Past, Present, and Future of Machine Learning and Artificial Intelligence for Breast Cancer Screening

被引:7
作者
Baughan, Natalie [1 ]
Douglas, Lindsay [1 ]
Giger, Maryellen L. [1 ]
机构
[1] Univ Chicago, Dept Radiol Comm Med Phys, Chicago, IL 60637 USA
关键词
artificial intelligence; computer-aided diagnosis; breast imaging; machine learning; screening; COMPUTER-AIDED DETECTION; ENHANCED MR-IMAGES; CLUSTERED MICROCALCIFICATIONS; AUTOMATED DETECTION; DECISION-SUPPORT; DETECTION SYSTEM; LESIONS; CLASSIFICATION; DIAGNOSIS; MAMMOGRAMS;
D O I
10.1093/jbi/wbac052
中图分类号
R73 [肿瘤学];
学科分类号
100214 ;
摘要
Breast cancer screening has evolved substantially over the past few decades because of advancements in new image acquisition systems and novel artificial intelligence (AI) algorithms. This review provides a brief overview of the history, current state, and future of AI in breast cancer screening and diagnosis along with challenges involved in the development of AI systems. Although AI has been developing for interpretation tasks associated with breast cancer screening for decades, its potential to combat the subjective nature and improve the efficiency of human image interpretation is always expanding. The rapid advancement of computational power and deep learning has increased greatly in AI research, with promising performance in detection and classification tasks across imaging modalities. Most AI systems, based on human-engineered or deep learning methods, serve as concurrent or secondary readers, that is, as aids to radiologists for a specific, well-defined task. In the future, AI may be able to perform multiple integrated tasks, making decisions at the level of or surpassing the ability of humans. Artificial intelligence may also serve as a partial primary reader to streamline ancillary tasks, triaging cases or ruling out obvious normal cases. However, before AI is used as an independent, autonomous reader, various challenges need to be addressed, including explainability and interpretability, in addition to repeatability and generalizability, to ensure that AI will provide a significant clinical benefit to breast cancer screening across all populations.
引用
收藏
页码:451 / 459
页数:9
相关论文
共 78 条
[1]  
ACR Data Science Institute AI Central, FDA CLEAR AI MED PRO
[2]   Case-based Diagnostic Classification Repeatability using Radiomic Features Extracted from Full-Field Digital Mammography Images of Breast Lesions [J].
Amstutz, Paul ;
Drukker, Karen ;
Li, Hui ;
Abe, Hiroyuki ;
Giger, Maryellen L. ;
Whitney, Heather M. .
MEDICAL IMAGING 2021: COMPUTER-AIDED DIAGNOSIS, 2021, 11597
[3]  
[Anonymous], 2014, WHO position paper on mammography screening
[4]  
[Anonymous], CODE FEDERAL REGULAT
[5]   Use of clinical MRI maximum intensity projections for improved breast lesion classification with deep convolutional neural networks [J].
Antropova, Natalia ;
Abe, Hiroyuki ;
Giger, Maryellen L. .
JOURNAL OF MEDICAL IMAGING, 2018, 5 (01)
[6]   A deep feature fusion methodology for breast cancer diagnosis demonstrated on three imaging modality datasets [J].
Antropova, Natalia ;
Huynh, Benjamin Q. ;
Giger, Maryellen L. .
MEDICAL PHYSICS, 2017, 44 (10) :5162-5171
[7]   Automatic breast lesion detection in ultrafast DCE-MRI using deep learning [J].
Ayatollahi, Fazael ;
Shokouhi, Shahriar B. ;
Mann, Ritse M. ;
Teuwen, Jonas .
MEDICAL PHYSICS, 2021, 48 (10) :5897-5907
[8]   Cancerous Breast Lesions on Dynamic Contrast-enhanced MR Images: Computerized Characterization for Image-based Prognostic Markers [J].
Bhooshan, Neha ;
Giger, Maryellen L. ;
Jansen, Sanaz A. ;
Li, Hui ;
Lan, Li ;
Newstead, Gillian M. .
RADIOLOGY, 2010, 254 (03) :680-690
[9]  
Bi WL., 2019, CA-CANCER J CLIN, V1, P31
[10]   Computer-aided detection system for breast masses on digital tomosynthesis mammograms: Preliminary experience [J].
Chan, HP ;
Wei, J ;
Sahiner, B ;
Rafferty, EA ;
Wu, T ;
Roubidoux, MA ;
Moore, RH ;
Kopans, DB ;
Hadjiiski, LM ;
Helvie, MA .
RADIOLOGY, 2005, 237 (03) :1075-1080