Candidiasis is a fungal infection caused by Candida species, most commonly Candida albicans. It can affect various parts of the body including the mouth, throat, esophagus, vagina, and bloodstream. Early detection and prediction of candidiasis risk is crucial for effective treatment and prevention of complications.
To address this issue, we use Singa to implement a machine learning model for predicting candidiasis disease. The model uses tabular data with various clinical features to predict the likelihood of candidiasis infection.
The dataset used in this task is MIMIC-III after preprocessed. Before starting to use this model for candidiasis disease prediction, download the sample dataset for candidiasis disease prediction: https://github.com/lzjpaul/singa-healthcare/tree/main/data/candidiasis
data includes the scripts for preprocessing Candidiasis datasets.
model includes the MLP model construction codes by creating a subclass of Module to wrap the neural network operations of each model.
train.py is the training script, which controls the training flow by doing BackPropagation and SGD update.
python train.py candidiasisnet -dir pathToDataset