Department of Biomedical Engineering

Yu Gan, an assistant professor in the Department of Biomedical Engineering, recently received a $600,000 National Science Foundation CAREER Award for his project, “Developing Algorithms for Object-Adaptive Super-Resolution in Biomedical Imaging.” The five-year project seeks to develop intelligent, computationally efficient algorithms to improve the clarity and quality of diagnostic biomedical images in a cost-effective and generalizable manner.

Advanced biomedical imaging technology has revolutionized medical diagnosis and treatment, but the amount of detail, or resolution, that can be captured in biomedical images can be insufficient for specific applications. Conventional software-based solutions for developing super-resolution images — which combine multiple low-resolution images to create one high-resolution image — are slow, expensive and inefficient.

Gan’s project seeks to advance national health by developing an optimized artificial intelligence framework that adaptively improves digital imaging resolution while minimizing the cost of computation in the super-resolution process. He will develop a robust neural network to detect regions of biomedical images to be super-resolved and computationally efficient algorithms to super-resolve those images when enlarged to different scales.

The framework will be generalized for use across multiple biological imaging categories, including optical coherence tomography (OCT), microscopic histology images, confocal images, magnetic resonance imaging (MRI) and ultrasound images for different applications.

Resources from Gan’s research will help develop cost-effective biomedical imaging methods that can be more widely distributed than current specialized biomedical imaging facilities allow, improving access to advanced diagnostic methods and medical care for underrepresented groups for whom such access is financially or geographically limited.

Professor and Chair of the Department of Biomedical Engineering Jennifer Kang-Mieler has been elected to be inducted into the 2023 Class of the American Institute for Medical and Biological Engineering College of Fellows.

Election to the AIMBE College of Fellows is among the highest professional distinctions accorded to a medical and biological engineer in the world.

Kang-Mieler was elected by peers and members of the College of Fellows “for innovation in ocular drug delivery, outstanding professional service and contributions to biomedical engineering education.” She was inducted among 140 colleagues into the Class of 2023.

Kang-Mieler's translational NIH-supported research projects include ocular drug delivery, retinal imaging and biomarkers, retinal blood flow and electrophysiology. Her clinical interests include retinal vascular diseases, such as age-related macular degeneration and diabetic retinopathy.

The College of Fellows, which is comprised of the top 2% of medical and biological engineers globally, honors professionals who have made outstanding contributions to and pioneering advances in engineering and medicine research, practice and education.

AIMBE Fellows are among the most distinguished medical and biological engineers in the world and include three Nobel Prize laureates and 17 recipients of the Presidential Medal of Science and/or Technology and Innovation.

Raviraj Nataraj, an assistant professor in the Department of Biomedical Engineering, recently received a $622,287 National Science Foundation CAREER Award for his project “Personalizing Sensory-Driven Computerized Interfaces to Optimize Motor Rehabilitation.” The five-year award is one of the most prestigious and competitive awards available to early-career faculty.

Although highly realistic or game-like virtual reality (VR) technologies are increasingly used to motivate participation in motor therapy by persons with brain or spinal cord injuries, VR motor therapy does not always achieve better outcomes than traditional methods, partially because the therapy and sensory feedback is not personalized to each user.

Nataraj’s research seeks to develop improved methods of VR motor therapy that are more personalized to, and therefore more effective for, the individual.

Focusing on persons with spinal cord injury specifically, this project will examine how impairment in function affects personal performance, physiologic responses and perceptions during VR training that personalizes task training complexity and augmented sensory feedback (such as visual or haptic cues) that provides movement direction and magnitude guidance. It will test the hypothesis that motor performance improves when VR training features are adapted based on individual measures indicating well-being and physical readiness for training.

The project will also train high-school students to develop, present and test custom VR rehabilitation applications.

Jennifer Kang-Mieler recently received a 2022 National Institutes of Health (NIH) Grant for $504,051 for her project, "Dynamic Tracer Kinetic Model to Detect Preclinical Diabetic Retinopathy (DR)." Early detection and treatment of DR can prevent more than 90% of vision loss, however, finding an appropriate tool or technology to detect preclinical signs (biomarkers) of DR is a challenge.

Since the retinal vessels are early and prevalent targets of diabetic damage, hold great potential as biomarkers. Recent advances in retinal imaging technology such as Adaptive Optics Scanning Laser Ophthalmoscopy (AOSLO) and OCT angiography (OCTA) have allowed a better visualization of sensitive identifiers of changes in structural and functional blood vessel characteristics. These types of non-invasive tools have limitations (e.g., long scan times, limited field-of-view, motion artifacts and need for an expert operator) that prevent these technologies from being effective preclinical detection tools in a clinical setting.

To address this, Kang-Mieler has developed a novel dynamic tracer kinetic model to measure quantitatively vascular permeability and blood flow changes based on fluorescein video-angiography (FVA). The approach is immediately translatable to FVA data collected in patients as demonstrated by her preliminary data.

In this project, she will demonstrate that her dynamic tracer kinetic model can detect preclinical DR with a higher sensitivity and specificity than other retinal imaging modalities such as OCTA and AOSLO.