Machine learning, predictive algorithms and artificial intelligence (AI) in healthcare are expected to drastically change the way medicine is practiced. While some worry that these programs will eventually replace doctors, the potential benefits to clinical care are expected to calm those fears — especially for radiologists.
The U.S. Department of Health and Human Services commissioned a report titled "Artificial Intelligence for Health and Health Care" in 2017 to, "consider how AI will shape the future of public health, community health and healthcare delivery." According to the report, one of the biggest challenges to the use of AI in medicine is whether the technology will be accepted in healthcare for use in clinical practice.
AI in Healthcare Is Closer Than Ever
In the near future, large data sets may allow AI to recognize abnormal ultrasound images. Algorithms embedded in machine systems will be programmed to know what normal scans are supposed to look like. Instead of a physician having to look at innumerable scans for each patient, the machine would recognize which scans are possibly abnormal and need to be viewed by a radiologist for further consideration. This technology is still being developed and needs to gain FDA approval before it is implemented into patient care.
How Patients Are Using AI for Their Health
AI is already used across the world in ways many people do not realize. For example, Amazon and Netflix use machine learning algorithms to make recommendations to their customers. Google Translate does not actually translate, but instead sorts quickly through documents that have already been translated to provide a phrase or sentence. Driverless cars, facial recognition software, some video games and even health applications all use different types of AI technology.
Smartphone apps and wearables are already helping patients track and monitor their health. However, when patients get the information from their devices, they may not know what to do with it. A gynecology patient may use an app to help her track her periods and ovulation, but she might not think to refer to her application data when she visits her physician.
If they aren't already, physicians should be prepared to request this information, be familiar with the available apps and know which ones are worthwhile. While some newer electronic medical records (EMRs) are ready to accept patient-gathered health data, many are not. The data these patients collect may be helpful to a physician who is trying to discern between a patient's memory and reality. According to the Pew Research Center, 77 percent of Americans own a smartphone, so physicians now have a great opportunity to harness this technology to help their patients improve their health.
A 2018 survey published by the University of Pittsburgh Medical Center found that healthcare organizations see AI as low priority for the coming year, but nearly all of the respondents were planning to offer health and wellness apps to their patients. Respondents felt that these apps promote a "consumer-centered environment" and "improve the patient experience."
Preparing for the Future of AI in Healthcare
Although AI is not quite ready for clinical implementation, gynecologists may want to prepare for it in several important ways. Image management systems that automatically place ultrasound images in the EMR, store them in the cloud, back up the images, and allow for remote viewing or sharing will allow an easier transition to an AI-backed system.
Whether a practice is just beginning or is expanding and updating, its owners should look to invest in equipment that was built with advanced analytics in mind. Practices should also focus on their patient information security protocols, since many AI systems depend on remote neural networks to parse information.
Perhaps most importantly, private practitioners should be open to the benefits that AI in healthcare may bring, ultimately creating more time to focus on patient care.
