Scalp Diagnostic System With Label-Free Segmentation and Training-Free Image Translation

2024-06-25

arXiv.org

AI Summary

Scalp disorders are highly prevalent globally, yet AI-assisted diagnosis remains challenging due to scarcity of dermatological experts, prohibitively high costs of pixel-level annotation, and severe class imbalance in clinical data. To address these issues, we propose a lightweight AI diagnostic system for scalp diseases and alopecia. Our method introduces: (1) a novel prompt-based, label-free hair segmentation approach driven by pseudo-image–label pairs—eliminating the need for pixel-wise ground truth; (2) DiffuseIT-M, a diffusion-based model enabling zero-shot, structure-preserving image augmentation and unsupervised image translation; and (3) an integrated framework combining prompt learning, pseudo-label generation, and feature-driven quantitative evaluation for multi-disease classification and alopecia staging. Evaluated on real-world clinical datasets, our system achieves significant improvements in diagnostic accuracy, demonstrating strong clinical applicability, cross-domain generalizability, and deployment efficiency.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: AI for Accessibility
• Machine Learning: Scalability of ML Systems

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Scalp diseases and alopecia affect millions of people around the world, underscoring the urgent need for early diagnosis and management of the disease. However, the development of a comprehensive AI-based diagnosis system encompassing these conditions remains an underexplored domain due to the challenges associated with data imbalance and the costly nature of labeling. To address these issues, we propose ScalpVision, an AI-driven system for the holistic diagnosis of scalp diseases and alopecia. In ScalpVision, effective hair segmentation is achieved using pseudo image-label pairs and an innovative prompting method in the absence of traditional hair masking labels. This approach is crucial for extracting key features such as hair thickness and count, which are then used to assess alopecia severity. Additionally, ScalpVision introduces DiffuseIT-M, a generative model adept at dataset augmentation while maintaining hair information, facilitating improved predictions of scalp disease severity. Our experimental results affirm ScalpVision's efficiency in diagnosing a variety of scalp conditions and alopecia, showcasing its potential as a valuable tool in dermatological care.

Problem

• Diagnosing scalp disorders with limited expert access
• Addressing data imbalance and lack of segmentation labels
• Developing training-free image translation for dataset augmentation

Innovation

• Label-free segmentation using pseudo image-label pairs
• Training-free image translation with DiffuseIT-M model
• AI-driven holistic diagnosis without traditional masking labels