A Robust Skin Cancer Detection and Classification Model Based on Improved Convolutional Neural Networks
The American Journal of Medical Sciences and Pharmaceutical Research · 23.05.2026 · 08 (05) · 79-93
DOI: 10.37547/tajmspr/volume08issue05-17 ↗Kurz gefasst
Fachartikel in der Zeitschrift The American Journal of Medical Sciences and Pharmaceutical Research, online erschienen am 23.05.2026. Verfasst von Gona Mohammed Hasan.
Abstract (Originalsprache)
Skin cancer (SC) is a major health problem that demands early detection and classification for the best chances of survival. Existing computer vision techniques lack the capacity to address the fine-grained variability exhibited by skin lesion features across surfaces. It is important to identify the SC signs early as the incidence and death rates of this disease are rising rapidly, along with the cost of care. Techniques for diagnosing SC often rely on visual analysis or techniques that do not have accurate results; this can be dangerous for the patient. Hence, deep learning (DL) methods have proved helpful for researchers to construct different methods for the early detection of SC. To distinguish SC from melanoma, these techniques used the lesion's color, size, form, symmetry, and other features. Using the HAM10000 dataset, a sizable and varied dataset, this work suggests a novel DL-based SC detection and diagnosis scheme (DL-SCDCM) that employs DL techniques, particularly CNN, to guarantee an accurate but efficient diagnosis. The proposed CNN model was trained before testing, and it demonstrated impressive performance, correctly diagnosing seven distinct skin lesion types with an accuracy of 96.9 %. Furthermore, the results were compared to those of other research that proposed a slightly different approach, and the proposed model outperformed the others in these comparisons.
Zitieren
APA
Hasan, G. M. (2026). A Robust Skin Cancer Detection and Classification Model Based on Improved Convolutional Neural Networks. The American Journal of Medical Sciences and Pharmaceutical Research, 08(05), 79-93. https://doi.org/10.37547/tajmspr/volume08issue05-17
Vancouver
Hasan GM. A Robust Skin Cancer Detection and Classification Model Based on Improved Convolutional Neural Networks. The American Journal of Medical Sciences and Pharmaceutical Research. 2026;08(05):79-93. doi:10.37547/tajmspr/volume08issue05-17
BibTeX
@article{Hasan2026,
author = {Hasan, Gona Mohammed},
title = {A Robust Skin Cancer Detection and Classification Model Based on Improved Convolutional Neural Networks},
journal = {The American Journal of Medical Sciences and Pharmaceutical Research},
year = {2026},
volume = {08},
number = {05},
pages = {79--93},
doi = {10.37547/tajmspr/volume08issue05-17}
}
EUVIMED