Summary of Informing Clinical Assessment by Contextualizing Post-hoc Explanations Of Risk Prediction Models in Type-2 Diabetes, By Shruthi Chari et al.
Informing clinical assessment by contextualizing post-hoc explanations of risk prediction models in type-2 diabetes
by Shruthi Chari, Prasant Acharya, Daniel M. Gruen, Olivia Zhang, Elif K. Eyigoz, Mohamed Ghalwash, Oshani Seneviratne, Fernando Suarez Saiz, Pablo Meyer, Prithwish Chakraborty, Deborah L. McGuinness
First submitted to arxiv on: 11 Feb 2023
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 This study examines how Artificial Intelligence (AI) systems can be designed to support medical practitioners’ decision-making processes by providing contextual explanations that connect AI-generated inferences to the practitioner’s context. The researchers focus on comorbidity risk prediction scenarios and explore how relevant information can be extracted from medical guidelines to answer typical questions posed by clinical practitioners. They identify this as a question-answering task and employ state-of-the-art Large Language Models (LLMs) to present contextual explanations around AI-generated inferences. The benefits of these explanations are evaluated by building an end-to-end AI pipeline, including data cohorting, AI risk modeling, post-hoc model explanations, and prototyping a visual dashboard to present insights from different context dimensions and data sources. The study demonstrates the feasibility of deploying LLMs like BERT and SciBERT to extract relevant explanations that support clinical usage. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary AI systems can help doctors make better decisions by providing clear explanations for their predictions. This study shows how AI can be used in a real-world medical setting to predict the risk of Chronic Kidney Disease, a common condition that affects people with type-2 diabetes. By using special algorithms and language models, the researchers were able to extract useful information from medical guidelines and provide it to doctors in a way that makes sense for their work. |
Keywords
» Artificial intelligence » Bert » Question answering
