Researchers Develop AI To Predict Hospital Readmission Rates From Clinical Notes 29
Researchers at New York University and Princeton have developed a framework that evaluates clinical notes and autonomously assigns a risk score indicating whether patients will be readmitted within 30 days. They claim that the code and model parameters, which are publicly available on Github, handily outperform baselines. VentureBeat reports: As the researchers point out in a preprint paper on Arxiv.org, clinical notes use abbreviations and jargon, and they're often lengthy, which poses an AI system design challenge. To overcome it, they used a natural language processing method -- Google's bidirectional encoder representations from transformers, or BERT -- that captures interactions between distant words in sentences by incorporating global, long-range information. Each clinical note is represented as a collection of tokens, or subword units extracted from text in a preprocessing step. From multiple sequences of these, ClinicalBERT identifies which tokens are associated with which sequence. It also learns the position of tokens from variables corresponding to the sequences, and inserts a special token used in classification tasks in front of every sequence.
To train ClinicalBERT, the team sourced a corpus of clinical notes and masked 15 percent of the input tokens, forcing the model to predict the concealed tokens and whether any two given two sentences were in consecutive order. Then, drawing on the Multiparameter Intelligent Monitoring in Intensive Care (MIMIC-III), an electronic health records data set comprising over two million notes from 58,976 hospital admissions of 38,597 patients, the researchers fine-tuned the system for clinical forecasting tasks. Tested on a sample set consisting of 30 pairs of medical terms designed to assess medical term similarity, the authors report, ClinicalBERT achieved a high correlation score, indicating that its tokens captured similarity between medical concepts terms. Heart-related concepts like myocardial infarction, atrial fibrillation, and myocardium were close together, they say, and renal failure and kidney failure were also close.
To train ClinicalBERT, the team sourced a corpus of clinical notes and masked 15 percent of the input tokens, forcing the model to predict the concealed tokens and whether any two given two sentences were in consecutive order. Then, drawing on the Multiparameter Intelligent Monitoring in Intensive Care (MIMIC-III), an electronic health records data set comprising over two million notes from 58,976 hospital admissions of 38,597 patients, the researchers fine-tuned the system for clinical forecasting tasks. Tested on a sample set consisting of 30 pairs of medical terms designed to assess medical term similarity, the authors report, ClinicalBERT achieved a high correlation score, indicating that its tokens captured similarity between medical concepts terms. Heart-related concepts like myocardial infarction, atrial fibrillation, and myocardium were close together, they say, and renal failure and kidney failure were also close.
Awesome (Score:2)
virtual doctor says "You get leprosy good bye" (Score:2)
virtual doctor says "You get leprosy good bye"
https://www.youtube.com/watch?... [youtube.com]
Re:virtual doctor says "You got leprosy good bye" (Score:2)
Re:virtual doctor says "You got leprosy good bye"
Re: (Score:2)
And just maybe, with a richer database, based on the patient's latest exam(s) and previous history, as found in their recorded electronic health records, AI might outperform doctors with better diagnoses, recommended prescriptions, rehabilitative care, etc.
Not that I'm saying AI should do the exam itself. Having AI administer any
doctor handwriting ocr with an 95+ rate? (Score:3)
doctor handwriting ocr with an 95+ rate?
Re: (Score:2)
Don't think that is possible. This is based on electronic records.
Researchers should work on something else (Score:3)
It's hilarious (Score:2)
Re: (Score:2)
Who do you think understands law better.
I can find you any number of layers that will argue either case.
Re: (Score:2)
Often in the same legal proceeding.
Background (Score:4, Informative)
And this stuff is honestly pretty useful. It helps earmark which patients might need a bit of extra monitoring or guidance, or which ones might benefit from having a nurse check in on them at home. It means figuring out where to proactively spend limited human resources in order to decrease the chances someone is going to end up right back in your ER. Good stuff.