• Soujenya.voggu
• Dr. Shadab Siddiqui
• Soujenya.voggu

: Skin diseases are the most common diseases than other diseases. Fungal infections, allergies, viruses, or bacteria are reasons for skin diseases. We have suggested a method of identification and classification of skin diseases that can be used for HAM10000 image datasets which amount to about (10,015) pictures and are posted by the International Skin Image Collaboration (ISIC). Various algorithms for determining and classifying skin disease images are learned in the literature. Still, multiple algorithms failed to extract lesion edges accurately and organize them. We propose ResNet-50 neural network with enhanced random forest algorithms (ERF) to improve skin image identification and classification reliability. The ResNet-50 is utilized for the task of extraction image features and ERF for the image classification task in this work and compares them with various algorithms based on the HAM10000 image dataset. The proposed method utilizes the feature map of ResNet-50 and gives it to the Enhanced Random Forest algorithm for classification. We found that the suggested algorithm is better accurate when compared with existing algorithms in this field and has strong artifacts in the skin images.

Skin disease detection, Deep learning, Random Forest, ResNet-50, HAM10000

"Skin Disease Detection Using Resnet-50 and Machine Learning Based Enhanced Random Forest Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 9, page no.f491-f505, September-2024, Available :http://www.jetir.org/papers/JETIR2409587.pdf

"Skin Disease Detection Using Resnet-50 and Machine Learning Based Enhanced Random Forest Approach", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 9, page no. ppf491-f505, September-2024, Available at : http://www.jetir.org/papers/JETIR2409587.pdf