Researchers Build Artificial Intelligence That Has The Ability To Detect Rare Cancers With 95% Accuracy
When it comes to cancer, scientists have yet to find the perfect cure and the perfect device for early detection. Those who have followed this development may take comfort in the fact that hope may be on the horizon.
So, what have scientists and researchers come up with? To date, they have designed an odor-based test that sniffs out vapors that emanates from blood samples. This test was able to distinguish between benign and pancreatic and ovarian cancer cells, and so far, it has been 95 percent accurate. This is according to a new study from a team of researchers that hail from the University of Pennsylvania and Penn’s Perelman School of Medicine.
The findings they came up are from a Penn-developed tool. This uses artificial intelligence and machine learning to decipher the mixture of volatile organic compounds (VOCs) that come off the cells in blood plasma samples. The results suggest how this could actually work as a non-invasive approach when it comes to screening harder-to-detect cancers, specifically pancreatic and ovarian cancer cells.
The results of the study were shown when the researchers took the podium at the annual American Society of Clinical Oncology meeting. Johnson explains, “It’s an early study but the results are very promising. The data shows we can identify these tumors at both advanced and the earliest stages, which is exciting. If developed appropriately for the clinical setting, this could potentially be a test that’s done on a standard blood draw that may be part of your annual physical.”
The Penn research team is working with VOC Health right now. They want to commercialize the device, along with others, for research and clinical purposes and applications. They call the device the electronic olfaction (or e-nose) system. This is built-in nanosensors that have been carefully calibrated to detect the composition of VOCs, a substance that emanate from all cells. Previous studies have been made from researchers and these all demonstrated how VOCs released from tissue and plasma from ovarian cancer patients are actually distinct from those released from samples of patients with benign tumors.
Several patients have been drafted for this study. Among 93 patients, including 20 patients with ovarian cancer, 20 with benign ovarian tumors and 20 age-matched controls with no cancer, as well as 13 patients with pancreatic cancer, 10 patients with benign pancreatic disease, and 10 controls, the vapor sensors that came from the cells had been discriminated the VOCs from ovarian cancer with 95 percent accuracy. It showed 90 percent accuracy for the sufferers of pancreatic cancer. The tool was also used for cancers that were in its early stages. It had correctly identified all eight patients accurately as well.
The technology makes use of a pattern recognition approach much like how our own sense of smell works. It detects a distinct mixture of compounds that tells the brain what it’s smelling. The tool developed to do this was trained and tested for the same thing, and it was able to identify the VOC patterns more associated with cancer cells and those associated with cells from healthy blood samples in just 20 minutes and even less.
The team worked in collaboration with Richard Postrel. He is the CEO and chief innovation officer of VOC Health. With his help, there was a marked improvement in the detection speed by 20-fold.
The researchers wanted to expedite its commercialization process in order to make it more available to establishments who need it. Postrel asserts how the “initial prototypes of commercial devices able to detect cancer from liquids and vapors will be ready soon and be provided to these Penn researchers to further their work.”
Other teams are working on related devices and as of now, researchers from McMaster and Brock universities in Canada are working on the development of a device that lets patients monitor their own blood. This is done to find the unique biomarkers of prostate cancer as seen in the picture below.
With all the research to help detect cancer, Johnson, along with his co-investigator Benjamin Abella, MD, a professor of Emergency Medicine, and the VOC Health were all awarded a two-year, $2 million grant. This was given by the National Institutes of Health National Center for Advancing Translational Sciences. Honor was given to them because of the development of a handheld device that can detect the signature “odor” of people with COVID-19. This also made use of the very same cancer-detection technology that was used in this study. And it is with hope that this can truly come to fruition in order to help those working nonstop in the medical field.