Virginia Company Develops Automated TB Detection System
The power of computer algorithms to help assist or automate traditional laboratory methods is one of the major value propositions for digital pathology. Microscopic techniques for the diagnosis of TB are often tedious and made more tedious by the fact that such cases are often "rare" events, meaning not only is the incidence low (depending on the community and hospital population) but the actual number of organisms that may be present is often very low. For the average community hospital in the US, microscopic diagnosis for TB in body fluids or tissues involves special stains that must be reviewed at high power while "looking for a needle in a haystack".
While some are concerned that computer algorithms may replace traditional diagnostic skills performed by pathologists, these I think are just the types of technologies that will help perform some very time consuming, tedious, poorly reimbursed laboratory tests in a manner that is cost-effective, faster and more accurate. Who can argue with that?
It should be noted that some molecular techniques have shortened the time to diagnose and classify mycobacterial infections but may not be available beyond more conventional slide-based techniques from tissue or body fluids.
If the results from the below press release (see link) hold up, I think this is an important milestone for pathology and laboratory medicine in the digital era.
Donald G. McNeil, Jr. writes in the New York Times (4/13, D6) Global Update column, "One of the difficulties of diagnosing tuberculosis is that there is no simple blood or urine test. Instead, a laboratory technician must take a sample of sputum coughed up from the lungs, stain it and inspect it under a microscope for the telltale bacteria, which resemble long-grain rice." This "takes expertise that is often rare in poor countries." That is why "Guardian Technologies, a Virginia company that was started to help airport X-ray scanners distinguish explosives in luggage from innocuous plastics and liquids, has developed a system that automatically scans microscope slides for the bacillus." Notably, the "company's software algorithms can spot distinctive shapes, colors and densities that untrained eyes may miss."
