Summary of Next Visit Diagnosis Prediction Via Medical Code-centric Multimodal Contrastive Ehr Modelling with Hierarchical Regularisation, by Heejoon Koo
Next Visit Diagnosis Prediction via Medical Code-Centric Multimodal Contrastive EHR Modelling with Hierarchical Regularisation
by Heejoon Koo
First submitted to arxiv on: 22 Jan 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
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 NECHO framework is a novel multimodal contrastive learning algorithm for predicting next visit diagnoses from Electronic Health Records (EHRs). This framework addresses the heterogeneous and hierarchical characteristics of EHR data, which previous studies have not adequately handled. NECHO integrates multifaceted information, including medical codes, demographics, and clinical notes, using a tailored network design and bimodal contrastive losses centered on medical code representations. The algorithm also incorporates hierarchical regularization through parental level information in medical ontology to learn the structure of EHR data. Experimental results on MIMIC-III data demonstrate the effectiveness of this approach. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary NECHO is a new way to use electronic health records (EHRs) to predict what might happen next with a patient’s health. Right now, doctors and patients don’t have good ways to make plans for the future because most computer programs can’t understand all the different types of information in EHRs. NECHO fixes this by combining many different kinds of data together in a special way. It uses a lot of information about each person, including their medical history, age, and what doctors have written down about them. This helps computers better predict what will happen next with someone’s health. |
Keywords
* Artificial intelligence * Regularization
