AI Depression Check
Private Mental Health AI: Users fill out a form to check for depression, securely uploaded to Nillion. A Neural Network provides a blind prediction. Trained on Kaggle's Depression Anxiety Stress Scales (95% accuracy), it predicts 5 classes. Deployed on Nillion.
Project Description
Users fill in a form to diagnose depression, this form is securely uploaded to Nillion as a secret and a prediction blindly is given by a Neural Network also uploaded on Nillion. This allows people who might want to check whether they are suffering from depressive symptoms before going to a specialist yet, in a private way, where the app they use for it will not sell the fact that they have or they think they have depression to third parties.
How it's Made
We use the https://www.kaggle.com/datasets/lucasgreenwell/depression-anxiety-stress-scales-responses dataset to train a simple but highly performant NN (95% accuracy) to predict 5 different classes: Normal, Mild, Moderate, Severe or Extremely Severe.
We upload this model to the NIllion devnet on the working version, since the testnet is momentarily congested and storing even an integer costs levels of magnitude above what can be obtained on the faucet. I cant link the devnet branch of the repo below, make sure to examine that one please.
