Surgical Safety's OR black box records everything; checklists skipped in 1 of 5 operations

In the 1990s, surgeon Teodor Grantcharov watched a VHS recording of his own operation and found his self-image shattered: 'My perception was that my performance was spectacular — until the moment I saw the video.' A 30-minute procedure had stretched to 90 through wrong instruments and rough dissection. His conclusion: surgeons are notoriously bad at acknowledging their own mistakes, and 'as long as the only barrier between success and failure is a human, there will be errors.'

The product he built at Surgical Safety Technologies is the operating-room equivalent of an airplane black box: panoramic cameras, microphones and anesthesia monitors record the whole OR — not just the intraoperative video that rivals like Medtronic, Johnson & Johnson and Intuitive Surgical capture — including how often the door opens and how many non-case-related conversations occur. AI helps analyse the flood. The boxes now run at almost 40 institutions in the US, Canada and Western Europe, from Mount Sinai to Duke to the Mayo Clinic.

The evidence is concrete. A 2022 study in Annals of Surgery found the surgical checklist had not been followed in a fifth of operations, likely contributing to excess infections. Black-box data showed auditory disruptions — doors, alarms, pagers — happen every minute in gynecology ORs, and a median 20 intraoperative errors occur per laparoscopic case. Hospitals use the data for fixes: Duke checked antibiotic timing; several hospitals cut the idle time between cases by flagging unexplained gaps and having analysts review the recordings.

The friction is human, not technical: some surgeons have refused to operate with black boxes in place and some systems have been sabotaged, over privacy, discipline and legal exposure. And as critic J. Dimick noted, there was not yet large peer-reviewed proof that black boxes reduce complications and save lives — the adoption case still leans on the analogy Grantcharov draws to aviation, where recording helped cut fatal accidents from 6.5 per million flights in 1970 to under 0.5 today.

An airplane-style black box for the operating room: panoramic cameras, microphones and anesthesia monitors record everything, and AI turns the flood into counts.

182|The OR is 'probably the most measured place in the hospital but also one of the most poorly captured' — errors surface only as post-hoc recollection, which underreports and flatters.

174|Capturing the whole environment, not just the patient's body, exposes systemic causes — door openings, interruptions, idle turnovers — that technical-only systems cannot see.

148|Automated counting converts blame into measurement: hospitals act on aggregate patterns like antibiotic timing rather than disciplining individuals.

Aviation supplied both design and sales pitch: recording helped cut flying's fatal accidents from 6.5 to under 0.5 per million flights, and Grantcharov predicts the same for surgery.

Systems fail where feedback depends on memory and self-interest. Instrument the environment, and improvement stops depending on who admits fault.

By June 2024 the system ran at nearly 40 institutions across North America and Western Europe, with a 'customer success' team added after early variations in how hospitals used the data. An upcoming study was scheduled to show improved checklist compliance and reduced ICU stays, reoperations, readmissions and mortality, according to Grantcharov. Refusals and sabotage at some centers, and the absence of large peer-reviewed mortality evidence, remained the open obstacles.