Five Things AI Is Doing In Healthcare, And More

Artificial intelligence is profoundly transforming healthcare, from drug discovery and medical imaging to patient analysis and personalized treatments. Dr. Eric Rosenthal, a leading neurologist, highlighted AI's role in neuroscience, describing it as "two lanterns" for interpreting patient signals and orchestrating care. AI can document doctor-patient conversations in real-time, freeing medical staff for deeper clinical connection. Crucially, AI analyzes speech and motor skills to detect neurological conditions earlier than traditional methods, extending medical care into patients' daily lives. It also empowers patients with severe disabilities to communicate and guides surgeons with precision during complex procedures. Rosenthal envisions a future of "offices without walls," fostering greater connectivity and truly personalized medicine.

What is AI doing in healthcare? Well, the short answer is, a lot. But the longer answer would probably be a bullet list. Here are some of the most important and comprehensive advances that I have seen as I listen to hundreds of presentations annually, on AI and the science around it:

That’s just a partial list. The list goes on.

We had an event in June where Eric Rosenthal showed off some of the things that AI is doing in healthcare, particularly in neuroscience.

Rosenthal has a lot of credentials. He is an assistant neurologist at Massachusetts General Hospital, an associate professor of neurology at Harvard Medical School, and acts as MGH’s director of both the Neurosciences Intensive Care Unit and the Neurology Data Science & AI Center. He’s also the director of the Brigham NeuroAI Center, which he referenced in his talk.

Rosenthal started off making the analogy to Paul Revere and Boston in the American Revolution, a connection that I have made repeatedly in noting the local capacity for innovation and invention, and tenacity of spirit.

“Paul Revere's job wasn't to create the information,” Rosenthal said, “his job was to carry it, and in a way, we're going to talk about artificial intelligence as … like two lanterns: one lantern is going to be our way to interpret those signals, and learn also to orchestrate them.”

He talked about the process of bringing healthcare to the patient.

“You'll see the cell phone on the desk, you know, listening and helping record the conversation, so the machine will carry that record, and you can actually talk to someone,” he said, referencing the ability of AI to do documentation in real-time in a doctor’s office, freeing up the human medical staff for deeper clinical connection to the patient.

One of Rosenthal’s points was that doctors don’t have long to do their work with a patient, and may not be fully up on everything that is happening with a patient’s home life.

“What percent of time do you spend in your life in doctor’s offices?” he asked. “Our healthcare system asks patients to come to us, in this scheduled time, and compress their life. Some people come with a list, and … you know what's happening to your mother, or your sister, or your daughter, better than your physicians do. So what's going to happen, what we're going to see in the future is that we're going to be able to listen in not on what exactly you're saying, but to hear your struggle.”

Presenting what he called “spectrogram of speech,” Rosenthal explained how this type of analysis works.

“We can differentiate between patients who are affected by neurologic disease, and patients who are not,” he said, of people using the tool. “So, for example, the amount of time you participate in communication, how your speech is, your cadence … we can actually understand what your struggle is, and we would know, just like your family would know, if you're kind of having trouble before, before your doctor will.”

The same, he said, is true for motor skills. Rosenthal showed some slides illustrating this.

“Our lives have totally been transformed in the patient space outside the office, and the key is for us to bring our medical care into that space,” he said.

“Our map today is hyper-connected,” Rosenthal continued, “but we need to help people understand if their social network is deteriorating. Similarly, when we sit together in the doctor's office, or the provider's office, we really want to chart a course together, and we want a map for that.”

He gave the example of patients with spinal cord injury or ALS, who can now have implants that help them communicate.

“They can manipulate computers when their arms don't work,” he said, “they can talk where their voice doesn't work, and we can actually look and see what their voice used to be.”

There’s another way that ground-breaking technology is pushing healthcare forward, that Rosenthal covered, as well.

“Our surgeons, who are doing brain surgery, are going from seeing a tumor to actually being guided by an ultrasound guided image of a tumor, so they know they can get the most out of it,” he said, showing a slide with a process of continuous care. “We're going from discovering treatments through trial and error, increasingly, to now matching therapies to your biology, so that we can actually design something that's truly personalized, and instead of applying for FDA approval like it's some finish line, it's actually an iterative process of co-design.”

He also mentioned a greater connectedness.

“Every revolution needs principles,” Rosenthal said, “and if we had our bill of rights, I think (one part) is thinking about how we can connect ourselves in that office, with a patient, having an office without walls. From that, I think we'll transition to thinking about not just an office without walls, but really about an institution without walls.”

That was Rosenthal’s “Vision for 250,” and it resonated with me, because we can use those principles to build better in the healthcare environment. Stay tuned for more.