Year: 2025
Type: article
Abstract: Hair loss impacts approximately 80 million people in the United States and arises from factors such as aging, genetic predisposition, stress, and medication. In many cases, early signs of scalp and hair disorders remain unnoticed, delaying timely diagnosis and treatment. Neural networks, particularly in image-based diagnostics, have shown great promise across medical domains, including oncology and dermatology. This research investigates the application of deep learning for identifying three major scalp conditions: alopecia, psoriasis, and folliculitis. Major challenges included limited literature, a small dataset, and non-uniform image quality from online sources. To address these, the study compiled 150 scalp images and employed preprocessing techniques such as denoising, histogram equalization, augmentation, and class balancing. A two-dimensional Convolutional Neural Network (CNN) was trained on this dataset, achieving a training accuracy of 96.2% and a validation accuracy of 91.1%, with consistently high precision and recall across all classes. Additionally, a new scalp scan dataset was introduced to support future research, contributing to the development of AI-driven diagnostic tools for hair and scalp health assessment
Source: Kufa Journal of Engineering
Institution: Karunya University
Language: English
Topic: Hair Growth and Disorders
Subfield: Dermatology
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4415768204