The human-AI hybrid is more accurate than humans or AI on their own.
The human-AI hybrid is more accurate than humans or AI on their own.

AI can now detect lungbreastbrainskin and cervical cancer. But in the world of medical AI, figuring out when to rely on experts versus algorithms is still tricky. It’s not merely a matter of who is “better” at making a diagnosis or prediction. Factors like how much time medical professionals have and their level of expertise also come into play. To address this, researchers from MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) developed a machine learning system that can decide to either make a prediction or defer to an expert.

Most importantly, the system can adapt when and how often it defers to a human expert, based on that teammate’s availability, experience and scope of practice. For instance, in a busy hospital setting, the system may ask for human assistance only when necessary.

The researchers trained the system on multiple tasks, including looking at chest X-rays to diagnose conditions like a collapsed lung. When asked to diagnose cardiomegaly (an enlarged heart), the human-AI hybrid model performed eight percent better than either the AI or medical professionals could on their own.

“There are many obstacles that understandably prohibit full automation in clinical settings, including issues of trust and accountability,” says David Sontag, lead author of a paper that the CSAIL team presented at the International Conference on Machine Learning. “We hope that our method will inspire machine learning practitioners to get more creative in integrating real-time human expertise into their algorithms.”

Next, the researchers will test a system that works with and defer to several experts at once. For instance, the AI might collaborate with different radiologists who are more experienced with diverse patient populations. 

The team also believes their system could have implications for content moderation because it can detect offensive text and images. As social media companies struggle to remove misinformation and hate, a tool like this could alleviate some of the burdens on content moderators without resorting to full automation.

This article is retrieved from

Leave a Reply

Your email address will not be published. Required fields are marked *