Abstract
Systemic Lupus Erythematosus (SLE) is a complex and often underdiagnosed autoimmune disease that affects multiple organs and presents with a wide range of symptoms-ranging from fatigue and joint pain to life-threatening organ damage. One of its most visible and diagnostically significant indicators is the Butterfly Malar Rash (BMR), a distinctive facial rash that often resembles other common dermatological conditions like rosacea, acne, eczema, and fifth disease. This overlap can lead to misdiagnosis or delayed detection, especially in busy clinical environments. To assist dermatologists in distinguishing BMR from similar facial rashes, this study explores the development of an AI-powered image classification tool using deep learning. We collected and curated a dataset of diverse facial images featuring BMR and other lookalike skin conditions. Several Convolutional Neural Network (CNN) architectures, including DenseNet121, MobileNetV2, ResNet50, VGG16, and Xception, were first evaluated using transfer learning. While CNNs showed promising results, their performance was ultimately outpaced by Vision Transformer (ViT)-based models, particularly ViT-Small. When trained with CutMix augmentation, the ViT-Small model achieved a classification accuracy of 96%, outperforming all other tested models by approximately 10%. The goal of this work is not to replace medical expertise, but rather to provide dermatologists with a fast, consistent, and explainable tool to support early detection of BMR. This kind of assistive technology can be especially valuable in time-constrained settings or where rapid triage is required. Our findings highlight the potential of transformer-based models in medical image analysis and point toward a future where AI can augment clinical workflows, reduce diagnostic uncertainty, and ultimately improve patient outcomes in the context of SLE care.
Advisor
Naseef Mansoor
Committee Member
Rajeev Bukralia
Committee Member
Mansi Bhavsar
Date of Degree
2025
Language
english
Document Type
Thesis
Degree
Master of Science (MS)
Program of Study
Data Science
Department
Computer Information Science
College
Science, Engineering and Technology
Recommended Citation
Farhat, S. L. (2025). A vision transformer based assistive system for dermatological diagnosis in systemic lupus erythematosus [Master’s thesis, Minnesota State University, Mankato]. Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University, Mankato. https://cornerstone.lib.mnsu.edu/etds/1524/
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Rights Statement
