Chemistry Related
#
Dr. Shishir Shah
Department of Computer Science
photo by Jerome Crowder
Home
Dean's Message
Targeted Drug Delivery Strikes Gold
UH Professor CARES
Shining a Light on Cancer
On the Move
Newsletter Archive
Giving
 
by Noelle Heinze
With over 1.2 million new cases diagnosed every year, cancer is second only to heart disease as the leading cause of death in the United States. The earlier cancer is detected, the greater the likelihood of eradicating the disease and restoring a patient’s quality of life.
Assistant Professor of Computer Science Shishir Shah and his UH colleagues in the Quantitative Imaging Laboratory are working with collaborators at The University of Texas Medical Branch in Galveston to develop new technology to rapidly image, analyze, and identify abnormal cells.
One project, called virtual microscopy, combines sophisticated software with a shoebox-size machine to image tissue samples, display them on a computer screen, and generate automated analyses.
“We are building networks,” says Shah. “If a clinician in the Texas Medical Center wants a colleague in New York or India to view a slide, it can be done instantaneously. Our software and imaging system allow clinicians on two different continents to consult on a diagnosis in real-time. A clinician can log on to a computer from anywhere in the world, and without the actual tissue sample or a microscope, they can zoom in and look at the structures of the cells.”
“It changes the paradigm of how we provide health care,” Shah says. “It will save money and time and will increase the accuracy of results by increasing the number of pathologists who view a particular tissue sample.”
While similar microscopy machines exist, they require a clinician to input data and cost about $150,000. Shah’s machine is fully automated and relatively inexpensive—it will cost less than $10,000.
Graduate student James Thigpen is working with Shah on a second project, multispectral microscopy, which is a system to analyze cells by their spectral signal, the degree to which a particular cell absorbs light.
“Organic material absorbs and reflects light under a microscope,” explains Shah. “How will a cancer cell, which has a different biochemical state than a healthy cell, reflect light? We want to identify a spectral signature for cancer cells by using 31 different wavelengths to image cells, instead of a single light source.”
“In early-stage prostate cancer and other solid tumors, cancer cells are found routinely in peripheral blood. However, distinguishing these cells from among as many as 100,000 healthy cells is a daunting task,” says Shah. “By looking at a cell’s morphology, its size and shape, and combining this information with its spectral signature, you can make a better detector for cancer cells.”
Both projects are currently under testing to refine the technology. If successful, virtual microscopy may be available in as little as nine months and multispectral microscopy in 18 months.
For more information about Shah’s research, please visit http://www2.cs.uh.edu/%7Eshah/.
© University of Houston 2006