Date of Award
Spring 5-2026
Language
English
Document Type
Honors Thesis
Advisor/Committee Chair
Kimberly Cornell
Abstract
Artificial intelligence (AI) has been used in various sectors, including healthcare. Machine learning (ML) is a subset of AI, and deep learning (DL) is a subset of ML. The capabilities of ML and DL have proven to be useful for classification, prediction, and clustering (An et al., 2023). Therefore, this technology can be leveraged to help classify popular psychiatric disorders. Major depressive disorder (MDD), bipolar disorder (BD), and Schizophrenia impact millions of people worldwide (Scangos et al., 2023). Their symptoms tend to overlap, making the disorders difficult to differentiate. Despite their similarities, each disorder is unique in some way, which makes identifying their differences worthwhile. A graph convolutional network was designed after a GNN on GitHub that achieved significant accuracy. The model was trained using a dataset for MDD, BD, and schizophrenia from genome-wide association studies (GWAS) and contains single-nucleotide polymorphisms (SNPs). The purpose of the data was to classify the three disorders based on SNP-SNP interactions. Core aspects of the GitHub model remained in the adapted version, such as the main libraries and edge classification. Other aspects were excluded completely or altered to fit the dataset. It was hypothesized that using preexisting successful GitHub models and adapting them to the GWAS dataset would lead to a high success rate. In actuality, the implemented version displayed severe overfitting. The model's classification ability was greater than random guessing, proving the possibility of such a task. Future research should be done to understand how machine learning and deep learning can be used to classify these extremely prevalent disorders meaningfully.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 International License.
Recommended Citation
Castillo, Nikaley, "Using a GCN to Classify Psychiatric Disorders" (2026). Informatics. 5.
https://scholarsarchive.library.albany.edu/honorscollege_informatics/5