Artificial Intelligence Disease Diagnostics
IPMD is currently developing artificial intelligence diagnostics platforms that are able to screen and diagnosis many complex diseases in early stages with a very high degree of accuracy. The target diseases include but are not limited to lung cancer, prostate cancer, pancreatic cancer, breast cancer, colon cancer, head & neck cancer, brain cancer, gastric cancer, cervical cancer, ovarian cancer, Alzheimer’s diseases, Autism Spectrum Disorders, and many other complex diseases.
Concerning non-invasive in vitro diagnostics for the complex diseases, too many biomarkers (known and/or unknown) are involved and interact each other differently from a person to person. This makes availability of accurate screening and diagnostics platforms for the complex diseases extremely difficult. In order to solve these major problems, IPMD has developed highly sophisticated artificial intelligence platforms based on DNN algorithms in order to allow a very low cost, non invasive, yet highly accurate diagnostics system to be made possible.
An award-winning law firm and representing IPMD, Inc., Nixon Peabody filed a patent application covering a system for diagnosing chronic diseases that uses cutting edge artificial intelligence algorithms. Called “Lablet Artificial Intelligence Platform" developed by IPMD, uses an algorithm similar to Google’s AlphaGo and IBM’s Watson to train their artificial intelligence. However, with one key difference. While Watson relies on medical images acquired from methods such as CT and MRI scans as input data, IPMD uses a data-driven multivariate statistical approach. Employing a highly sophisticated biochip to analyze a small amount of human sample (fingertip blood, saliva, urine, etc.). Within the human sample, Lablet will simultaneously analyze thousands of biomarkers using short single strands of DNA or RNA oligonucleotides, which act similarly to antibodies. This is conducted by folding and adopting three-dimensional structures based on the interactions of complementary base pairs. This three-dimensional configuration is key to allowing oligonucleotides to have the ability to bind to specific biomarkers. By using this process, Lablet captures significant biomarkers found in the sample, thus allowing for artificial intelligence to then correlate identified patterns to either a diseased or not diseased state. The more input data inserted into the Lablet device, the greater the accuracy of the diagnostic results. The company’s current goal is to achieve at least 99% accuracy.
IPMD has formed a research consortium with VA Medical Center Greater Los Angeles Healthcare System, and includes many renowned scientists from the Veterans Affairs Greater LA Healthcare System, UCLA School of Medicine, University of California Institute of Prediction Technology, and many other Research Institutes.