# Novel imaging genetic biomarkers for sporadic frontotemporal dementia through machine learning

> **NIH NIH K01** · UNIVERSITY OF CALIFORNIA, SAN DIEGO · 2021 · $120,420

## Abstract

Project Summary
This is an application for a K01 award for Dr. Iris Broce-Diaz, a neuroimaging genetics postdoctoral fellow at the
University of California, San Diego and University of California, San Francisco. Dr. Broce-Diaz is establishing
herself as a young imaging geneticist conducting clinical research on neurodegenerative disease. This K01 will
provide Dr. Broce-Diaz with the support necessary to accomplish the following goals: (1) gain proficiency in
machine learning and computational modeling techniques, (2) gain proficiency in clinical and genetic research
methodology for cognitive aging and complex spectrum of neurodegenerative diseases, including clinical
characterization of frontotemporal dementia (FTD) and other Alzheimer’s Disease-Related Dementias,
differential diagnosis, risk prediction, and biomarker development, and (3) develop an independent research
career. To achieve these goals, Dr. Broce-Diaz has assembled an expert mentoring team, including her primary
mentors: Dr Anders Dale (renowned computational neuroimaging genetics scientist) and co-primary mentor
Bruce Miller (internationally recognized behavioral neurologist and leader in FTD), co-mentors: Drs. Jennifer
Yokoyama (expert in FTD genetics) and Chun Chieh Fan (expert in epidemiology/biostatistics), and two
collaborators: Drs. Adam Boxer (leader in clinical trials for FTD-spectrum disorders) and Wesley Thompson
(expert in advanced statistics).
The goal of the proposed project is to develop novel imaging genetics biomarkers for predicting individuals at
risk of developing sporadic (non-familial) FTD and improving classification accuracy of sporadic FTD. Dr. Broce-
Diaz will achieve this goal through the following specific aims: (1a) utilize a polygenic hazard approach to develop
and validate a novel genetic biomarker for predicting age-specific risk of sporadic FTD; (1b) leverage pleiotropic
information to increase accuracy of the genetic risk scores and derive biologically-based genetic risk scores; (2)
use machine learning approaches to reliably and accurately classify FTD clinical subtypes and obtain
personalized atrophy scores from these brain maps; and (3) improve FTD classification by integrating atrophy
scores with genetic risk scores. This proposed study uses highly innovative methodological approaches for
informing FTD prognosis, diagnosis, and, ultimately, clinical trial design. If validated, these biomarkers will make
significant contributions by assisting clinicians in identifying patients at elevated risk for sporadic FTD and
assisting in diagnosing sporadic FTD in its earliest stages—reducing diagnostic delays, accelerating the
discovery of novel treatments, and improving recruitment accuracy in clinical trials. This K01 research project
will provide Dr. Broce-Diaz with the protected research time and opportunity to train with leaders in the field she
needs to master the skills required to establish an independent, patient-oriented, imaging genetics and bi...

## Key facts

- **NIH application ID:** 10106384
- **Project number:** 1K01AG070376-01
- **Recipient organization:** UNIVERSITY OF CALIFORNIA, SAN DIEGO
- **Principal Investigator:** Iris J Broce-Diaz
- **Activity code:** K01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $120,420
- **Award type:** 1
- **Project period:** 2021-03-15 → 2026-02-28

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10106384

## Citation

> US National Institutes of Health, RePORTER application 10106384, Novel imaging genetic biomarkers for sporadic frontotemporal dementia through machine learning (1K01AG070376-01). Retrieved via AI Analytics 2026-05-21 from https://api.ai-analytics.org/grant/nih/10106384. Licensed CC0.

---

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