Summary of Smart: Towards Pre-trained Missing-aware Model For Patient Health Status Prediction, by Zhihao Yu et al.
SMART: Towards Pre-trained Missing-Aware Model for Patient Health Status Prediction
by Zhihao Yu, Xu Chu, Yujie Jin, Yasha Wang, Junfeng Zhao
First submitted to arxiv on: 15 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 The proposed SMART approach is a self-supervised representation learning method designed for patient health status prediction using electronic health record (EHR) data. Existing methods struggle with missing data in EHRs, which can lead to inaccurate predictions and correlations. To address this issue, SMART learns to impute missing values through a novel pre-training approach that reconstructs representations in the latent space. This approach focuses on learning higher-order representations and promotes better generalization and robustness to missing data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary SMART is an approach for patient health status prediction using EHR data with missing information. The method learns to predict patients’ health based on their electronic records, which often have missing data. Other methods can make wrong predictions because of this missing data. SMART does better by learning how to fill in the gaps and making more accurate predictions. |
Keywords
» Artificial intelligence » Generalization » Latent space » Representation learning » Self supervised
