Deep Learning Algorithms Versus Radiologists in Digital Breast Tomosynthesis for Breast Cancer Detection: Systematic Review and Meta-Analysis - Report - MDSpire
Advertisement
Deep Learning Algorithms Versus Radiologists in Digital Breast Tomosynthesis for Breast Cancer Detection: Systematic Review and Meta-Analysis
Clinical Report: Comparative Analysis of Deep Learning Models and Radiologists in Detecting Breast Cancer
Overview
This systematic review and meta-analysis evaluates the diagnostic performance of deep learning (DL) algorithms versus radiologists in detecting breast cancer via digital breast tomosynthesis (DBT). The analysis includes data from 38,565 patients, highlighting the potential of DL to enhance detection rates while addressing challenges such as false positives and negatives.
Background
Breast cancer remains the most commonly diagnosed cancer and a leading cause of cancer mortality among women globally. The transition to digital breast tomosynthesis (DBT) has improved detection rates but also introduced challenges in interpretation, particularly regarding false positives and negatives. The integration of deep learning algorithms aims to assist radiologists in overcoming these challenges and improving diagnostic accuracy.
Data Highlights
The meta-analysis included data from 38,565 patients across various studies, comparing the performance of DL algorithms and radiologists.
Key Findings
DBT significantly improves cancer detection rates compared to traditional digital mammography.
Deep learning algorithms show potential in enhancing lesion detection and classification.
Challenges with false positives and false negatives persist, particularly for certain breast cancer subtypes.
The performance of DL algorithms varies based on the experience level of radiologists.
Increased image volume from DBT may lead to radiologist fatigue and cognitive overload.
Clinical Implications
Radiologists should consider the integration of deep learning algorithms to enhance diagnostic accuracy in breast cancer detection. Continuous training and experience remain crucial in interpreting DBT images effectively, particularly in light of the challenges posed by false positives and negatives.
Conclusion
The findings underscore the importance of combining deep learning technologies with radiologist expertise to improve breast cancer detection rates. Ongoing research and development in this area are essential to optimize diagnostic performance.
