# Integrating brain, neurocognitive, and computational tools in Opioid Use Disorder (OUD) to characterize executive function and to predict clinical outcomes

> **NIH NIH K01** · UNIVERSITY OF PENNSYLVANIA · 2024 · $175,534

## Abstract

Project Summary. The 5-year K01 Mentored Research Scientist proposal will employ brain, neurocognitive,
and computational tools (e.g., deep learning) to understand the impact of opioid-use disorder (OUD) and
common co-occurring issues on executive function and clinical outcomes. There have been record numbers of
fatal and non-fatal overdoses (ODs) associated with opioids (and other drugs) in the past 12-months. Improving
classification and predictive capabilities to enhance treatment and prevent relapse is of the upmost importance.
Deficits in neurocognition often are associated with poor treatment outcomes (e.g., more drug use, medication
non-adherence), yet co-occurring issues associated with OUD (e.g., depression, anxiety, physical/sexual abuse,
neglect) make it difficult to parse which contributing factors lead to worse executive function (EF) and poorer
treatment outcomes. Novel brain, neurocognitive, and computational tools are needed to help determine these
differences, in order to lay the foundation for better treatments. This need has shaped both the training plan and
the associated research project in a 5-year K01 Mentored Research Scientist proposal, building on Dr. Regier's
prior preclinical and clinical addiction neuroscience experience (focused mostly on cocaine-use disorders, cue-
reactivity, subcortical networks, prior adversity, and univariate imaging (fMRI) techniques). Mentor Dr. Childress
will guide career development, and will coordinate training and individualized mentoring from a group of top-tier
experts centered around 4 areas: Training Aim 1) opioid use disorder (OUD), its treatments, and comorbidities
(Dr. Kampman, mentor), Training Aim 2) neurocognition (Dr. Gur, mentor), the impact of mental health, and its
relationship to clinical outcomes, Training Aim 3) functional near-infrared spectroscopy (fNIRS), a mobile, non-
invasive cortical brain imaging technology (Dr. Ayaz, Mentor), and Training Aim 4) advanced computational
techniques (deep learning; Drs. Ayaz and Curtin) in outcome prediction. The training aims will be enabled by the
Research Project Aims. Research Aim 1 (Conventional Approach): Examine differences between OUD vs HC
on EF scores and PFC activity during EF tasks (Aim 1a); Using step-wise regression, examine relationship of
brain (PFC) data and/or co-occurring variables with EF (Aim 1b) and clinical outcomes (Aim 1c). Research Aim
2 (Deep Learning): Examine whether multi-task, spatiotemporal brain data can distinguish OUD from HCs (Aim
2a). Within the OUD population, examine whether multi-task, spatiotemporal brain data can classify better or
worse EF (Aim 2b) and/or drug-use outcome groups (Aim 2c). Exploratory: Add co-occurring variables into the
deep learning pipeline to determine whether they improve classification of either EF and/or drug-use outcomes.
The proposed K01 will facilitate Dr. Regier's transition to an independent research career focused on brain-
behavioral vulnerabilities in relapse an...

## Key facts

- **NIH application ID:** 10781898
- **Project number:** 5K01DA056700-02
- **Recipient organization:** UNIVERSITY OF PENNSYLVANIA
- **Principal Investigator:** Paul Regier
- **Activity code:** K01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2024
- **Award amount:** $175,534
- **Award type:** 5
- **Project period:** 2023-03-01 → 2028-02-29

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10781898

## Citation

> US National Institutes of Health, RePORTER application 10781898, Integrating brain, neurocognitive, and computational tools in Opioid Use Disorder (OUD) to characterize executive function and to predict clinical outcomes (5K01DA056700-02). Retrieved via AI Analytics 2026-06-12 from https://api.ai-analytics.org/grant/nih/10781898. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*