS. PROSPERO registration code: CRD42024507843. Studies comparing the sensitivity, specificity, and AUC of AI algorithms versus conventional radiological interpretation methods were included. Given the heterogeneity between studies, the meta-analysis included only the sensitivity and specificity of AI algorithms for detecting breast cancer (Dersimonian-Lard random effects model).
Results: Out of 1 156 records identified, 32 studies were included in the review (32 for qualitative analysis and 26 for meta-analysis). The AUC, sensitivity, and specificity ranges of the AI algorithms for detecting breast cancer, classifying breast lesions, and discriminating molecular subtypes of breast cancer were 0.706-0.98, 63.7-96.89%, and 45-97.9%, 0.838-0.978, 76-100%, and 78. 71-100%, and 0.535-0.978, 50-96.6%, and 52.4-95.56%, respectively. As for the meta-analysis, the pooled sensitivity and specificity of AI for breast cancer detection were 87% and 89%, respectively.
Conclusion: AI demonstrated a high diagnostic performance in breast cancer screening and classification, surpassing human performance in several studies. Therefore, AI may serve as a complementary tool to enhance the efficiency and quality of screening programs. However, prospective studies are needed to assess its clinical implementation, standardize protocols, and determine its impact on long-term health outcomes.
Introducción. Los programas de tamizaje poblacional de cáncer de mama basados en la interpretación directa de mamografías y ecografías por radiólogos presentan limitaciones como carga de trabajo elevada, variabilidad interobservador, falsas alarmas y escasez de especialistas. La inteligencia artificial (IA), mediante redes neuronales profundas y otros algoritmos, podría optimizar la detección precoz y la caracterización de lesiones mamarias.
Objetivo. Evaluar la eficacia, en términos de características operativas (sensibilidad, especificidad y área bajo la curva [AUC]), del uso de algoritmos de IA en la interpretación de estudios de imagen (mamografía, ultrasonido y resonancia magnética) para 1) la detección de cáncer de mama en programas de cribado, 2) la clasificación de lesiones mamarias (benignas vs. malignas y clasificación BI-RADS) y 3) la diferenciación de subtipos moleculares de cáncer de mama (p. ej., triple negativo) en comparación con la interpretación convencional por radiólogos (i.e., sin asistencia de IA).
Materiales y métodos. Revisión sistemática con metaanálisis de estudios observacionales y ensayos clínicos publicados hasta el 29 de mayo de 2024 en Medline, Embase y LILACS. Código de registro en PROSPERO: CRD42024507843. Se incluyeron trabajos que compararan sensibilidad, especificidad y AUC de algoritmos de IA versus métodos radiológicos convencionales. Dada la heterogeneidad entre estudios, el metaanálisis incluyó únicamente la sensibilidad y especificidad de la IA para detectar cáncer de mama (modelo de efectos aleatorios de Dersimonian-Lard).
Resultados. De 1 156 registros identificados, 32 fueron incluidos en la revisión (32 para el análisis cualitativo y 26 para el metaanálisis). Los rangos de AUC, sensibilidad y especificidad de la IA para detectar cáncer de mama, clasificar lesiones mamarias y discriminar subtipos moleculares de cáncer fueron 0.706-0.98, 63.7-96.89% y 45-97.9%, 0.838-0.978, 76-100% y 78.71-100%, y 0.535-0.978, 50-96.6% y 52.4-95.56%, respectivamente. En cuanto al metaanálisis, la sensibilidad y especificidad combinadas de la IA para la detección de cáncer de mama fueron 87% y 89%, respectivamente.
Conclusión. La IA demostró un elevado desempeño diagnóstico en el tamizaje y la clasificación de lesiones mamarias, superponiéndose en varios estudios al rendimiento humano. Por lo tanto, la IA tiene el potencial de convertirse en una herramienta complementaria para mejorar la eficiencia y la calidad de los programas de cribado, aunque se requieren estudios prospectivos que evalúen su implementación clínica, estandaricen protocolos y determinen su impacto en resultados de salud a largo plazo.
Referencias
Ferlay J, Ervik M, Lam F, Laversanne M, Colombet M, Mery L, et al. Global Cancer Observatory: Cancer Today [Internet]. Lyon: International Agency for Research on Cancer; 2022 [cited 2025 Jul 2]. Available from: https://gco.iarc.who.int/media/globocan/factsheets/populations/900-world-fact-sheet.pdf.
Harbeck N, Gnant M. Breast cancer. The Lancet. 2017;389(10074):1134-50. doi: 10.1016/S0140-6736(16)31891-8. PMID: 27865536.
Chen Y, Lu J, Li J, Liao J, Huang X, Zhang B. Evaluation of diagnostic efficacy of multimode ultrasound in BI-RADS 4 breast neoplasms and establishment of a predictive model. Front Oncol. 2022;12:1053280. doi: 10.3389/fonc.2022.1053280. PMID: 36505867; PMCID: PMC9730703.
Akselrod-Ballin A, Chorev M, Shoshan Y, Spiro A, Hazan A, Melamed R, et al. Predicting Breast Cancer by Applying Deep Learning to Linked Health Records and Mammograms. Radiology. 2019;292(2):331-42. doi: 10.1148/radiol.2019182622. PMID: 31210611.
World Health Organization (WHO). Breast cancer [Internet]. Geneva: WHO; 2024 [cited 2024 Sep 3]. Available from: https://www.who.int/news-room/fact-sheets/detail/breast-cancer.
Manson EN, Achel DG. Fighting breast cancer in low-and-middle-income countries - What must we do to get every woman screened on regular basis? Scientific Africa. 2023;21:e01848. doi: 10.1016/j.sciaf.2023.e01848.
McKinney SM, Sieniek M, Godbole V, Godwin J, Antropova N, Ashrafian H, et al. International evaluation of an AI system for breast cancer screening. Nature. 2020;577(7788):89-94. doi: 10.1038/s41586-019-1799-6. PMID: 31894144.
Rodriguez-Ruiz A, Lång K, Gubern-Merida A, Broeders M, Gennaro G, Clauser P, et al. Stand-Alone Artificial Intelligence for Breast Cancer Detection in Mammography: Comparison With 101 Radiologists. J Natl Cancer Inst. 2019;111(9):916-22. doi: 10.1093/jnci/djy222. PMID: 30834436; PMCID: PMC6748773.
Verburg E, van Gils CH, Van Der Velden BHM, Bakker MF, Pijnappel RM, Veldhuis WB, et al. Validation of Combined Deep Learning Triaging and Computer-Aided Diagnosis in 2901 Breast MRI Examinations from the Second Screening Round of the Dense Tissue and Early Breast Neoplasm Screening Trial. Invest Radiol. 2023;58(4):293-8. doi: 10.1097/RLI.0000000000000934. PMID: 36256783; PMCID: PMC9997620.
Lehman CD, Wellman RD, Buist DSM, Kerlikowske K, Tosteson AN, Miglioretti DL, et al. Diagnostic Accuracy of Digital Screening Mammography With and Without Computer-Aided Detection. JAMA Intern Med. 2015;175(11):1828-37. doi: 10.1001/jamainternmed.2015.5231. PMID: 26414882; PMCID: PMC4836172.
Watanabe AT, Lim V, Vu HX, Chim R, Weise E, Liu J, et al. Improved Cancer Detection Using Artificial Intelligence: a Retrospective Evaluation of Missed Cancers on Mammography. J Digit Imaging. 2019;32(4):625-37. doi: 10.1007/s10278-019-00192-5. PMID: 31011956; PMCID: PMC6646649.
Schaffter T, Buist DSM, Lee CI, Nikulin Y, Ribli D, Guan Y, et al. Evaluation of Combined Artificial Intelligence and Radiologist Assessment to Interpret Screening Mammograms. JAMA Netw Open. 2020;3(3):e200265. doi: 10.1001/jamanetworkopen.2020.0265. Erratum in: JAMA Netw Open. 2020;3(3):e204429. doi: 10.1001/jamanetworkopen.2020.4429. PMID: 32119094; PMCID: PMC7052735.
Hsu W, Hippe DS, Nakhaei N, Wang PC, Zhu B, Siu N, et al. External Validation of an Ensemble Model for Automated Mammography Interpretation by Artificial Intelligence. JAMA Netw Open.2022;5(11):e2242343. doi: 10.1001/jamanetworkopen.2022.42343. PMID: 36409497; PMCID: PMC9679879.
Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ. 2009;339:b2700. doi: 10.1136/bmj.b2700. PMID: 19622552; PMCID: PMC2714672.
Centro Cochrane Iberoamericano, traductores. Manual Cochrane de Revisiones Sistemáticas de Intervenciones, versión 5.1.0 [Internet]. Barcelona: Centro Cochrane Iberoamericano; 2012 [cited 2024 Apr 29]. Available from: https://training.cochrane.org/es/manual-cochrane-de-revisiones-sistem%C3%A1ticas-de-intervenciones.
Hillen MA, Medendorp NM, Daams JG, Smets EMA. Patient-Driven Second Opinions in Oncology: A Systematic Review. Oncologist. 2017;22(10):1197-211. doi: 10.1634/theoncologist.2016-0429. PMID: 28606972; PMCID: PMC5634767.
Blanchard L, Ray S, Law C, Vega-Salas MJ, Bidonde J, Bridge G, et al. The effectiveness, cost-effectiveness and policy processes of regulatory, voluntary and partnership policies to improve food environments: an evidence synthesis. Southampton (UK): National Institute for Health and Care Research; 2024 Sep. (Public Health Research, No. 12.08.) Available from: https://www.ncbi.nlm.nih.gov/books/NBK607533/. doi: 10.3310/JYWP4049
Ma M, Liu R, Wen C, Xu W, Xu Z, Wang S, et al. Predicting the molecular subtype of breast cancer and identifying interpretable imaging features using machine learning algorithms. Eur Radiol. 2022;32(3):1652-62. doi: 10.1007/s00330-021-08271-4. PMID: 34647174.
Sun Y, Qu Y, Wang D, Li Y, Ye L, Du J, et al. Deep learning model improves radiologists’ performance in detection and classification of breast lesions. Chin J Cancer Res. 2021;33(6):682-93. doi: 10.21147/j.issn.1000-9604.2021.06.05. PMID: 35125812; PMCID: PMC8742176.
Gu Y, Xu W, Liu T, An X, Tian J, Ran H, et al. Ultrasound-based deep learning in the establishment of a breast lesion risk stratification system: a multicenter study. Eur Radiol. 2023;33(4):2954-64. doi: 10.1007/s00330-022-09263-8. PMID: 36418619.
Schönenberger C, Hejduk P, Ciritsis A, Marcon M, Rossi C, Boss A. Classification of Mammographic Breast Microcalcifications Using a Deep Convolutional Neural Network: A BI-RADS-Based Approach. Invest Radiol. 2021;56(4):224-31. doi: 10.1097/RLI.0000000000000729. PMID: 33038095.
Leibig C, Brehmer M, Bunk S, Byng D, Pinker K, Umutlu L. Combining the strengths of radiologists and AI for breast cancer screening: a retrospective analysis. Lancet Digit Health. 2022;4(7):e507-19. doi: 10.1016/S2589-7500(22)00070-X. PMID: 35750400; PMCID: PMC9839981.
Lee SE, Son NH, Kim MH, Kim EK. Mammographic Density Assessment by Artificial Intelligence-Based Computer-Assisted Diagnosis: A Comparison with Automated Volumetric Assessment. J Digit Imaging. 2022;35(2):173-9. doi: 10.1007/s10278-021-00555-x. PMID: 35015180; PMCID: PMC8921363.
Sasaki M, Tozaki M, Rodríguez-Ruiz A, Yotsumoto D, Ichiki Y, Terawaki A, et al. Artificial intelligence for breast cancer detection in mammography: experience of use of the ScreenPoint Medical Transpara system in 310 Japanese women. Breast Cancer. 2020;27(4):642-51. doi: 10.1007/s12282-020-01061-8. PMID: 32052311.
Elhakim MT, Stougaard SW, Graumann O, Nielsen M, Lång K, Gerke O, et al. Breast cancer detection accuracy of AI in an entire screening population: a retrospective, multicentre study. Cancer Imaging. 2023;23(1):127. doi: 10.1186/s40644-023-00643-x. PMID: 38124111; PMCID: PMC10731688.
Marinovich ML, Wylie E, Lotter W, Lund H, Waddell A, Madeley C, et al. Artificial intelligence (AI) for breast cancer screening: BreastScreen population-based cohort study of cancer detection. EBioMedicine. 2023;90:104498. doi: 10.1016/j.ebiom.2023.104498. PMID: 36863255; PMCID: PMC9996220.
Raafat M, Mansour S, Kamel R, Ali HW, Shibel PE, Marey A, et al. Does artificial intelligence aid in the detection of different types of breast cancer? Egypt J Radiol Nucl Med. 2022;53(1):182. doi: 10.1186/s43055-022-00868-z.
Wang Q, Chen H, Luo G, Li B, Shang H, Shao H, et al. Performance of novel deep learning network with the incorporation of the automatic segmentation network for diagnosis of breast cancer in automated breast ultrasound. Eur Radiol. 2022;32(10):7163-72. doi: 10.1007/s00330-022-08836-x. PMID: 35488916.
Jiang Y, Edwards AV, Newstead GM. Artificial intelligence applied to breast MRI for improved diagnosis. Radiology. 2021;298(1):38-46. doi: 10.1148/radiol.2020200292. PMID: 33078996.
Shia WC, Lin LS, Chen DR. Classification of malignant tumours in breast ultrasound using unsupervised machine learning approaches. Sci Rep. 2021;11(1):1418. doi: 10.1038/s41598-021-81008-x. PMID: 33446841; PMCID: PMC7809485.
Shia WC, Chen DR. Classification of malignant tumors in breast ultrasound using a pretrained deep residual network model and support vector machine. Comput Med Imaging Graph. 2021;87:101829. doi: 10.1016/j.compmedimag.2020.101829. PMID: 33302247.
Trang NTH, Long KQ, An PL, Dang TN. Development of an Artificial Intelligence-Based Breast Cancer Detection Model by Combining Mammograms and Medical Health Records. Diagn (Basel). 2023;13(3):346. doi: 10.3390/diagnostics13030346. PMID: 36766450; PMCID: PMC9913958.
Yoon JH, Han K, Suh HJ, Youk JH, Lee SE, Kim EK. Artificial intelligence-based computer-assisted detection/diagnosis (AI-CAD) for screening mammography: Outcomes of AI-CAD in the mammographic interpretation workflow. Eur J Radiol Open. 2023;11:100509. doi: 10.1016/j.ejro.2023.100509. PMID: 37484980; PMCID: PMC10362167.
Gu Y, Xu W, Lin B, An X, Tian J, Ran H, et al. Deep learning based on ultrasound images assists breast lesion diagnosis in China: a multicenter diagnostic study. Insights Imaging. 2022;13(1):124. doi: 10.1186/s13244-022-01259-8. PMID: 35900608; PMCID: PMC9334487.
Koch HW, Larsen M, Bartsch H, Kurz KD, Hofvind S. Artificial intelligence in BreastScreen Norway: a retrospective analysis of a cancer-enriched sample including 1254 breast cancer cases. Eur Radiol. 2023;33(5):3735-43. doi: 10.1007/s00330-023-09461-y. PMID: 36917260; PMCID: PMC10121532.
O’Connell AM, Bartolotta TV, Orlando A, Jung SH, Baek J, Parker KJ. Diagnostic Performance of an Artificial Intelligence System in Breast Ultrasound. J Ultrasound Med. 2022;41(1):97-105. doi: 10.1002/jum.15684. PMID: 33665833.
Park HJ, Kim SM, La Yun B, Jang M, Kim B, Jang JY, et al. A computer-aided diagnosis system using artificial intelligence for the diagnosis and characterization of breast masses on ultrasound: Added value for the inexperienced breast radiologist. Medicine (Baltimore). 2019;98(3):e14146. doi: 10.1097/MD.0000000000014146. PMID: 30653149; PMCID: PMC6370030.
Busaleh M, Hussain M, Aboalsamh HA, Amin FE. Breast Mass Classification Using Diverse Contextual Information and Convolutional Neural Network. Biosensors (Basel). 2021;11(11):419. doi: 10.3390/bios11110419. PMID: 34821634; PMCID: PMC8615673.
Cai X, Li X, Razmjooy N, Ghadimi N. Breast Cancer Diagnosis by Convolutional Neural Network and Advanced Thermal Exchange Optimization Algorithm. Comput Math Methods Med. 2021;2021:5595180. doi: 10.1155/2021/5595180. PMID: 34790252; PMCID: PMC8592754.
Shen L, Margolies LR, Rothstein JH, Fluder E, McBride R, Sieh W. Deep Learning to Improve Breast Cancer Detection on Screening Mammography. Sci Rep. 2019;9(1):12495. doi: 10.1038/s41598-019-48995-4. PMID: 31467326; PMCID: PMC6715802.
Suh YJ, Jung J, Cho BJ. Automated Breast Cancer Detection in Digital Mammograms of Various Densities via Deep Learning. J Pers Med. 2020;10(4). doi: 10.3390/jpm10040211. PMID: 33172076; PMCID: PMC7711783.
Guo YY, Huang YH, Wang Y, Huang J, Lai QQ, Li YZ. Breast MRI Tumor Automatic Segmentation and Triple-Negative Breast Cancer Discrimination Algorithm Based on Deep Learning. Comput Math Methods Med. 2022;2022:2541358. doi: 10.1155/2022/2541358. PMID: 36092784; PMCID: PMC9453096.
Rodríguez-Ruiz A, Krupinski E, Mordang JJ, Schilling K, Heywang-Köbrunner SH, Sechopoulos I, et al. Detection of Breast Cancer with Mammography: Effect of an Artificial Intelligence Support System. Radiology. 2019;290(2):305-14. doi: 10.1148/radiol.2018181371. PMID: 30457482.
Pang T, Wong JHD, Ng WL, Chan CS. Semi-supervised GAN-based Radiomics Model for Data Augmentation in Breast Ultrasound Mass Classification. Comput Methods Programs Biomed. 2021;203:106018. doi: 10.1016/j.cmpb.2021.106018. PMID: 33714900.
Ye H, Hang J, Zhang M, Chen X, Ye X, Chen J, et al. Automatic identification of triple negative breast cancer in ultrasonography using a deep convolutional neural network. Sci Rep. 2021;11(1):20474. doi: 10.1038/s41598-021-00018-x. PMID: 34650065; PMCID: PMC8517009.
Ferre R, Elst J, Senthilnathan S, Lagree A, Tabbarah S, Lu FI, et al. Machine learning analysis of breast ultrasound to classify triple negative and HER2+ breast cancer subtypes. Breast Dis. 2023;42(1):59-66. doi: 10.3233/BD-220018. PMID: 36911927.
Feng H, Cao J, Wang H, Xie Y, Yang D, Feng J, et al. A knowledge-driven feature learning and integration method for breast cancer diagnosis on multi-sequence MRI. Magn Reson Imaging. 2020;69:40-8. doi: 10.1016/j.mri.2020.03.001. PMID: 32173583.
Ciritsis A, Rossi C, Eberhard M, Marcon M, Becker AS, Boss A. Automatic classification of ultrasound breast lesions using a deep convolutional neural network mimicking human decision-making. Eur Radiol. 2019;29(10):5458-68. doi: 10.1007/s00330-019-06118-7. PMID: 30927100.
Li C, Zhang H, Chen J, Shao S, Li X, Yao M, et al. Deep learning radiomics of ultrasonography for differentiating sclerosing adenosis from breast cancer. Clin Hemorheol Microcirc. 2023;84(2):153-63. doi: 10.3233/CH-221608. PMID: 36373313.
Hickman SE, Woitek R, Le EPV, Im YR, Mouritsen Luxhøj C, et al. Machine Learning for Workflow Applications in Screening Mammography: Systematic Review and Meta-Analysis. Radiology. 2022;302(1):88-104. doi: 10.1148/radiol.2021210391. PMID: 34665034; PMCID: PMC8717814.
Liu J, Lei J, Ou Y, Zhao Y, Tuo X, Zhang B, et al. Mammography diagnosis of breast cancer screening through machine learning: a systematic review and meta-analysis. Clin Exp Med. 2023;23(6):2341-2356. doi: 10.1007/s10238-022-00895-0. PMID: 36242643.
Yoon JH, Strand F, Baltzer PAT, Conant EF, Gilbert FJ, Lehman CD, et al. Standalone AI for Breast Cancer Detection at Screening Digital Mammography and Digital Breast Tomosynthesis: A Systematic Review and Meta-Analysis. Radiology. 2023;307(5):e222639. doi: 10.1148/radiol.222639. PMID: 37219445; PMCID: PMC10315526.
Kim HE, Kim HH, Han BK, Kim KH, Han K, Nam H, et al. Changes in cancer detection and false-positive recall in mammography using artificial intelligence: a retrospective, multireader study. Lancet Digit Health. 2020;2(3):e138-48. doi: 10.1016/S2589-7500(20)30003-0. PMID: 33334578.
Aparecida-Roela R, Vansuita-Valente G, Shimizu C, Mendoza-Lopez RV, de Melo-Tucunduva TC, Koike-Folgueira G, et al. Deep learning algorithm performance in mammography screening: A systematic review and meta-analysis. J Clin Oncol. 2021;39(Suppl 15):e13553. doi: 10.1200/JCO.2021.39.15_suppl.e1355.
Dimensions
PlumX
Visitas a la página del resumen del artículo
Descargas
Cómo citar
Licencia
Derechos de autor 2025 El(Los) autor(es).
Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.
-
