Computer-Aided Detection for Breast Cancer Screening in Clinical Settings: Scoping Review

Abstract

Background With the growth of machine learning applications, the practice of medicine is evolving. Computer-aided detection (CAD) is a software technology that has become widespread in radiology practices, particularly in breast cancer screening for improving detection rates at earlier stages. Many studies have investigated the diagnostic accuracy of CAD, but its implementation in clinical settings has been largely overlooked. Objective The aim of this scoping review was to summarize recent literature on the adoption and implementation of CAD during breast cancer screening by radiologists and to describe barriers and facilitators for CAD use. Methods The MEDLINE database was searched for English, peer-reviewed articles that described CAD implementation, including barriers or facilitators, in breast cancer screening and were published between January 2010 and March 2018. Articles describing the diagnostic accuracy of CAD for breast cancer detection were excluded. The search returned 526 citations, which were reviewed in duplicate through abstract and full-text screening. Reference lists and cited references in the included studies were reviewed. Results A total of nine articles met the inclusion criteria. The included articles showed that there is a tradeoff between the facilitators and barriers for CAD use. Facilitators for CAD use were improved breast cancer detection rates, increased profitability of breast imaging, and time saved by replacing double reading. Identified barriers were less favorable perceptions of CAD compared to double reading by radiologists, an increase in recall rates of patients for further testing, increased costs, and unclear effect on patient outcomes. Conclusions There is a gap in the literature between CAD’s well-established diagnostic accuracy and its implementation and use by radiologists. Generally, the perceptions of radiologists have not been considered and details of implementation approaches for adoption of CAD have not been reported. The cost-effectiveness of CAD has not been well established for breast cancer screening in various populations. Further research is needed on how to best facilitate CAD in radiology practices in order to optimize patient outcomes, and the views of radiologists need to be better considered when advancing CAD use.

What does it mean to provide decision support to a responsible and competent expert?

“…According to their results, the radiologists were more favorable to double reading, even though most of them perceived that DDSSs were better at improving recall rates than double reading. The meta-analysis proposed by Masud et al (2019), on the use of DDSSs in radiology departments, showed similar results on the low acceptability of DDSSs despite an improvement of performances perceived by radiologists.…”

“…To sum-up, although there are exceptions in specific situations, it appears that DDSSs are generally poorly accepted in customary situations, where support appears to be redundant with physicians' capabilities (Masud et al, 2019). It hence appears that we are currently in a paradoxical situation.…”

Automated Detection of Vestibular Schwannoma Growth Using a Two‐Dimensional U‐Net Convolutional Neural Network

“…We believe a tool like our automated segmentation method could play a role in this situation. For example, in computer‐aided detection in breast mammography, the radiologist combines their interpretation of a study with an algorithm's interpretation to develop a final read 21 . In a similar and simpler fashion, the radiologist measuring the greatest linear dimension of a VS to detect tumor growth could also run an automated segmentation protocol on the scan, and use the two sets of data to evaluate growth without resorting to time‐ and labor‐intensive manual segmentation.…”

“…For example, in computer-aided detection in breast mammography, the radiologist combines their interpretation of a study with an algorithm's interpretation to develop a final read. 21 In a similar and simpler fashion, the radiologist measuring the greatest linear dimension of a VS to detect tumor growth could also run an automated segmentation protocol on the scan, and use the two sets of data to evaluate growth without resorting to time-and laborintensive manual segmentation. For reference, it takes on average 1.34 seconds to run our segmentation algorithm on an identified VS using the graphics card described in the Materials and Methods section, versus minutes to manually segment a tumor, depending on slice thickness and tumor size.…”

Assessment of Raman Spectroscopy for Reducing Unnecessary Biopsies for Melanoma Screening

“…Although our results demonstrate the feasibility of reducing unnecessary biopsies using our in vivo Raman-based tool in a typical skin cancer screening setting, we have not asked dermatologists to hypothesize about whether they would have changed their decision to biopsy if our results suggested that a biopsy was not needed. Research in other areas of medical imaging demonstrates that computer aided detection or diagnosis recommendations do not necessarily change a physician’s actions [ 30 ]. Therefore, future studies are needed on the acceptability of our tool in dermatological practice.…”