Comprehensive coverage

Artificial intelligence has contributed to the increase in finding a match for experimental breast cancer treatments

Implementation of the Watson system of IBM in the work processes at the hospital resulted in a significant shortening of the scanning time required to match a patient to the appropriate test for his condition, and to improve the accuracy of the scanning process

Illustration: Thomas B. Shea / Feature Photo Service for IBM / Flickr.
Illustration: Thomas B. Shea / Feature Photo Service for IBM / Flickr.

IBM and the Mayo Clinic in Rochester, Minneapolis, reveal first results of using the Watson artificial intelligence platform to match cancer patients to clinical trials. In the first 11 months of the system's operation, the hospital recorded an 80% increase in the recruitment of patients for medical trials in breast cancer treatments.

Clinical trials offer patients access to new treatments, but the processes of matching patients to trials that suit them currently require dealing with a lot of manual work. Today, only 5% of cancer patients in the US participate in medical trials - which means that the trials are relatively slow, and some will not even be completed at all. These delays stop the development of medical research, and prevent access to better treatments for all patients with the disease. In this reality, innovative solutions are required that will enable us to deal with the clinical need that has not yet been met, advance cancer research and treatments for the disease and improve treatment results.

The implementation of the Watson system in the work processes of the hospital resulted in a significant shortening of the scanning time required to match a patient to the appropriate test for his condition. In this way manual work is avoided and all patients can be reviewed in an attempt to match them with the most appropriate treatment from among the variety of treatment options that exist within the framework of active medical trials. The speed and accuracy of IBM's integrated activity together with the team of trial coordinators, allows the hospital's doctors to develop treatment plans for patients that reflect the entire range of options available in any given case.

The two organizations also announced a new agreement designed to expand and deepen the training processes and use of the system. Until now, the system has been trained to support the adjustment processes of patients for clinical trials in the treatment of breast cancer, lung cancer and bowel cancer. Today, Watson training is already underway to identify suitable patients for medical trials in other types of cancer.

The collaboration between IBM and Mayo Clinic was presented in detail at the annual conference of the American Society for Information Systems and Medical Management, which took place this weekend in Las Vegas. The Watson system for matching patients to clinical trials is designed to provide systematic matching of patients to medical trials relevant to their condition as part of the overall treatment plan in that program. Mayo Clinic's team of experts developed optimal processes for managing work processes and scanning patients, and began a systematic process of educating patients about the treatment options within these trials. Starting in June of last year, the hospital operates the Watson systems together with a team of medical trial coordinators for breast cancer patients.

Dr. Michal Rosen Zvi, director of medical information at IBM's global research laboratories: "Artificial intelligence enhances human capabilities thanks to its ability to deal with huge amounts of data, identify patterns based on many variables and offer smart insights. When these capabilities are harnessed to the fields of medicine, we allow every doctor to have a powerful surgical tool that is only getting better and better."

See more on the subject on the science website:

Leave a Reply

Email will not be published. Required fields are marked *

This site uses Akismat to prevent spam messages. Click here to learn how your response data is processed.