Summary of Doctor: a Multi-disease Detection Continual Learning Framework Based on Wearable Medical Sensors, by Chia-hao Li and Niraj K. Jha
DOCTOR: A Multi-Disease Detection Continual Learning Framework Based on Wearable Medical Sensors
by Chia-Hao Li, Niraj K. Jha
First submitted to arxiv on: 9 May 2023
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Human-Computer Interaction (cs.HC); Signal Processing (eess.SP)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Modern advances in machine learning (ML) and wearable medical sensors (WMSs) have enabled ML-driven disease detection for smart healthcare. The proposed DOCTOR framework addresses challenges in conventional ML-driven methods, which lack adaptability to distribution shifts and new task classification classes, require retraining from scratch for each new disease, consume excessive memory, drain the battery faster, and complicate the detection process. DOCTOR is a multi-disease detection continual learning (CL) framework based on WMSs, employing a multi-headed deep neural network (DNN) and a replay-style CL algorithm. This framework continually learns new missions, counteracting catastrophic forgetting with data preservation and synthetic data generation methods. The multi-headed DNN enables simultaneous disease detection based on user WMS data. DOCTOR demonstrates efficacy in maintaining high disease classification accuracy in various CL experiments. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary DOCTOR is a smart healthcare system that uses machine learning to detect diseases from wearable medical sensors. Right now, it’s hard to make these systems adapt to new situations or learn from experiences. The DOCTOR team created a special kind of artificial intelligence (AI) called a multi-headed deep neural network that can learn and remember multiple things at the same time. This AI is trained using a special algorithm that helps it forget less as it learns more, so it gets better over time. The result is a system that can accurately detect diseases from wearable sensors without needing to be re-trained every time something new comes along. |
Keywords
* Artificial intelligence * Classification * Continual learning * Machine learning * Neural network * Synthetic data
