Researchers from the University of Texas at Dallas have shown that imaging technology that is used to map the universe might potentially be used to identify cancer cells more accurately in the operating room. The research, conducted by Dr. Baowei Fei and colleagues shows that hyperspectral imaging, along with artificial intelligence, can predict the presence of cancer cells with 80 to 90% precision in 293 tissue specimens from 102 patients undergoing head and neck cancer surgery. The team has published the study in the journal Cancers. Fei, Professor, Bioengineering, and Cecil H. and the Cecil H. and Ida Green Chair, Systems Biology Science, Erik Jonsson School of Engineering and Computer Science, was recently awarded a $1.6 million grant by the Cancer Prevention & Research Institute of Texas (CPRIT) for further development of the technology dubbed a ‘smart surgical microscope.’
The technique, when fully developed, will be tested in clinical studies before it can be used in operating rooms. Fei hopes to employ the technology to help surgeons better detect cancer during surgery, reduce the operating time, decrease medical costs, and save lives. Hyperspectral imaging is non-invasive, portable, and does not need radiation or a contrast agent, explains Fei. Hyperspectral imaging, which was initially meant for satellite imagery and orbiting telescopes, among other applications, is capable of going beyond what the human eye can see as cells are inspected under ultraviolet and near-infrared lights at micrometer resolution. Currently, pathologists use a process known as intraoperative frozen section analysis, wherein they analyze tissue samples from a patient undergoing cancer surgery and is still under the effects of anesthesia.
This is a complex process, and in some cases, the surgeons are unable to sample or detect cancer cells during surgery, which might result in the need for additional surgery. By analyzing the way cells reflect and absorb light through the electromagnetic spectrum, experts can achieve a spectral image of cells, which is just as unique as a fingerprint. Fei says that the CPRIT grant will support the team’s efforts to train the microscope to recognize cancer by using images of cancerous and non-cancerous cells in an extensive database.