Related Experiment Video
Updated: Jun 18, 2025
09:37
Combining Reflectance Confocal Microscopy with Optical Coherence Tomography for Noninvasive Diagnosis of Skin Cancers via Image Acquisition
Published on: August 18, 2022
2.3K
Deep learning algorithms for melanoma detection using dermoscopic images: A systematic review and meta-analysis
Zichen Ye1, Daqian Zhang1, Yuankai Zhao1
1School of Population Medicine and Public Health, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Artificial Intelligence in Medicine
|August 1, 2024
Summary
Deep learning (DL) algorithms show high accuracy in diagnosing melanoma from dermatoscopic images, matching senior dermatologists. This technology can support clinical decisions, but more research is needed.
Area of Science:
- Dermatology
- Artificial Intelligence
- Medical Imaging
Background:
- Melanoma diagnosis relies on early identification for successful treatment.
- Deep learning (DL) shows promise in analyzing medical images for cancer detection.
- Assessing DL performance in melanoma diagnostics is crucial.
Approach:
- Systematic review and meta-analysis of studies on DL for melanoma detection using dermatoscopic images.
- Searched Ovid-Medline, Embase, IEEE Xplore, and Cochrane Library up to December 2021.
- Extracted diagnostic accuracy data (sensitivity, specificity, AUC) and analyzed human-machine comparison and cooperation.
Key Points:
- Pooled sensitivity: 82%, specificity: 87%, AUC: 0.92 for DL models.
- DL performance was comparable to senior dermatologists (AUC 0.90 vs 0.88).
- DL-assisted dermatologists achieved higher accuracy (AUC 0.87) than unassisted ones (AUC 0.76).
Conclusions:
- Deep learning algorithms demonstrate diagnostic accuracy on par with senior dermatologists for melanoma.
- DL can serve as a valuable tool to aid dermatologists in diagnostic processes.
- Further large-scale, multicenter studies are necessary to address AI implementation challenges in medical diagnostics.
