Relevance-Based Prediction: A Transparent, Patient-SpecificApproach to Predicting Medical Outcomes
Robinson, C. L.; Turkington, D.; Lee, L.; Kritzman, M.; Yong, R. J.
Show abstract
Accurate prediction of individual medical outcomes is essential for optimizing treatment allocation amid rising costs, coverage denials, and limited clinical resources. Traditional predictive models, including linear regression and neural networks, rely on average effects and cannot tailor predictions to the specific circumstances of individual cases. We present relevance-based prediction (RBP), a model-free and task-specific method that predicts outcomes as weighted averages of observed cases, with weights determined by a rigorously defined and theoretically justified measure of relevance. Unlike model-based methods that rely on fixed calibrated parameters, relevance-based prediction revisits the original data for each prediction and customizes both the cases and variables used. We illustrate RBP by applying it to predict opioid treatment outcomes and demonstrate that it provides case-specific insights unavailable from conventional models, including how each prior case informs a prediction, how each variable affects each predictions reliability and value, and how reliable each prediction is before it is made. These individualized insights may prevent misleading average-based decisions and reduce harmful or suboptimal treatment.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Continuous-Time and Dynamic Suicide Attempt Risk Prediction with Neural Ordinary Differential Equations 94%
- Enhancing Privacy-Preserving Deployable Large Language Models for Perioperative Complication Detection: A Targeted Strategy with LoRA Fine-tuning 92%
- Predicting critical state after COVID-19 diagnosis: Model development using a large US electronic health record dataset 92%
Similar papers in this journal
- Predicting diabetes second-line therapy initiation in the Australian population via timespan-guided neural attention network 92%
- A scaling approach to estimate the COVID-19 infection fatality ratio from incomplete data 91%
- Significant Relaxation of SARS-CoV-2-Targeted Non-Pharmaceutical Interventions Will Result in Profound Mortality: A New York State Modelling Study 91%
Similar papers in this journal
- Interpretable deep recommender system model for prediction of kinase inhibitor efficacy across cancer cell lines 92%
- Leveraging Temporal Learning with Dynamic Range (TLDR) for Enhanced Prediction of Outcomes in Recurrent Exposure and Treatment Settings in Electronic Health Records 92%
- Large Language Models Improve the Identification of Emergency Department Visits for Symptomatic Kidney Stones 91%
Similar papers in this journal
- Model-informed Deep Q-Networks to Guide Infliximab Dosing in Pediatric Crohn's Disease 90%
- Algorithmic identification of treatment-emergent adverse events from clinical notes using large language models: a pilot study in inflammatory bowel disease 89%
- Phenotypic models of drug-drug-gene interactions mediated by cytochrome drug-metabolizing enzymes 87%
Similar papers in this journal
- Optimizing the Implementation of Clinical Predictive Models to Minimize National Costs: A Sepsis Case Study 93%
- Optimal policy determination in sequential systemic and locoregional therapy of oropharyngeal squamous carcinomas: A patient-physician digital twin dyad with deep Q-learning for treatment selection 93%
- Dynamics and Development of the COVID-19 Epidemics in the US: a Compartmental Model with Deep Learning Enhancement 90%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.