arXiv subjects

Brita Elvevåg

Publications and source records attributed to Brita Elvevåg.

3 recordsLinked to original sources

arXiv2026

From population norms to personalized trajectories: interpretable Bayesian forecasting for cognitive decline

The timing and rate of cognitive decline vary substantially between individuals, limiting the ability of fixed population-level thresholds to determine whether a new observation represents meaningful change for an individual. We developed Personalized Risk Inference via Sequential Monitoring (PRISM), an interpretable framework for individualized longitudinal forecasting of cognitive decline. PRISM estimates a personalized cognitive baseline from routinely collected demographic, health, and functional variables using an Explainable Boosting Machine, then updates this expectation through Bayesian inference with temporal decay as cognitive scores accrue. Decline is evaluated relative to an age-adjusted personal anchor, with uncertainty quantified through posterior probabilities. We evaluated PRISM in 30,664 adults from the Health and Retirement Study and externally validated it in 1,866 Alzheimer's Disease Neuroimaging Initiative participants. Forecasting performance was compared with demographic-norm and cumulative-average baselines, and discrimination with a linear mixed-effects model. PRISM identified emerging decline before study-defined cognitive worsening in 31% of sustained decliners in the Health and Retirement Study and 41% in the Alzheimer's Disease Neuroimaging Initiative, with median lead times of 6 and 2 years, respectively. By the time of worsening, 68% and 56% had been identified. PRISM also achieved lower forecasting error than demographic-norm and cumulative-average baselines and distinguished worsening from stable trajectories better than a linear mixed-effects model, particularly early in follow-up. PRISM enables earlier, interpretable, uncertainty-aware detection of cognitive decline relative to each individual's expected trajectory using routinely collected data. It may support closer monitoring and timely assessment when personal longitudinal history is limited.

cs.ET↗

arXiv2025

From research to clinic: Accelerating the translation of clinical decision support systems by making synthetic data interoperable

The translation of clinical decision support system (CDSS) tools from research settings into the clinic is often non-existent, partly because the focus tends to be on training machine learning models rather than tool development using the model for inference. To develop a CDSS tool that can be deployed in the clinical workflow, there is a need to integrate, validate, and test the tool on the Electronic Health Record (EHR) systems that store and manage patient data. Not surprisingly, it is rarely possible for researchers to get the necessary access to an EHR system due to legal restrictions pertaining to the protection of data privacy in patient records. We propose an architecture for using synthetic data in EHR systems to make CDSS tool development and testing much easier. In this study, the architecture is implemented in the SyntHIR system. SyntHIR has three noteworthy architectural features enabling (i) integration with synthetic data generators, (ii) data interoperability, and (iii) tool transportability. The translational value of this approach was evaluated through two primary steps. First, a working proof-of-concept of a machine learning-based CDSS tool was developed using data from patient registries in Norway. Second, the transportability of this CDSS tool was demonstrated by successfully deploying it in Norway's largest EHR system vendor (DIPS). These findings showcase the value of the SyntHIR architecture as a useful reference model to accelerate the translation of "bench to bedside" research of CDSS tools.

cs.LG↗

arXiv2024

Prompt Engineering a Schizophrenia Chatbot: Utilizing a Multi-Agent Approach for Enhanced Compliance with Prompt Instructions

Patients with schizophrenia often present with cognitive impairments that may hinder their ability to learn about their condition. These individuals could benefit greatly from education platforms that leverage the adaptability of Large Language Models (LLMs) such as GPT-4. While LLMs have the potential to make topical mental health information more accessible and engaging, their black-box nature raises concerns about ethics and safety. Prompting offers a way to produce semi-scripted chatbots with responses anchored in instructions and validated information, but prompt-engineered chatbots may drift from their intended identity as the conversation progresses. We propose a Critical Analysis Filter for achieving better control over chatbot behavior. In this system, a team of prompted LLM agents are prompt-engineered to critically analyze and refine the chatbot's response and deliver real-time feedback to the chatbot. To test this approach, we develop an informational schizophrenia chatbot and converse with it (with the filter deactivated) until it oversteps its scope. Once drift has been observed, AI-agents are used to automatically generate sample conversations in which the chatbot is being enticed to talk about out-of-bounds topics. We manually assign to each response a compliance score that quantifies the chatbot's compliance to its instructions; specifically the rules about accurately conveying sources and being transparent about limitations. Activating the Critical Analysis Filter resulted in an acceptable compliance score (>=2) in 67.0% of responses, compared to only 8.7% when the filter was deactivated. These results suggest that a self-reflection layer could enable LLMs to be used effectively and safely in mental health platforms, maintaining adaptability while reliably limiting their scope to appropriate use cases.

cs.CL↗