Abstract
Systemic Lupus Erythematosus (SLE) poses significant challenges due to its complex and varied symptoms making diagnosis extremely challenging and time consuming. Symptoms of SLE often mimics other autoimmune or physical conditions and around 5 million people worldwide suffers from this condition, as reported by the Lupus Foundation of America during their study in 2019. However, diagnosis is much more difficult in developing countries with backdated clinical technology and setup therefore, making it virtually unknown the exact number of SLE patient count worldwide. Among all the heterogeneous symptoms presented by SLE, Butterfly Malar Rash (BMR) is one of the symptoms that is prevalent in the majority of SLE patients, which is a butterfly shaped malar rash that appears on the face. This BMR is often misdiagnosed since it is mimicked by other skin conditions like Rosacea, Acne, Eczema, Fifth disease and so on. To speed up the diagnosis process, we experimented with multiple CNN models in this study to create a tool for dermatologists that can classify BMR from other similar facial rashes. We proposed a customized state-of-the-art Densenet121-C model that outperformed the existing models achieving an accuracy of 97.3% and 98% precision rate for detecting BMR caused by SLE. We also scoped out a system architecture for model deployment in a clinical setup that can act as a tool to classify BMR from facial images in real-time, thus helping the dermatologists speed up the process of diagnosis. For model training, we constructed two custom datasets based on various online sources. During our experiments we noticed improvements in the performance of our model, especially when the models were retrained on a better-quality dataset. Medical images, especially facial images, are difficult to get because of patient confidentiality and validity issues. Therefore, we experimented with StyleGAN3 technology and laid out an outline for augmentation to generate synthetic images of BMR, which can be used for training such classifier as a future scope.
Advisor
Naseef Mansoor
Committee Member
John Burke
Date of Degree
2024
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
Dey, S. B. (2024). A deep learning model for early diagnosis of Systemic Lupus Erythematosus from facial images [Master’s alternative plan paper, Minnesota State University, Mankato]. Cornerstone: A Collection of Scholarly and Creative Works for Minnesota State University, Mankato. https://cornerstone.lib.mnsu.edu/etds/1470/
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Included in
Rights Statement
