Literature-Augmented Clinical Outcome Prediction

11/16/2021
by   Aakanksha Naik, et al.
0

Predictive models for medical outcomes hold great promise for enhancing clinical decision-making. These models are trained on rich patient data such as clinical notes, aggregating many patient signals into an outcome prediction. However, AI-based clinical models have typically been developed in isolation from the prominent paradigm of Evidence Based Medicine (EBM), in which medical decisions are based on explicit evidence from existing literature. In this work, we introduce techniques to help bridge this gap between EBM and AI-based clinical models, and show that these methods can improve predictive accuracy. We propose a novel system that automatically retrieves patient-specific literature based on intensive care (ICU) patient information, aggregates relevant papers and fuses them with internal admission notes to form outcome predictions. Our model is able to substantially boost predictive accuracy on three challenging tasks in comparison to strong recent baselines; for in-hospital mortality, we are able to boost top-10 of over 25

READ FULL TEXT
research
05/19/2017

Learning Effective Representations from Clinical Notes

Clinical notes are a rich source of information about patient state. How...
research
02/08/2021

Clinical Outcome Prediction from Admission Notes using Self-Supervised Knowledge Integration

Outcome prediction from clinical text can prevent doctors from overlooki...
research
01/24/2019

Extracting PICO elements from RCT abstracts using 1-2gram analysis and multitask classification

The core of evidence-based medicine is to read and analyze numerous pape...
research
11/14/2022

Learning predictive checklists from continuous medical data

Checklists, while being only recently introduced in the medical domain, ...
research
05/25/2023

Patient Outcome Predictions Improve Operations at a Large Hospital Network

Problem definition: Access to accurate predictions of patients' outcomes...
research
07/25/2020

Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes with Guided Multi-Headed Attention

Problem lists are intended to provide clinicians with a relevant summary...
research
11/30/2021

What Do You See in this Patient? Behavioral Testing of Clinical NLP Models

Decision support systems based on clinical notes have the potential to i...

Please sign up or login with your details

Forgot password? Click here to reset