AI Odor-Based Blood Test Detects Cancer with 95% Accuracy

• Researchers have created an odor-based blood test that can sniff out cancer cells from blood samples.
• The device uses artificial intelligence and machine learning to distinguish cancer cells, including those that are harder to detect.
• The research team hopes that it can be used as a standard blood test in the near future.

A new AI odor-based blood test was able to detect cancer cells from vapors emanating from blood samples. According to the study by researchers from the University of Pennsylvania and Penn’s Perelman School of Medicine, the test has an accuracy of up to 95 percent.

Through artificial intelligence and machine learning, the test studies the mixture of volatile organic compounds (VOCs) emitting off cells in blood plasma samples. Study authors suggest that this non-invasive approach could help diagnose cancers that are harder to detect, such as ovarian and pancreatic.

While it is still an early study, author Johnson called the results “very promising.”

The test can detect “tumors at both advanced and the earliest stages,” Johnson added. If properly developed, it could serve as a standard blood test included in an annual physical exam.

The researchers explained that all cells emanate VOCs. Tissue and plasma from ovarian cancer release VOCs that are distinct from samples of patients with benign tumors. The composition of these VOCs are then detected by the test’s electronic olfaction or “e-nose” system, which is equipped with calibrated nanosensors. It works in a similar way to how the human nose identifies odors.

The study included 93 patients with different to no stages of pancreatic or ovarian cancer. The vapor sensors were able to distinguish between the VOCs from ovarian cancer with 95 percent accuracy and pancreatic cancer with 90 percent accuracy.

The test also correctly identified all patients with with early-stage cancers.

The researchers are now working with VOC Health’s CEO and chief innovation officer Richard Postrel to commercialize the device for clinical and other research applications.

So far, the commercialization process has improved the detection speed by 20-fold. The system can currently distinguish cancer cells from healthy cells in 20 minutes or less.

Postrel expects “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.”

The study results were presented in June at the annual meeting of the American Society of Clinical Oncology.

Another device created by researchers from McMaster and Brock universities in Canada helps patients monitor their blood for unique biomarkers of prostate cancer.

The cancer-detection technology applied in this study is currently being used to develop a similar handheld device used to detect the signature “odor” of people with COVID-19.

Source: Good News Network