Later this year Walker and his colleague and wife, Dianne, hope to show that canine cancer detection can be done.
The husband-and-wife team intend to use a special technique as they study the ability of dogs to detect prostate cancer in human urine samples.
The training program uses a chemical stimulus, n-amyl acetate, which smells like bananas.
Working with the bananalike scent, which the dogs already recognize, will allow the researchers to prove their dogs are well-trained. Put simply, the duo will steadily lower the concentration of the banana-smelling chemical in test samples, then slowly introduce urine samples with and without cancer cells into the training regimen.
"If the dog goes from getting it right about half the time to doing it much better than that, or even showing perfect performance—let's say it takes two months to learn—what that would show is the dog is learning to categorize the urine samples into two classes: normal versus cancer," Walker said.
At that point, the researchers would phase out n-amyl acetate altogether and only test dogs on urine samples.
Since the urine samples will have already been screened by doctors, successfully trained dogs should only be as good as their medically trained human counterparts.
The final step in the dogs' training will require several years of rigorous analysis: Canines must be tested on unscreened urine. Researchers would record the dogs' analysis and track human patients to determine if the dogs are able to diagnose cancer any earlier than conventional medical techniques allow.
Walker cautions that the work is preliminary. He adds that it will be at least another five years before dogs, or any canine-inspired technology, greet people who visit their doctor's office for cancer screening.
|
SOURCES AND RELATED WEB SITES
|
