Benchmarking Continuous Time Models for Predicting Multiple Sclerosis Progression
Abstract: Multiple sclerosis is a disease that affects the brain and spinal cord, it can lead to severe disability and has no known cure. The majority of prior work in machine learning for multiple sclerosis has been centered around using Magnetic Resonance Imaging scans or laboratory tests; these modalities are both expensive to acquire and can be unreliable. In a recent paper it was shown that disease progression can be predicted effectively using performance outcome measures and demographic data. In our work we build on this to investigate the modeling side, using continuous time models to predict progression. We benchmark four continuous time models using a publicly available multiple sclerosis dataset. We find that the best continuous model is often able to outperform the best benchmarked discrete time model. We also carry out an extensive ablation to discover the sources of performance gains, we find that standardizing existing features leads to a larger performance increase than interpolating missing features.
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Submission Length: Regular submission (no more than 12 pages of main content)
Changes Since Last Submission: Since our last revision of the paper, we have produced the camera-ready version of our manuscript. The major change we made is to test RNN on the MSOAC dataset, as requested by the reviewers and action editor to include a further baseline. This represents a deep learning model designed for sequences not tested in Roy et al. 2022. We see that RNN performs worse than TCN and Neural CDE. The ranking being Neural CDE > TCN > RNN, further supporting the use of Neural CDE in the Multiple Sclerosis setting. The minor changes we made are fixing typos and adding the sections required for a TMLR camera-ready manuscript (author information, acknowledgements and author contributions).
Assigned Action Editor: ~Qibin_Zhao1
Submission Number: 1182