Source code for CS598 Deep learning for healthcare
Final.ipynb is the implementation notebook.
-
Original peper: Application of deep and machine learning techniques for multi-label classification performance on psychotic disorder diseases, Israel Elujide, Stephen G. Fashoto, Bunmi Fashoto, Elliot Mbunge, Sakinat O. Folorunso, Jeremiah O. Olamijuwon, Informatics in Medicine Unlocked, Volume 23, 2021, 100545, ISSN 2352-9148, https://doi.org/10.1016/j.imu.2021.100545. (https://www.sciencedirect.com/science/article/pii/S2352914821000356)
-
Data retrieved from the original paper can be downloaded here https://drive.google.com/file/d/1-4K2WLgzaEggCIphVCD7zyGpJMjTepWU/view?usp=sharing
-
Result
The project reproduces three claims chosen from the original paper.
-
Deep learning model with class imbalance outperforms machine learning models in multi-label classification used in the paper.
-
The best performance without class imbalance is the random forest model, among deep learning and machine learning models.
-
Insomnia and bipolar disorder are highly correlated.
