Research Projects

The RainFrog ecosystem offers an array of evidence-based online therapy modules and smartphone tools designed to prevent mental health problems. Therapy is tailored by selecting specific technique-based modules and adapting content to individual needs. Special thanks to the UCLA Depression Grand Challenge (DGC) for their support in developing and implementing this project.

This study leverages passive data from Apple Watches and iPhones to develop person-specific digital phenotypes for common mental health disorders (CMDs). The goal is to create scalable and personalized assessment and treatment methods for a diverse community college population. We are currently recruiting participants from the STAND program at East Los Angeles College (ELAC). Our second-year graduate student has already presented promising findings at the International Society for Research on Internet Interventions (ISRII) and the Society for Digital Mental Health (SDMH) conferences.

The OPTIMA-ILIAD (Operationalizing Digital PhenoTyping in the Measurement of Anhedonia – Investigating Low-Intensity Focused Ultrasound Pulsation in Anhedonic Depression) project aimed to measure and track behavioral features of anhedonia using digital phenotyping and neuroimaging. Enrolling up to 600 adults, the study collected longitudinal behavioral health data, self-report assessments, and ecological momentary assessments (EMAs) to predict and assess changes in anhedonia. The ILIAD component focused on a randomized, sham-controlled trial using low-intensity focused ultrasound pulsation (LIFUP) to treat anhedonic depression. By combining behavioral, neuroimaging, and neuromodulation data, the OPTIMA-ILIAD project aimed to improve the understanding and treatment of anhedonia, offering insights into the potential of LIFUP as a therapeutic intervention.

The Youth Mindful Awareness Program address the need for accessible, effective treatment of internalizing disorders in youth through an online MBI. This randomized controlled trial involves 360 youths aged 12-17, evaluating participants at baseline, post-intervention, and follow-up at 6 months. Apple Watches will collect objective data alongside ecological momentary assessments (EMAs). The goal is to prevent mental health issues by reducing negative affectivity, examining intervention outcomes, and exploring physiological stress reactivity. YMAP is a multisite project with participants recruited by Vanderbilt, Northwestern, and UCLA. The University of Arizona is responsible for EMA build through the MyDataHelps App and incorporating digital phenotyping.

The Personalized Advantage Index (PAI) was a pioneering project aimed at improving mental health outcomes by matching individuals to their optimal treatments. Developed to address the limitations of single-variable moderation, the PAI approach integrated multiple predictive factors to enhance treatment selection. The project involved a team of international collaborators who used data from randomized controlled trials to develop and illustrate methods of treatment matching. The PAI approach significantly impacted the field of depression treatment by providing clinicians with a robust tool to personalize therapy, ultimately leading to better patient outcomes.

Our predictive modeling project utilized advanced machine learning techniques to identify prognostic factors and improve treatment outcomes in psychiatry. One of the most notable initiatives was the SMART (Stratified Medicine Approaches foR Treatment Selection) Mental Health Prediction Tournament. In this project, teams of experts from around the world were given the same training dataset to predict psychotherapy outcomes. Using a large naturalistic dataset from the UK’s NHS IAPT program, the project overcame sample size limitations, demonstrating that some modeling approaches could generate robust, generalizable prognostic models. These models significantly outperformed simpler methods, showing great promise in personalizing mental health treatments. This project laid the groundwork for future predictive modeling efforts in mental health care.

StratCare led by Jaime Delgadillo, Ph.D. was a project focused on personalized treatment approaches for mental health. The project emphasized the importance of matching patients to optimal therapies based on individual characteristics, such as symptom profiles and treatment preferences. By utilizing stratified care models, StratCare aimed to improve treatment outcomes and efficiency. This approach was validated through various studies and clinical trials, highlighting its potential to enhance the precision of mental health care. StratCare’s findings were published in leading journals and presented at international conferences, contributing significantly to the field of personalized mental health treatment and informing best practices for future interventions.