“Cancer” is a catch-all phrase used to describe any of the hundreds of diseases that fall under the umbrella of this condition. Cancer itself can manifest itself in different parts of the body and in different ways, and, as a result, each distinct variation must be treated in a particular way. However, with so many subtleties in the different forms of cancer which exist today, it is easy to see how errors can occur in both diagnosis and treatment.
Researchers from the MIT Computer Science and Artificial Intelligence Laboratory have teamed up with doctors from Massachusetts General Hospital to develop a computer model which utilizes algorithms to narrow down possible diagnoses. In a test study, researchers looked at patients with three different types of lymphoma. The computer model pulled information from patients’ diagnostic tests, previous cases, patient history, and information from the World Health Organization’s diagnosis parameters. The model’s extensive testing and diagnostic processes were ten percent more accurate than current techniques. Researchers are now looking to expand the scale of the model to continue improving diagnosis and treatment of various forms of cancer.