Seven projects funded by 2021 Precision Health Investigators Awards

December 8, 2021  //  FOUND IN: News

The Precision Health Investigators Awards support new, collaborative research projects rooted in precision health to advance science and develop health innovations. This awards program aspires to nurture groundbreaking multidisciplinary research projects that advance the field of precision health through the use and/or enrichment of Precision Health data, tools, methods and techniques.

The awards for 2021 recognize seven thoroughly collaborative interdisciplinary research projects that explore topics such as surgery-associated kidney injury, point-of-care diagnosis for glioma genetic subtypes, and two projects focused on data integration: one involving electronic health records, and the other aimed at pan-cancer analysis.

Below, the grant recipients describe the aims, impacts and reasons for their research. You can also read the project abstracts.

“Using Artificial Intelligence to Broaden and Diversify Outdated Standards for the Determination of Skeletal Maturation in Growing Children”
Anouck Girard, Ph.D., associate professor of aerospace engineering; Josephine Kasa-Vubu, M.D., clinical professor of pediatrics; Michael DiPietro, M.D., professor emeritus of radiology

“Blending drone technology, traditional radiological imaging, and current-day pediatric endocrinology practice, this research challenges century-old dogmas about skeletal maturation in children. Interdisciplinary approaches have been a long tradition at Michigan, and few places would fathom the feasibility of associating aerospace engineering and pediatric medicine for groundbreaking research… only at Michigan!” —Josephine Kasa-Vubu

“It’s amazing to see how technology that was originally studied for military and aerospace applications is helping make precision health a reality, leading us to develop more inclusive and diverse medical standards, and improving the lives of children.” — Anouck Girard

“Rapid Intraoperative Molecular Diagnosis of Diffuse Gliomas Using Stimulated Raman Histology and Deep Neural Networks”
Todd Hollon, M.D., assistant professor of neurosurgery; Honglak Lee, Ph.D., associate professor of computer science; Sandra Camelo-Piragua, M.D., associate professor of pathology

“Molecular classification has transformed the diagnosis and treatment of diffuse gliomas, creating targets for precision therapies. However, timely and efficient access to molecular diagnostic methods remains difficult, causing a significant barrier to delivering molecularly targeted treatments. Our laboratory aims to develop an innovative point-of-care diagnostic screening method that provides rapid and accurate molecular classification of diffuse gliomas through artificial intelligence and optical imaging, to improve the comprehensive care of brain tumor patients.” —Todd Hollon

“Statistical and Computational Methods for Asymmetric Integration of Datasets from Different Cancers for the Identification of Cancer-related Genes and Biomarkers in Case-control Analyses”
Hui Jiang, Ph.D., associate professor of biostatistics
Collaborators: J. Chad Brenner, Ph.D., associate professor of otolaryngology-head and neck surgery; Kevin (Zhi) He, Ph.D., research associate professor of biostatistics

“Pan-cancer analysis has the potential to identify common driver genes and biomarkers with greater statistical power and accuracy by taking advantage of the increased sample size when integrating datasets from different cancers. In this project, we will develop novel statistical and computational methods, as well as software tools, to analyze data collected from cancer patients and matched controls in the Michigan Genomics Initiative (MGI). Based on this dataset, we aim to find germline variants and related genes that are associated with increased risk of cancer, which can be further utilized for predictive modeling to facilitate scientific investigations and clinical uses. This research relies on an interdisciplinary collaboration involving expertise in both biostatistics and cancer genetics.” —Hui Jiang

“Predicting Cardiac Surgery-Associated Acute Kidney Injury Using Federated Learning”
Michael Mathis, M.D., assistant professor of anesthesiology
Collaborators: Karandeep Singh, M.D., M.M.Sc., assistant professor of learning health sciences, internal medicine, urology and information; Donald Likosky, Ph.D., professor of cardiac surgery; Rahul Ladhania, Ph.D., assistant professor of health informatics and biostatistics; Paramveer Dhillon, Ph.D., assistant professor of information

“Kidney injury after cardiac surgery is a serious complication that could be better understood if hospitals could combine their data on such patients. However, combining multi-hospital data runs counter to patients’ privacy expectations. In our study, our team of clinicians and methodologists – brought together by Precision Health– uniquely overcomes this challenge, through the use of federated learning. Developed for mobile technologies, federated learning enables the development of multi-hospital prediction models without the need to share data. Just as mobile apps such as Waze can crowdsource information on traffic conditions without sharing sensitive individual-level cellphone data, our approach to understanding kidney injury after cardiac surgery crowdsources information from multiple hospitals without compromising patient-level protected health information.” —Michael Mathis

“Assessing the Impact of Germline Pharmacogenetics (PGx) on Medication Outcomes and Clinician Prescribing Decisions in Patients with Cancer”
Amy Pasternak, Pharm.D., clinical assistant professor of pharmacy; Vaibhav Sahai, MBBS, M.S., associate professor of medical oncology and hematology
Collaborators: Daniel Hertz, Pharm.D., Ph.D., assistant professor of pharmacy; Valerie Gunchick, M.S., clinical research project manager

“Many patients with cancer experience side effects to their cancer treatment. With this award, we will investigate how pharmacogenetics could help to decrease the risk of a patient experiencing these side effects for different chemotherapies and for medications intended to help manage chemotherapy side effects.” —Amy Pasternak

“Deep Learning for Prediction of Mild Cognitive Impairment and Dementia of the Alzheimer’s Type”
Scott Peltier, Ph.D., research scientist of biomedical engineering/functional MRI laboratory; Zhongming Liu, Ph.D., associate professor of biomedical engineering and electrical engineering and computer science
Collaborators: Benjamin Hampstead, Ph.D., ABPP/CN, professor of psychology and psychiatry; Jeffrey Fessler, Ph.D., professor of electrical engineering and computer science; Douglas Noll, Ph.D., professor of biomedical engineering

“This project will use functional MRI and machine learning methods to advance individualized diagnosis and treatment of Alzheimer’s disease.

The unique collaborative environment at the University of Michigan, including the Functional MRI Laboratory, Departments of Biomedical and Electrical Engineering, and the Michigan Alzheimer’s Disease Research Center, makes this research possible.” —Co-PIs Scott Peltier and Zhongming Liu

“Automated Harmonization of Multi-institutional Electronic Health Records Data”
Xu Shi
, Ph.D., assistant professor of biostatistics
Collaborator: V.G. Vinod Vydiswaran, Ph.D., associate professor of learning health sciences and information

“Despite federal initiatives incentivizing electronic health records (EHR) data harmonization across healthcare institutions, it is notorious that EHRs do not talk to each other. Such a lack of interoperability can decrease a model’s performance and lead to biases in biomedical research. Our team will adopt principles in how humans talk to each other to address the inherent heterogeneity in multi-institutional EHR data and implement the proposed methods to generate and transfer knowledge between the Michigan Genomics Initiative and UK Biobank.” —Xu Shi

RELATED STORIES