Melanoma and Nevus Skin Lesion Classification Using Handcraft and Deep Learning Feature Fusion via Mutual Information Measures
importance 6/10
ai assisted
Cutaneous Melanoma Detection and Management dermatology stage: research
Problem
Detecting and classifying dangerous melanoma skin lesions from dermoscopic images is difficult and error-prone.
Why it matters
Accurate early melanoma detection is critical for patient survival.
What the AI did
A computer-aided detection system fuses handcrafted ABCD-rule features with CNN deep learning features using mutual information measurements for classification.
Results
not stated in source
Impact
Peer-reviewed CAD approach; clinical adoption not stated in source.
Record
| Venue | Entropy |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Jose A. Almaraz-Damian, Volodymyr I. Ponomaryov, Sergiy Sadovnychiy, Heydy Castillejos |
| Citations | 172 (OpenAlex, at import time) |
| References | 62 |
| Model or method | CNN pre-trained on ImageNet |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/e22040484 |
| Open access | gold |
| Keywords | melanoma, skin lesion classification, CNN, mutual information, feature fusion, CAD |
Importance rating
6
significant — thousands to millions affected
Melanoma detection from dermoscopic images; abstract calls accurate early detection critical for patient survival but gives no scale; significant clinical problem, narrower methodological advance.