Virtual patients help doctors improve bedside manner

Doctors can hone their people skills on the AIs before treating real patients.

Good doctors are able to diagnose and treat patients. Great doctors do so without forgetting that their patients are human beings who might be scared, confused, and in pain.

By treating patients like people — empathizing with their situations, and ensuring that they feel heard and understand what’s happening to them — doctors’ bedside manner can not only improve patients’ experience, but also patient outcomes.

Now, a new platform called Virti is helping doctors hone these soft skills with patients who aren’t real human beings — they’re virtual patients powered by AI.

Teaching Bedside Manner

Medical schools traditionally teach bedside manner by having students examine actors pretending to suffer from a specific illness.

Med students talk to the simulated patients about their symptoms before providing a diagnosis. After the encounter, the fake patients provide feedback on the doctor-to-be’s communication, interpersonal skills, and where they have room for improvement.

Playing out these scenarios with actors from different demographics can help med students identify any unconscious biases that might have that could affect their judgment. But it requires med schools to find — and pay — a variety of actors to pretend to be patients.

The subjectivity of the actors is another limitation to this method of teaching bedside manner — while the fake patients are trained to be as objective as possible, two actors still might provide different feedback on the same doctor interaction.

The pandemic has put some of these training programs on hold entirely, as having students and simulated patients in the same room would increase their risk of contracting COVID-19.

Virti can help overcome all of these issues by helping med students and healthcare providers develop good bedside manner remotely.

Virti’s Virtual Patients

Virti provides hospitals with virtual patients — computer-generated characters that can interact with med students or doctors the same way simulated patients can, but through a phone, computer screen, or VR headset.

The virtual patients are powered by AI software that lets them provide realistic responses to users’ questions — they can describe their symptoms, for example, based on the illness they’re supposed to have.

The AIs are programmed to have realistic mannerisms, and their physical characteristics (skin color, age, height, etc.) are customizable.

During a session, Virti records data on not only what the trainee says, but also their tone and the quality of their answers to the virtual patients’ questions. If the user is wearing a virtual reality headset, Virti can even track eye contact between them and the virtual patient.

After the session, the platform provides a score based on all this data, as well as the trainee’s ability to correctly diagnose the patient.

Several major healthcare systems, including Britain’s NHS, are now using Virti, and while its virtual patients aren’t perfect training tools — it can be difficult for digital creations to replicate the emotions of humans — they have several advantages over patient actors.

The feedback from the platform is always objective, and practicing bedside manner on different demographics of patients is as easy as tweaking a few parameters.

Because all of the patients exist in silica, large numbers of students or doctors can train with them anytime, anywhere a screen is available — no need for hospitals or med schools to coordinate schedules with actors or worry about paying them for every session.

“What we wanted to do with the virtual patient was create a scalable, data-driven way for people to practice their soft skills and communication,” Virti founder Alex Young told the Washington Post.

Well-Rounded Doctors

The Virti platform isn’t limited to bedside manner training, either.

Since the beginning of the pandemic, several institutions have used it to teach healthcare providers the hard skills they need to interact with COVID-19 patients, such as how to use PPE and safely perform CPR.

“We’ve been using Virti’s technology in our intensive care unit to help train staff who have been drafted in to deal with COVID-19 demand,” Tom Woollard, the clinical skills and simulation tutor at West Suffolk Hospital, said in a statement.

“The tech has helped us to reach a large audience and deliver formerly labour-intensive training and teaching which is now impossible with social distancing,” he added.

In an NHS study involving 50 participants, those trained using Virti on COVID-19 skills outperformed those trained using traditional methods — suggesting that AI can teach doctors both how to treat patients and how to treat patients like people.

We’d love to hear from you! If you have a comment about this article or if you have a tip for a future Freethink story, please email us at [email protected].

Target is now offering the world’s first “robot manicure”
A robot that uses AI and 3D cameras to paint fingernails is now giving Target customers 10-minute manicures for just $8.
Meta can (kinda) guess what you’ve heard via your brain waves 
Meta has created an AI that can tell what you’re hearing based on non-invasive brain scan measurements.
Six benefits that the metaverse offers to colleges and universities 
Colleges and universities are keen to jump on the metaverse bandwagon, even before knowing what the metaverse really is.
First-of-its-kind trial shows AI beat humans at analyzing heart scans 
Echonet, an AI trained to assess a measure of heart function, has outperformed trained technicians in both accuracy and efficiency.
How AI could learn from Aesop’s fables 
When USC researchers tried to teach an AI simple fables, they discovered that the process is a lot harder than it sounds.
Up Next
artificial genomes
Subscribe to Freethink for more great stories