2016 Galvanizing Engineering in Medicine (GEM) Awards

The Clinical and Translational Research Institute (CTRI) announces the selection of four physician-engineer teams as the 2016 recipients of the Galvanizing Engineering in Medicine (GEM) awards. GEM, an initiative of the CTRI and UC San Diego Institute of Engineering in Medicine (IEM), supports projects that identify clinical challenges for which engineering solutions can be developed and implemented to improve health care. The GEM recipients and their projects are below.

Bhaskar Rao, PhD Sebastian Obrzut, MD Sebastian Obrzut, MD
Bhaskar Rao, PhD
Jacobs School of Engineering
UC San Diego
Sebastian Obrzut, MD
Department of Radiology
UC San Diego
Bongyong Song, PhD
Department of Radiation Medicine and Applied Sciences
UC San Diego

Title: Compressed Sensing Image Acquisition and Processing for Single Photon Emission Tomograph

In Single Photon Emission Tomography (SPECT) clinical images are reconstructed from acquired projection images of radioactivity distribution onto detectors at various angles. Due to inherent inefficiency of gamma cameras, SPECT images are corrupted by noise and SPECT acquisition requires either long imaging time or greater injected radiopharmaceutical dose. This project aims to evaluate Compressed Sensing (CS) image acquisition and a processing algorithm for SPECT to significantly reduce image acquisition time.

By exploiting spatial locality of the radiotracer as prior information about the SPECT image to be reconstructed, the image can be transformed into a sparse image, allowing for utilization of a CS framework for reconstructing sparse signals (signals with many zeros). Our CS framework will utilize sparse angular sampling and/or shortened measurement time per projection angle, thereby enabling a greatly reduced SPECT scanning time. Furthermore, our framework will take advantage of the 3-D spatial locality of radioisotope distribution and will jointly reconstruct a 3-D image with a 3-D prior in order to enhance the image reconstruction quality. Although this joint reconstruction requires iteratively solving a large scale non-linear optimization problem, this involves a very large number of similar operations that can be greatly accelerated using parallel computing (e.g. GPU). Since any useful 3-D specific properties have a potential Bayesian interpretation, this project will investigate a new framework for developing state-of-the-art effective algorithms that incorporates a class of priors promoting specific image properties for different clinical applications.

The new framework will be evaluated using both the real and virtual Jaszczak phantoms and clinical SPECT bone images. The new framework will greatly reduce the image acquisition time. Specifically, this project aims to: reduce clinical brain SPECT imaging from 30 minutes to less than 15 minutes; reduce clinical chest, abdomen or pelvis SPECT imaging from 30 minutes to less than 15 minutes, respectively; and reduce small animal whole body SPECT imaging from 45 minutes to less than 20 minutes. These will not only increase patient comfort but also contribute to a potential reduction of patient motion artifacts. Furthermore, the increase in the patient throughput will reduce the overall clinic operating cost.

Frank Talke, PhD Thomas Savides, MD
Frank Talke, PhD
Jacobs School of Engineering
UC San Diego
Thomas Savides, MD
Chief Experience Officer, UC San Diego Health
Division of Gastroenterology
UC San Diego

Title: Disposable, 3-D Printable Endoscope

Endoscopes are instruments used to view an inner part of the body such as a colon. Difficulty with disinfecting a type of endoscope called a duodenoscope has been associated with contamination issues and subsequent bacterial infections. Duodenoscopes are flexible lighted tubes used to diagnose and treat pancreatic and bile duct diseases. This project focuses on the design and manufacture of a 3-D printable and disposable endoscope that is biocompatible and programmable. Shape Memory Alloy (SMA) wires actuate the newly designed endoscope, which can be guided through an “obstacle course” resembling human internal organs. The instrument is manufactured using 3-D printing, which is faster and less costly than other manufacturing methods presently used, and its dimensions can be easily modified to suit a patient’s needs. The team has implemented two different devices so far: a flexible endoscope and a disposable overtube, which is a cylindrical device that encloses an endoscope and guides it.

Karen Christman, PhD   Marianna Alperin, MD, MAS
Karen Christman, PhD
Department of Bioengineering
UC San Diego
Marianna Alperin, MD, MAS
Department of Reproductive Medicine
UC San Diego

Title: Injectable Biomaterial for Treating Pelvic Floor Disorders

Pelvic floor disorders (PFD), which include urinary and fecal incontinence, and pelvic organ prolapse, are debilitating and costly conditions that affect a quarter of the U.S. female population to date. By 2050, a continual dramatic increase in PFD prevalence is predicted to result in 43.8 million women suffering from pelvic floor dysfunction. Maternal childbirth trauma and consequent dysfunction of urethral sphincter, external anal sphincter and pelvic floor skeletal muscles, is a leading risk factor for urinary incontinence, fecal incontinence, and pelvic organ prolapse, respectively. Currently, there are no preventive measures, beyond Cesarean section, and the existing treatments do not address the underlying pathophysiology. Furthermore, the available therapies are associated with significant morbidities, while offering marginal promise at best. The proposed engineering solution is to develop an injectable biomaterial scaffold and a minimally invasive delivery system to potentiate the acute recovery of injured urethral sphincter, external anal sphincter, and pelvic floor skeletal muscles. Upon injection, the material will set up into a porous and fibrous scaffold that will facilitate endogenous cell infiltration to regenerate and heal the damaged muscles post-vaginal delivery thereby preventing the development of PFD.

Sheng Xu, PhD Robert Owens, MD
Sheng Xu, PhD
Jacobs School of Engineering
UC San Diego
Robert Owens, MD
Division of Pulmonary, Critical Care and Sleep Medicine
UC San Diego

Title: Battery-free Wireless Wearable Sensors for Sleep Monitoring

Many patients, and most physicians, think that a “good night’s sleep” is an essential part of recovery during and after medical and psychiatric illness. Adequate sleep can promote attention, vigilance and a sense of well-being. Conversely, sleep deprivation can lead to decreased neurocognitive performance and alertness, and disorganized thinking. Although the science is still emerging, sleep appears to be crucial for maintaining homeostasis and immune function, with decreased sleep associated with insulin resistance, risk of myocardial infarction, and increased infection risk. Despite these known or suspected benefits, sleep is not prioritized in most intensive care units (ICUs). As a result, the most critically ill patients are denied sufficient sleep, with a total sleep time of approximately five hours per night, and a median sleep time of just three minutes.

Sleep can be characterized by measuring local field potentials on the body, such as electrocardiogram, electrooculogram, and electroencephalogram that capture heart rate, eye movement, and brain activity, respectively. Conventional tools for collecting local field potentials typically involve bulky electronics with bundles of connection wires. This cumbersome and uncomfortable set up creates an environment that is deviating from the natural sleep condition of the subject, which may bias the data collected. Furthermore, in the ICU environment, maintaining the recording equipment may not be feasible, or may hinder best clinical management practices.

The objective of this project is to design and fabricate a low profile minimally invasive wearable patch for sleep monitoring. The approach is to hybridize customized soft membrane sensors and rigid commercial-off-the-shelf chips that will be connected by wavy metal wires on a stretchable substrate. The overall system will be rigid locally but compliant globally, with elastic properties similar to our skin. The on-board coil antenna will serve for dual functions: sending the acquired data wirelessly to the external backend receiver, and harvesting power from the reading coil by resonant inductive coupling. This wearable patch will enable checking the sleep quality of the subject in an unobtrusive manner, and thus provide critical capabilities in collecting unbiased data for sleep studies.