Summary of Unifying Interpretability and Explainability For Alzheimer’s Disease Progression Prediction, by Raja Farrukh Ali and Stephanie Milani and John Woods and Emmanuel Adenij and Ayesha Farooq and Clayton Mansel and Jeffrey Burns and William Hsu
Unifying Interpretability and Explainability for Alzheimer’s Disease Progression Prediction
by Raja Farrukh Ali, Stephanie Milani, John Woods, Emmanuel Adenij, Ayesha Farooq, Clayton Mansel, Jeffrey Burns, William Hsu
First submitted to arxiv on: 11 Jun 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 Reinforcement learning (RL) has shown promise in predicting Alzheimer’s disease progression, particularly due to its ability to model domain knowledge. However, it is unclear which RL algorithms are well-suited for this task. Our work addresses this question by comparing the performance of four contemporary RL algorithms in predicting brain cognition over 10 years using only baseline data. We also apply SHAP to explain the decisions made by each algorithm, providing insights into key factors influencing AD progression at both global and individual patient levels. While one RL method satisfactorily models disease progression, post-hoc explanations indicate that all methods fail to capture the importance of amyloid accumulation, a pathological hallmark of Alzheimer’s disease. Our work aims to merge predictive accuracy with transparency, assisting clinicians in enhancing disease progression modeling for informed healthcare decisions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses a type of artificial intelligence called reinforcement learning (RL) to try to predict how quickly people will develop Alzheimer’s disease. RL is good at understanding complex information and making predictions. The researchers compared four different types of RL algorithms to see which one works best for this task. They also used a technique called SHAP to understand why each algorithm made certain predictions. This helps us learn more about what factors are important in predicting Alzheimer’s disease progression. While one type of algorithm did a good job, all the methods had trouble capturing an important part of the disease, which is the buildup of amyloid plaques in the brain. |
Keywords
» Artificial intelligence » Reinforcement learning
