• Version
• Download 20
• File Size 333.26 KB
• File Count 1
• Create Date 1 December 2025
• Last Updated 1 December 2025

Deep Learning-Based Detection of Atopic Dermatitis from Whole-Body

Trishala AG 1 , Dr.P.Amudha2, Dr.R.Ahila3

1PG Scholar,2Professor and Head, 3Associate Professor

Department of Computer Science and Engineering, Avinashi lingam Institute for Home Science and Higher

Education for Women, Coimbatore, Tamil Nadu, India

 

 

 ABSTRACT

Atopic dermatitis (AD) is a chronic inflammatory skin disease that causes red, itchy, inflammatory lesions and is heavily exacerbated by environmental factors. An early and accurate diagnosis is fundamental to commencing the appropriate therapy and preventing complications. It detect atopic dermatitis from generic dermatological images automatically using convolutional neural networks (CNNs). Pre-trained DenseNet201 architecture is used on a collection of high-resolution skin images labeled "Atopic Dermatitis" and "Healthy". To improve generalization, the data is  undergoes  pre-processing methods including normalization, de-noising and augmentation techniques. The model is evaluated on strong performance metrics of accuracy, precision, recall and F1-score. The entire pipeline is deployed using stream lit through an interactive web application enabling users and medical professionals to make real-time predictions .

Keywords: Atopic Dermatitis, DenseNet201, Deep Learning, Classification, Skin Disease Detection, CNN, AI in Healthcare, Image Preprocessing.

Download