Aifred Health, a Deep Learning Powered Clinical Decision Support System for Mental Health

Abstract

INTRODUCTION:The pharmacological treatment of Major Depressive Disorder (MDD) continues to rely predominantly on a trial-and-error approach. Here, we introduce an artificial intelligence (AI) model aiming to personalize treatment and improve outcomes, which was deployed in the Artificial Intelligence in Depression -Medication Enhancement (AID-ME) Study.OBJECTIVES: 1) Develop a model capable of predicting probabilities of remission across multiple pharmacological treatments for adults with at least moderate major depression. 2) Validate model predictions and examine them for amplification of harmful biases. METHODS: Data from previous clinical trials of antidepressant medications were sourced from the NIMH, collaborating researchers, and pharmaceutical open science platforms, and standardized into a common framework. Our analysis included 9,042 adults with moderate to severe major depression from 22 studies, each lasting 8-14 weeks, and covering 10 different treatments. The data was divided into training, validation, and held-out test sets. Feature selection selected 25 clinical and demographic variables. Using Bayesian optimization, a deep learning model was trained on the training set and refined using the validation set before being tested once on the held-out test set. Pre-specified post-hoc testing was performed to assess for potential clinical utility as well as risk of the amplification of harmful biases and biased subgroup performance.RESULTS: In the evaluation on the held-out test set, the model demonstrated a performance with an AUC (Area Under the Curve) score of 0.65. The model outperformed a null model on the test set (p = 0.01). The model demonstrated notable clinical utility, achieving an absolute improvement in population remission rate from 43.15% to 53.99% under hypothetical "naive" analysis and an improvement from 43.21 to 55.08% improvement under actual improvement "conservative" analysis on the testing data. While the model did identify one drug (escitalopram)

Using a simulation centre to evaluate preliminary acceptability and impact of an artificial intelligence-powered clinical decision support system for depression treatment on the physician–patient interaction

“…Physicians (both psychiatrists and primary care physicians) are faced with a large selection of effective treatments, as well as guidelines which help to manage treatments once they are chosen, but they do not currently have access to tools that can help them effectively choose between the existing first-line agents to optimise chances of treatment success and minimise the need for repeated trial-and-error treatment trials. 10 This need for improved decision support has led to a number of projects aimed at improving the personalisation of treatment selection, notably pharmacogenomics. 11 However, pharmacogenomics may be expensive, and samples may take time to be processed, which could be used to treat the patient.…”

“…This artificial intelligence helps support clinicians by considering complex interactions between multiple patient variables to help personalise treatment in order to improve upon a trial-and-error treatment approach and reduce the number of failed treatment trials. 10 , 15 It also tracks symptoms by using standardised questionnaires such as the Patient Health Questionnaire-9. 16 Major depressive disorder (MDD) was chosen, given its high prevalence, 17 , 18 status as the leading cause of disability globally 19 and poor remission rates following initial treatment.…”

Using a Simulation Centre to Evaluate the Effect of anArtificial Intelligence-Powered Clinical Decision Support System for Depression Treatment on the Physician-Patient Interaction

“…We investigated the use of Aifred, a clinical decision support software (CDSS) which includes an operationalized version of the 2016 Canadian Network for Mood and Anxiety Treatments (CANMAT) guidelines for depression treatment [5] and provides AI decision support when treatments are chosen. This AI helps support clinicians by considering complex interactions between multiple patient variables to help personalize treatment in order to improve upon a trial-and-error treatment approach and reduce the number of failed treatment trials [6,7]. It also tracks symptoms using standardized questionnaires such as the PHQ-9 [8].…”

Evaluating the Clinical Feasibility of an Artificial Intelligence-Powered Clinical Decision Support System: A Longitudinal Feasibility Study

“…Furthermore, treatments are essentially equally effective at the population level, when to improve outcomes treatment selection must address the individual's specific characteristics (Cipriani et al, 2018;. As such, there is a clear need for improved and personalized decision support for mental healthcare (see Benrimoh et al, 2018 for further discussion).…”