Modern artificial intelligence systems can be used not only for complex mathematical vysilani, but in order to improve the quality of medical care. For example, recently, researchers from the Massachusetts Institute of technology (MIT) were able to apply AI DeepMind for more effective cancer therapy by reducing the number of complications in several times.
For the implementation of their ideas, the experts chose 4 drug: temozolomide, procarbazine, lomustine and vincristine. After that, they allowed the DeepMind algorithms to calculate the schedule of these drugs, based on the facts about treatment with these agents, and information about the patients, biochemical interaction, and so on. In the result, AI has developed a course of treatment and dosage individually for each patient-participant experience.
The initial stage of medication did not differ from “traditional”. But later, the AI adjusted the dosage: has repealed some drugs, for some, increased the dose, but for others reduced. Just experiment was attended by 50 people. For each DeepMind counted several thousands of treatment options, excluding those that may be ineffective. Moreover, during treatment, DeepMind continued to study, receiving points if patients stanovilos better and losing them if the health of people uhudshilos.
As a result, the frequency of use of chemotherapy drugs has decreased from a monthly to a few treatments per year, the dosage is decreased by half, while preserving the positive dynamics of cancer treatment and fewer side effects from chemotherapy. Researchers believe that a similar approach can be applied in other areas of medicine in which treatment requires regular medication.
AI Google reduced the number of adverse effects in anticancer therapy
Vladimir Kuznetsov