EUVIMEDEUVIMED entdecken
Systematische Übersicht

A Systematic Review of Machine Learning and Deep Learning Approaches for Skin Cancer Diagnosis

Geeta Mahadev Mangela and Dr. Poonam N. Sonar

International Journal of Advanced Research in Science Communication and Technology · 15.06.2026 · 163

DOI: 10.48175/ijarsct-36717 ↗

Kurz gefasst

Systematische Übersicht in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 15.06.2026. Verfasst von Geeta Mahadev Mangela and Dr. Poonam N. Sonar. Die Arbeit nennt 20 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Skin cancer continues to be one of the most common and clinically difficult cancers around the world, owing to its fast-growing rate, extremely high mortality rates in cases of malignant melanomas, and the essential need to diagnose at the initial stage. Traditionally, the process of diagnosing skin cancer was performed using visual inspection, dermoscopy, and histopathology biopsy, which formed the core of the clinical assessment of the disease. These methods continue to rely heavily on the experience of the dermatologist, take much time to perform, and are subjective in nature. Recently, the diagnostic approach has moved away from traditional clinical procedures towards automated systems that leverage artificial intelligence and enable highly precise identification of malignancies. The present literature review aims to examine recent developments and methodologies adopted in ML and DL-based systems that can be used in diagnosing and classifying skin cancer. A systematic analysis is made based on a taxonomy, comprising medical image data sources and preprocessing approaches, algorithm architectures encompassing traditional ML models, CNN models, transfer learning, and deep neural architectures, as well as performance evaluation measures adopted in research. Moreover, this review transcends traditional algorithmic comparisons by addressing fundamental issues that hinder the successful adoption of automation technology for real-world applications on a large scale. Issues such as data scarcity in labeled medical datasets, class imbalance, overfitting in deep neural networks, the black box nature of neural models, and computational complexity, among others, have been investigated in this context.

Zitieren

APA

Sonar, G. M. M. A. D. P. N. (2026). A Systematic Review of Machine Learning and Deep Learning Approaches for Skin Cancer Diagnosis. International Journal of Advanced Research in Science Communication and Technology, 163. https://doi.org/10.48175/ijarsct-36717

Vancouver

Sonar GMMADPN. A Systematic Review of Machine Learning and Deep Learning Approaches for Skin Cancer Diagnosis. International Journal of Advanced Research in Science Communication and Technology. 2026:163. doi:10.48175/ijarsct-36717

BibTeX
@article{Sonar2026,
  author = {Sonar, Geeta Mahadev Mangela and Dr. Poonam N.},
  title = {A Systematic Review of Machine Learning and Deep Learning Approaches for Skin Cancer Diagnosis},
  journal = {International Journal of Advanced Research in Science Communication and Technology},
  year = {2026},
  pages = {163},
  doi = {10.48175/ijarsct-36717}
}
Kostenlos starten →