When it comes to surgery, there is a common procedure that most people who have undergone surgery are familiar with. First, you wait for the staff to bring you to the surgery area. Then, you wait there until your vitals are checked and the anesthesiologist visits you. Finally, you are taken to the operating room. Once the surgery is completed, it’s not uncommon to find out that the procedure took longer than expected due to miscommunication between the staff and an intern who is still learning.
Dr. John W. Cromwell, the director of the division of gastrointestinal surgery at the University of Iowa Hospitals & Clinics, wants to improve this procedure at his hospital. He has three goals: optimizing patient safety and comfort, increasing efficiency, and developing skills through team coaching. Cromwell also runs an informatics lab at the university and has a keen interest in integrating IT and digital systems with medical procedures.
While the hospital already tracks quality metrics, Cromwell believes that he needs more detailed data to gain insights into the skills of the surgical teams in the operating room (OR) – insights that would lead to important coaching moments. He envisions a new system that includes video and audio data, which would be analyzed using artificial intelligence. Without AI, the data cannot be analyzed or interpreted quickly enough.
With a setup like that, the hospital could work on improving surgical outcomes. “We know that keeping patients warm while they are under anesthesia is an important factor in reducing surgical infections, so we collect data on the core temperature of patients when they come out of the OR,” Cromwell explained. “If their core temperature is lower than we want it to be, we want to determine what factors in the OR allowed that to happen.” The team’s approach and teamwork can be objectively evaluated and scored, enabling targeted coaching on processes and procedures that could have prevented such occurrences.
In a busy hospital where surgeries happen back-to-back, ensuring safe and efficient surgeries is a major goal. That means being able to analyze “turnovers,” which refers to the time it takes to move a patient out of the OR and prepare the room for the next one. “If we can shave 10 to 15 minutes off of the turnover, by the end of the day we will have been able to get one more patient into the OR,” Cromwell said.
Cromwell’s vision started taking shape when he attended a conference in 2018 and encountered a company with a promising idea but limited technology advancement at the time. Over the next few years, that company, Caresyntax, developed its technology and approached Cromwell. Caresyntax aimed to collect audio and video data through its digital surgery platform, processed in real-time using AI inferencing models, and extract valuable insights to improve processes and facilitate coaching moments.
Caresyntax’s Approach to Surgical Data Analysis
Cromwell believed that the Caresyntax system had a high chance of success, and it was implemented at the hospital in July 2021.
To gather the necessary data for analysis, the system follows a multi-step process. It collects video and audio data during procedures using cameras and microphones in the OR. This data is then transmitted and streamed to Caresyntax’s secure cloud. In the cloud, the collected data is combined with basic patient information obtained from electronic health records. Caresyntax’s CXAdvance software adds scores based on objective measures of teamwork.
Surgeons and clinical leaders receive a link to the surgical video, allowing them to access the video-based assessment within the platform. Clinical staff can review the recorded video to identify both exemplary work and potential areas of improvement related to team dynamics. Additionally, staff can use a module called CX-Insight to analyze ways to optimize patient access, improve efficiencies, and reduce infection rates.
The effectiveness of Caresyntax’s approach relies heavily on the use of AI and automation. Timothy Lantz, president of Caresyntax, explained that their system can process both unstructured data (like video and audio) and structured data (like the information in electronic health records). This processing converts all the data into a structured format so it can be quickly analyzed.
Enhancing Data Analysis in Surgical Care
To expand the scope of data analysis, the next step involved pulling more data into the analytics chain. This data included various timestamps, which capture crucial moments such as the start and end of anesthesia and the completion of the surgical procedure. By incorporating timestamps, it is much easier for staff to analyze different phases of care during a surgical case.
“If we’re looking at how the team performed during a specimen handoff, for example, we want to see if it was labeled the right way and sent to the right lab,” Cromwell explained.
With all the necessary systems in place, it was time to put them to use. Starting from the moment a patient gives consent, every step of the surgical process is now recorded, including the patient’s check-in, transfer to the operating table, the operation itself, and the turnover process. Valuable insights are generated, leading to coaching moments. According to Cromwell, they are only scratching the surface of what the system can help the hospital achieve.
“Our goal is to come away from reviews with a list of specific actions that we need to take or incorporate into our OR education programs,” Cromwell said. To this end, Cromwell’s team is even developing simulation-based tools that provide detailed training models for leaders to guide their teams.
Caresyntax is also committed to advancing its technology. For example, the company has begun automating the real-time detection of distinct phases of surgery. Although the system could already use AI to detect these phases after the operation, being able to do so in real time can potentially offer turn-by-turn guidance during surgery, Lantz said.
About the author
Karen D. Schwartz is a technology and business writer with more than 20 years of experience. She has written on a broad range of technology topics for publications including CIO, InformationWeek, GCN, FCW, FedTech, BizTech, eWeek and Government Executive.
