How The Ottawa Hospital uses AI ambient voice capture to reduce physician burnout by 70%, achieve 97% patient satisfaction

Clinical and patient satisfaction (or its absence), patients in the field of health – are a remarkable issue for patients, clinicians, clinicians simply to fight to make clinicians simply to fight.

Ottawa Hospital (TOH) set out to solve this problem with integration Microsoft’s Dax Copilot last year. And it has already been a great impact: 70% of the show, 70% in a match in the early results, 70% in the clinical-notification and fatigue, 93% of patients say better or equivalent care experience.

“The use of care is one of the greatest issues,” Glen Kearns, EVP and Tohun Cio, Tohun Cio, TOH is one of the biggest issues. “If we can improve the transmission, a few patients each in the queue, even a few patients, the more than 10 doctors who care, and then increase to 365 – it is not an ineffective increase to care.”

AMENT AI as an active assistant

Toh was the first Canadian Hospital for Microsoft DAX Copilot, which is directly integrated with a widely used electronic health record (EHR) platform platform platform. In March, DAX Copilot in Microsoft Dragon Copilot, which is used by one of 600,000 doctors of Microsoft, Dragon Medical (DMO), Tech Giant.

Dax Copilot, who comes out of the box, seizes Doctors-sick conversations Creates a real-time clinical notes project through a mobile application, Kenn Harper, project manager for Microsoft dragon, told Venturebeat.

“Clinicians can start a celebration from a mobile phone, can put their phones on the ground, explore the patient, talk to the patient, not different from whatever they do.”

After the system, the details of the visit (symptoms, diagnosis, treatment plans, follow-up plans), immediately available EHR; All clinic needs to quickly review and finalize the work.

“Instead of writing something from scratch and remember something in front of a patient, it only works automatically,” he said, “Once” a pretty clear “first time” he said.

To tune the Dragon Copilot, Microsoft uses “Mass Deposit of Cleaned Clinical Data for Years”. Engineers are constantly fixing large language models (LLMS) to understand the system’s medical information and can safely generalize and generalize safely.

To further improve accuracy, models, emergency doctors, a dermatologist, a dermatologist, or optimized for whether or not any other medical care.

In contact loops, teams analyze the first draft prepared by the machine, and compare how much of this and how much the clinic was changed.

“In this case, on average, when it comes to information and learning from this information, editing reduces editing,” Harper said.

Thoh is how to reduce the doctor’s exhaustion, how to improve visits

According to the Canadian Medical Union, doctors spend around 10 hours per week About administrative tasks as patient meetings update the schedules.

Kearns explained that the TOH has developed a solid valuation plan for DAX Copilot, which contains monthly updates via Microsoft Power Bi dashboard. This includes opinions from the information of clinicians, patient surveys and epics.

“This framework helps us to influence and follow the development of guidance,” he said.

Typically at the end of the day or in the day, doctors should then return and complete the documents from patient visits and explained. However, after the timeline, graphics and documentation works “All categories Doctors“It doesn’t only keep time, but it helps to reduce because they have less tiring work.

The tool also improved the clinician’s code of clinic loading during the visit: instead of inserting patient details and navigate the documents and forms, said they said “different and better.” In addition, “We have seen the growing level of the passage, every turn, in every turn.”

Kearns stressed that all patients were asked to be agreed before being appointed, and I got access to the notes on mycart patient portals. They are also provided with literature on the program and make sure the health notes are always confidential and reliable.

The patient receiver was also “very positive” with a 97% report, noting that the AI ​​instrument is a typical destination or better. “It is often involved in the opportunity to communicate with them and deliberately, during a visiting trip over time,” he said.

He said he looked at the back, he could be used in scenarios such as the biomarker detection and social identification of health (non-medical problems such as a person’s health, transportation or lack of transport.

“The documents have a lot to go down the road in the seizure space,” Kearns said.

To overcome digital teammates personnel problems

Another area of ​​AI is combined is “digital teammates.” Last summer, the enterprise used work with Deloitte and began Sophie, speaking in several languages.

Interestingly, you can interpret the response of a sick feeling and behavior.

“I hate to say that, but patients lie to the clinicians,” Kearns acknowledged. “You know, ‘What’s the scale of the pain?’ ‘Oh, I’m doing well, five out of 10.’ But then Sophie will have the ability to take a look at your face and go to the face, ‘Well, that’s not like five.’ “

Then you can continue to ask what the patient’s definition of five and pivot based on objective information.

Toh, in addition, another avatar will help patients visit and access to the health care system and in advance.

“I don’t think that no one is aware of the human resource crisis, which is in a global scale,” Kearns said. “Really try to support and serve healthier patients than we can do today.”

For example, the object calls patients after certain procedures. However, due to resource restrictions, only the highest risky patients can follow. One goal for Kearns is watching with each patient, and the avatar confirmed that they clarify any sick questions and tried to follow their clinical orders. Avatar may increase to the nurse or the patient’s clinical team as needed.

“One of the health works is human touch,” said Kearns. “This is a way to make sure that this man is maximizing and optimizing the sources of touch, but also ensure that patients are well supported through health trips.”

But he still noted that the game was early. An important future step allows you to interact with digital teammate epic information and environment.

“We have done more there, we still pay a lot of attention to the adoption,” Kearns said. “It’s still a health system that is very reactive to patients’ health. We would like to reach the place where we are proactive.”