In today’s hospitals and healthcare clinics, a new doctor’s new assistant is now often on the job — in the form of artificial intelligence. Whether it’s analyzing medical images or guiding robots that assist with surgeries, AI is making steady inroads into our hospitals and clinics. Need an online nursing assistant or a watchdog that helps detect dosage errors? There’s an AI application for that.
The advent of AI in healthcare is a promising trend in terms of both patient care and economic efficiency. AI can help us address a forecasted shortage of physicians, particularly in specialty-care fields, while containing the costs of caring for an aging and growing population. A recent study by a team of researchers from the consulting firm Accenture found that the use of 10 promising AI applications could create up to $150 billion in annual savings for U.S. healthcare by 2026.
For an example of the potential of AI in healthcare, we need to look no further than Gustave Roussy, a leading European center for cancer research and care. In a study published this summer in the medical journal The Lancet Oncology, a team of medical researchers from Gustave Roussy and a few other institutions demonstrated that AI can process medical images to extract biological and clinical information to help with immunotherapy treatment, according to a news release on the study.
In this groundbreaking study, the researchers used an algorithm they designed and developed to analyze CT scan images and create a “radiomic signature.” This signature defines the level of lymphocyte infiltration of a tumor — or the degree to which immune cells have moved from the blood into a tumor cell. The radiomic signature also provides a predictive score for the efficacy of immunotherapy in the patient.
Here’s how AI played into this research: Using an approach based on machine learning, the team first taught the algorithm to use relevant information extracted from CT scans of patients participating in the study. Then, based solely on images, the algorithm learned to predict what the genome might have revealed about the tumor immune infiltrate, and it established the radiomic signature.
The announcement summarizing the findings of the study notes that in the future, physicians might be able to use imaging to identify biological phenomena in a tumor located in any part of the body without having to perform a biopsy.
At Dell EMC, this research is particularly close to our hearts because we are active supporters of Gustave Roussy.