A Novel Adaptive Neuro-Gradient Model for Accurate Lupus Disease Detection using Intelligent Feature Optimization
Year: 2026
Type: conference-paper
Abstract: The systemic Lupus Erythematosus is a heterogeneous and multifaceted autoimmune disease, which is challenging to diagnose, because of the heterogeneous symptoms and the high-dimensional biological data. Conventional machine learning methods would also tend to have redundancy of features, noise, and lack of flexibility hence less diagnostic performance. In this regard, to overcome these constraints, this paper will present a new Adaptive Neuro-Gradient Model coupled with intelligent optimization of features to help detect lupus disease. The methodology is aimed at processing gene expression data, identifying most informative features, and using a gradient-based learning mechanism of neural type to improve the accuracy of classification. The dimensionality of data is greatly minimized in the process of feature optimization and information that is very important is maintained thus enhancing the computational efficiency. It has been experimentally proven that the proposed model has an accuracy of 96%, and a high level of precision, recall, F1-score, and AUC. It is also demonstrated that the model has a better generalization ability as compared to other approaches. On the whole, the framework suggested is a valid and effective approach to the early and precise identification of the lupus disease.
Authors: Flemitha Mary D, Govindaraj Ramkumar
Institution: Saveetha University
Language: English
Subfield: Rheumatology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W7203483961