The Challenges of Training AI for Healthcare Applications

The development of machine learning and precision medicine applications requires a training data set with an outcome variable, such as disease onset. AI engineers and IT executives in the health technology field face major challenges in determining the right data strategy for training models. These challenges include the need for health training data to train AI models, the potential for the strategy to increase existing inequalities in the health system, and the need to create innovative algorithms that consume less data and use more human-like artificial general intelligence. Data lakes are an entry point to what could eventually become artificial intelligence, and they're already taking hold in healthcare.

A machine with artificial intelligence must be able to accept information about the problem from its environment, generate a list of actions it could take and maximize its chances of achieving its objectives by using logic and probability to choose the activities with the highest probability of success. Remote patient monitoring could also benefit from an artificial intelligence program that is responsible for coordinating Internet of Things equipment in the home for elderly, disabled or frail patients. IBM has been dedicating health data to Watson for several years and has disbursed billions of dollars to acquire big data analysis companies that will promote its goal of creating a truly intelligent partner for quality care. Technology companies need to create innovative algorithms that consume less data and use more human-like artificial general intelligence.

Consumers are already familiar with voice-answered phone menus and automated chat bots on websites that can answer questions or establish connections with varying degrees of success, but healthcare could offer a much more robust AI experience if Amazon CEO Jeff Bezos takes Alexa big time. As healthcare organizations begin to focus on customer expectations in response to increased out-of-pocket costs and value-based reimbursements, providers must learn to personalize the patient experience, reduce unnecessary expenses, and maintain open lines of communication between office visits to keep patients as healthy as possible. How Alexa will improve the healthcare experience remains to be seen, but it is possible that the hospitals of the future will have an AI listening device in every patient room, replacing nurse calling systems, doctor locators and public address announcements of yesteryear with an intelligent, discreet and highly responsive communication system. Providers should also discuss liability issues when an AI program makes a mistake that causes harm to patients, how to measure and manage risk by introducing AI to a new task, and how to safely test novel technologies in the healthcare environment without exposing patients to potentially dangerous situations. While this bleak vision of the future is still firmly embedded in the realm of science fiction novels and summer blockbuster movies, recent advances in artificial intelligence (AI) and machine learning cause some to wonder if Isaac Asimov's Three Laws of Robotics will be applicable to everyday life sooner rather than later.

Deanna Trueman
Deanna Trueman

Social media practitioner. Devoted music fan. Certified zombie ninja. Total bacon expert. Hardcore zombie practitioner.