International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Skin Disease Classification System |
|---|---|
| ABSTRACT | Skin cancer remains one of the most prevalent cancers worldwide, and timely diagnosis plays a crucial role in improving patient outcomes. However, the traditional examination of dermatoscopic images is often labor- intensive and prone to differences in interpretation among dermatologists. This study introduces a binary classification framework for skin disease detection that automatically differentiates dermatoscopic images as benign or malignant. The model utilized transfer learning based on pretrained EfficientNetB0, B1, and B3 architectures, which were fine-tuned using a balanced dataset comprising 34,000 annotated images. The dataset was divided into training (70%), validation (10%), and testing (20%) sets to ensure a reliable performance assessment. Experimental evaluations revealed a high level of classification accuracy—achieving [90%] on the test data—along with strong sensitivity and specificity, confirming the model’s ability to accurately identify malignant lesions. Overall, the proposed method presents a scalable, cost-efficient, and effective tool to assist dermatologists in early-stage skin cancer screening, potentially enhancing diagnostic precision and speed in real-world clinical applications. |
| TITLE | |
| AUTHOR | ROSHAN KOTKONDAWAR, VIVEK KUMAR, RUDRAKSHA ASATI, SHASHWAT LAUTAWAR, FAHAD A. KHAN |
| PUBLICATION DATE | 2025-11-06 |
| VOLUME | 176 |
| DOI | DOI: 10.15680/IJIRCCE.2025.1311011 |
| pdf/11_Skin Disease Classification System.pdf | |
| KEYWORDS |
