Identification of Schizophrenia using LSTM Recurrent Neural Network
| Autor: | Abinaya Sundari R, C M Sujatha |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Computer science
business.industry Schizophrenia (object-oriented programming) Deep learning Disease Machine learning computer.software_genre Identification (information) Recurrent neural network nervous system Classifier (linguistics) Effective treatment Artificial intelligence business computer Temporal information |
| Zdroj: | 2021 Seventh International conference on Bio Signals, Images, and Instrumentation (ICBSII). |
| DOI: | 10.1109/icbsii51839.2021.9445189 |
| Popis: | Schizophrenia is a severe psychiatric illness that greatly affects the quality of life. Early discovery of psychiatric onset and progression are of great significance for bestowing effective treatment to prevent or mitigate further degeneration in the disease. Functional MRI possesses large number of subtle information which helps in early identification of disease. Current fMRI analysis focuses on spatial maps and connectivity patterns rather than temporal information. In this work an attempt has been made to identify schizophrenia with the use of LSTM deep learning model using ROI based time series extracted from fMRI. The classifier accuracy of 81.3% demonstrates that, use of ROI time series can aid as better diagnostic measure in nearby future. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|