Autism spectrum disorder (ASD) is a group of complex neurodevelopment disorders affecting nearly 1.5 percent of American children. The underlying and contributing mechanisms of ASD are largely undetermined and there are no FDA-approved medications to treat the core social-communicative symptoms of the disorder. Unlike other brain disorders such as Alzheimer?s disease (AD), there is a gross disparity between the high prevalence of ASD and the small number of PET research studies; this condition should be addressed since the opportunities for PET imaging to have an impact in ASD research are substantial. Synapses allow for the transmission of electrochemical information, which is the essential means of communication in the brain. Synaptic pathology has been associated with brain disorders including AD, epilepsy, autism, depression, and schizophrenia. Our recent development of 11C-UCB-J, a PET radioligand that binds to synaptic vesicle glycoprotein 2A (SV2A), provides the ability to quantify synaptic density in the living human brain. 11C-UCB-J for SV2A PET imaging has the potential to make game-changing contributions to ASD research and potential diagnosis. However, because of the current imaging dose, the technique cannot be applied regularly in autism research of children due to the requirement of low radiation exposure especially for longitudinal studies in the same individuals at different developmental stages. The goal of this proposal is to reduce the clinical 11C-UCB-J PET imaging dose to be lower than the radiation one receives from a round-trip cross-country flight. To accomplish this goal, we will develop artificial neural network-based dose-reduction techniques. We hypothesize that 1) the proposed techniques will enhance (parametric) images from 1/10 of the dose so that they are comparable to the full-dose (parametric) images in terms of noise and resolution and 2) with the reduced dose, the test-retest reproducibility and the ability to differentiate patients from controls will not be diminished. The proposed research will be the first study to demonstrate that PET imaging dose can be reduced to a point of acceptability for ASD research in children and other diseases of adolescence. This effort is in line with the NIBIB I/START program to develop clinical technologies for imaging development in children and synergizes with the NIH supported large scale ABCD (Adolescent Brain Cognitive Development) longitudinal study for brain development in children, where MRI has been widely used but PET has not. If successful, we will apply for a larger grant to use SV2A PET to characterize the mechanisms of ASD neuropathology toward more effective diagnosis and personalized treatment.

Public Health Relevance

A newly developed positron emission tomography (PET) imaging tracer can be used to measure synaptic density in the living human brain. Synaptic pathology has been associated with many brain disorders including autism spectrum disorder (ASD), which affects 1.5% of American children. The project will use machine learning techniques to explore whether the radiation dose of synaptic PET imaging can be effectively lowered to roughly that of a round-trip cross-country airline flight for its regular utilization in ASD research in children and adolescents.

National Institute of Health (NIH)
National Institute of Biomedical Imaging and Bioengineering (NIBIB)
Small Research Grants (R03)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Duan, Qi
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Oakland University
Engineering (All Types)
Biomed Engr/Col Engr/Engr Sta
United States
Zip Code