Software & Machine Intelligence | ECE | Virginia Tech


Software & Machine Intelligence

ECE researchers seek to improve diagnosis of muscular dystrophies

The Computational Bioinformatics and BioImaging Laboratory (CBIL) is working with biomedical researchers to improve the diagnosis of muscular dystrophies (MD). Muscular dystrophies are a group of rare genetic diseases that cause progressive weakness and degeneration of the muscles used for voluntary movement.

There are nine different identified types of MD, each with varying degrees of progression and disability. Some forms of MD are mild and progress slowly throughout a normal lifespan, while others progress quickly and can lead to death in the third decade of life. There is no treatment for the diseases, other than physical and respiratory therapy and sometimes surgery to improve quality of life.

Computer renders of tissue engineering scaffold and skeletal, fetal mouse

Left: A visualization of the structure of a tissue engineering scaffold derived from a microCT image volume. Microcomputed tomography images of poly(L-lactic acid scaffolds for tissue engineering applications. The scaffolds were prepared by Michaela Schultz at the University of Graz, Austria and the image rendering done by the Bioimaging Systems Laboratory.

Right: A research project in the BioImaging Systems Laboratory is aimed at developing an open-source package that quickly, automatically, and consistently analyzes microCT images of fetal mice. The software will ultimately be able to count ribs, estimate rib fusion, or measure total bone volume or density, measure the skull, arms or limbs

MD is typically diagnosed and classified with blood tests, testing the electrical activity of the muscle (electromyography), ultrasound, muscle biopsy, and genetic testing.

The ECE researchers have teamed with researchers from the Children's National Medical Center, The Catholic University of America, and the University of Maryland to develop molecular diagnostic tools, build biochemical models of the different MD diseases, and identify new classifications of MD.

The team has worked on previous muscle biopsy projects, and is using its existing 107 biopsy data set for developing sensitive and specific methods for biomarker identification and subsequent molecular diagnostics. They will use 200 muscle biopsies of unknown diagnosis to determine the accuracy of the tests.

The second goal of developing biochemical pathway models is to eventually be able to identify interacting proteins and modification states in both normal and dystrophic muscles. The team is interested in identifying new classifications of MD. "Muscular Dystrophies are very hard to diagnose," said (Joseph) Yue Wang, director of CBIL. "Of 200 patient biopsies received at Children's National Medical Center, only 40 can be specifically diagnosed. We hope to identify previously unknown causes of MD, using comparisons of individual patients to the large data warehouse we build for the pathway modeling."

The team has a five-year, $3.5 million grant from the National Institutes of Health for the project, $469,000 of which is Virginia Tech's share.

Print Version

image of article

(124KB .pdf) Print Version