Data scientists design OR Black Box to monitor operating rooms and streamline surgeries

Two Columbia Data Science Institute (DSI) graduates work for the startup Surgical Safety Technologies (SST), whose data-driven technology is helping hospitals become more efficient, streamlined and safer.

Peter Grantcharov and Manksh Gupta are data scientists at SST, whose A.I. platform records data from cameras, microphones and sensors that are installed in operating rooms. The platform, the "OR Black Box," captures data that identify inefficiencies and errors in operating rooms that, in turn, can help hospitals streamline surgical procedures. It's similar to the black boxes used in airplanes, but the OR Black Box records a gallery view of the entire operating room, including conversations between nurses and doctors, distractions, and the room's temperature and decibel levels.

The black box can even capture data about what is happening inside a patient's body, for instance, generating a detailed data log of laparoscopic surgery.

As for the experience of working at SST, so far it has been a great ride. I'm developing solutions to problems that can save actual lives in the hospitals we serve. It's an amazing and rewarding experience."

Manksh Gupta, Data Scientist at SST

Gupta graduated from the Data Science Institute in 2019.

He and Grantcharov code and analyze the data using techniques they learned in the DSI master's program--A.I. models, computer-vision algorithms, and machine learning. They mine the data for variables in operating rooms such as the surgical instruments being used, staffing levels and inefficiencies. The two also identify factors that ensure surgeries continue smoothly, such as when surgical team members delegate tasks or need to have backup instruments prepared. Their data analysis captures entire patterns of behavior and workflow in operating rooms. SST uses their analysis to determine best practices for hospitals, helping administrators adjust to the unprecedented demand for resources as a result of the Covid-19 pandemic.

"SST engineered the OR Black Box platform to radically improve intraoperative safety and culture through deep analysis of intraoperative decisions and performance," says Grantcharov, who graduated in 2020. "Our mission is to extract as much valuable information that we can from all of our pre-, intra-, and post-operative data streams."

SST has received major recognition for its data-intensive technology. The OR Black Box was named one of Time Magazine's Best 100 Inventions for 2019. It has been installed in hospitals in Europe, Canada, and launched recently in the U.S. It's being used in Long Island Jewish Medical Center, which is part of Northwell Health, and will soon be installed in major medical centers including Duke and UT Southwestern, Grantcharov says.

Dr. Louis Kavoussi, a surgeon at Long Island Jewish Medical Center who uses the OR Black Box in his operating room, says the platform will make operating rooms more efficient. "There's no question that this [technology] is going to improve the quality for patients," Kavoussi told CBS news. "Who should or should not be coming in and out of the OR? Should telephone calls come in, should there be music in the operating room? We all think as surgeons we are perfect, but in reality we aren't, and this is an opportunity to use artificial intelligence to analyze what's going on in the operating room."

Grantcharov and Gupta also use techniques they learned at DSI to help hospitals optimize certain surgical procedures. They are, for instance, developing a model to optimize the use of surgical trays--a preselected set of instruments that surgeons use during surgery. Preparing the trays and sterilizing the instruments are expensive, and there is "no mathematical basis for the selection of instruments that go into these trays," says Grantcharov.

The two apply computer vision to determine which surgical instruments were used in a surgical procedure, and create predictive models to show which instruments are required for different surgeries. This data-driven approach is helping hospitals become more streamlined and efficient.

"Sub-optimal surgical tray configurations result in hospitals wasting tens of thousands of dollars per operating room a year," adds Grantcharov, "so we're developing a data-driven solution to reduce medical waste and the costs associated with it."

Gupta also enjoys using data science to enhance healthcare. While he was a student at DSI, he wanted to work for a big tech company. But he soon realized he could have a more-profound impact working for a medical startup that uses data science for good--in this case helping hospitals become safer and more efficient.

"Here at Surgical Safety Technologies, Peter [Grantcharov] and I work on data-driven projects that make operating rooms more efficient and safer for all of us," Gupta says. "To hear that the A.I. products we're developing improves healthcare and even saves lives is immensely rewarding, and we hear that every day!"