How to Use Digital Signals to Facilitate Cancer Surgery

© Kheng Guan Toh / shutterstock.com

A surgeon’s job is to look, alter, and/or remove—in other words, to recognize and take action. But distinguishing cancer from normal tissue can be complicated without context, said Baran D. Sumer, MD, associate professor and chief of the division of head and neck oncology in the department of otolaryngology at the University of Texas Southwestern Medical Center in Dallas. And a determination, for example, of the extent of base of tongue cancer during surgery can be obscured by the background noise of uncertainty. Nanotechnology, however, can help diminish that noise.

Dr. Sumer’s H. Bryan Neel III, MD, PhD Distinguished Research Lecture, presented September 16, 2019 at the American Academy of Otolaryngology–Head and Neck Surgery Annual Meeting in New Orleans and entitled “Digitization of a Physiologic Signal to Facilitate Oncologic Surgery,” gave attendees a basic explanation of information theory and the mathematics of communicating information, made the distinction between an analog and a digital signal, and showed how macromolecular processes can lead to emergent chemical properties. The presentation also showed that molecular cooperativity can be leveraged for cancer imaging and discussed the digital nanoprobe Dr. Sumer and his collaborators developed.

Bayesian Inference 101

Bayesian inference describes a method of statistical inference in which the probability of an event is updated based on previous knowledge of conditions possibly related to it, according to Dr. Sumer’s presentation. “Bayesian theory is that the probability of events are not just independent. They’re based on a prior knowledge of other things that may be going on,” he said, adding that as new information arises, we can update our idea of that probability.

Dr. Sumer illustrated the human mind’s ability to navigate abstractions by filling in missing information from previous experiences using a series of images—first a highly pixelated, abstract image of Abraham Lincoln, then a less pixelated image, and finally a photograph of the president. “If we give just a little more information, a lot of people can look at that picture and say, “That looks like President Abraham Lincoln.”

We want high sensitivity, we want high specificity, and we want to suppress the noise. —Baran D. Sumer, MD

Similarly, surgeons can fill in information from previous surgical experience to recognize structures. “We probably do it without even knowing it,” he said. “Is it tumor or tonsil tissue? We can palpate it and see if it’s soft or firm. We can look at preop PET scans. With that knowledge, we can update what we’re doing as we proceed in surgery.”