This repository is the PUBLIC repo for students of the 2018 MIT BWSI Medlytics course for WEEK 2. Many notebooks do not include solutions, and instead has [# Your code here] where students should write their own code.

This repository does not allow pull requests so we recommend you fork this repository and work on your own copy.

Datasets used in this repo:

Sleep data challenge from Massachusetts General Hospital’s (MGH), Computational Clinical Neurophysiology Laboratory (CCNL), and the Clinical Data Animation Laboratory (CDAC): https://www.physionet.org/physiobank/database/challenge/2018/ Human Recognition Using Smartphone Software Dataset from UCI Machine Learning Repository: https://archive.ics.uci.edu/ml/datasets/human+activity+recognition+using+smartphones

• Contains the Week 2 Signal Processing Challenge. This challenge requires students to work with segments of signals from electroencephalography, electrooculography, electromyography, respiratory airflow, and electrocardiography. From these segments they must predict whether the patient was in an aroused (awake), in non-REM1, non-REM2, non-REM3, or REM state. Created by Brian Xia

• Contains lesson notebooks (and solutions) on the SIR model, Time Series Exercises 1 and 2, Signal Processing, Fourier Transformations, Signal Cleanup, ML on Signals, ANNs on Signal Data Using Keras, Introduction to CNNs, and 1D Convolutional Nets on Signal Data.