In a word, “innovation”- the kind that changes the world and improves the lives of people. It’s been 14 years since an original article first appeared in The Missile Defense Agency’s “Tech Update” discussing the origins of CDx Diagnostics’ Neural Network. The imaging algorithm used was originally part of the “Star Wars” missile defense program – helping to differentiate between warheads and decoys. The catalyst for the innovation for CDx Diagnostics came when the Founder and Chief Science Officer, Mark Rutenberg, decided to think about potential applications in healthcare and point that algorithm at tissue to differentiate between healthy and unhealthy cells instead.
The Strategic Defense Initiative (SDI), also known as Star Wars, was a program first initiated on March 23, 1983 under President Ronald Reagan. The intent of this program was to develop a sophisticated anti-ballistic missile system in order to prevent missile attacks from other countries, specifically the Soviet Union. A research-and-development effort that emerged from SDI was Brilliant Pebbles, which was focused on technology that would enable basing interceptors in space. One of the proprietary uses that came out of Brilliant Pebbles helps power WATS3D and other tests from CDx today.
The nickname “Star Wars” may have been attached to the program for some of its abstract and farfetched ideas, many of which included lasers. Furthermore, the previously released science fiction movie titled “Star Wars,” caused the public to easily associate this program with new and creative technologies. The weapons required included space- and ground-based nuclear X-ray lasers, subatomic particle beams, and computer-guided projectiles fired by electromagnetic rail guns—all under the central control of a supercomputer system. By using these systems, the United States planned to intercept intercontinental ballistic missiles while they still flew high above the Earth, minimizing their effects.
What are Neural networks, you ask? Neural networks are a set of algorithms, modeled loosely after the human brain, that are designed to recognize patterns. They interpret sensory data through a kind of machine perception, labeling or clustering raw input. The patterns they recognize are numerical, contained in vectors, into which all real-world data, be it images, sound, text or time series, must be translated. The artificial intelligence learns by being fed examples. In the case of cancer prevention, this brain food was images of healthy and unhealthy cells.
What does techy space science have to do with cancer prevention?
The development of the Pap smear test, the first widely used screening test for cancer,
is now recognized as one of the most significant advances in the control of cancer in the 20th century. The Pap test (or Pap smear), a well-proven way to prevent cervical cancer. It finds conditions that may lead to cancers and can find pre-cancers before they can turn into invasive cancer. If a pre-cancer is found it can be treated, keeping it from turning into a cervical cancer.
CDx Diagnostics created a life-saving version of the Reeses Peanut Butter Cup – the candy that merged together two things that seem to have nothing to do with one another. One, a snack or a dessert. The other, the perfect complement to jelly and the staple lunch for many children.
When you combine a super-computer, AI-powered, detection algorithm developed to keep Americans safe from missile attacks with the Pap smear model of looking for abnormality before it can become cancer (which is often like finding the needle in a haystack), you have developed the CDx Diagnostics preemption platform – a reliable process to find pre-cancer (dysplasia) and help prevent cancer.