The discrepancy in the results of the surgery was staggering. Within a few months, Tamir Wolff diagnosed his wife and former boss with appendicitis, and the two drove to two different New York City hospitals—just miles away—for an appendectomy, a common surgical procedure where the small organ is inflamed. from the abdomen.
“The approach and treatment were very, very different, which led to different results,” says Wolf.
For his wife, there’s as seamless emergency medicine management as anyone would want – 12 hours from the emergency room door to their home, free of the inflamed organ and any major complications.
For his boss at the time, there was a horrific streak, he says – septic shock, a prolonged stay.
For Wolf, this is a confirmation of a shocking fact in medicine: that – in his words – determines where you live if You live.
He was head of the medical branch of his unit as a lieutenant commander in Shaetet 13, an Israeli special forces unit comparable to the Navy’s SEAL, whose bat-winged insignia exemplifies its fast and silent capabilities. He was awarded a medal for his service in the 2006 war in Lebanon, for treating soldiers under fire.
Theator is a platform to help surgeons improve their performance using video, similar to a professional athlete or fighter pilot – creating, arranging and applying what Theator calls “surgical intelligence.”
It is decisions that determine outcomes; The right decisions were made in his wife’s case and her operation and recovery went smoothly. Something goes wrong along the line for the boss, and a bad decision leads to another, a huddled snowball.
“These are two notable hospitals in the same city,” says Wolf. “You’d think that for something like bread and butter…like an appendectomy, things would be different.”
These hidden asymmetries inspired Wolf to take advantage of technology to level the playing field.
Wolf and Dotan Asselmann, a computer vision and machine learning expert and entrepreneur who served in Israel’s 81st unit focused on covert technology, founded Theator, a platform to help surgeons improve their performance using video, similar to a professional athlete or fighter pilot — create, arrange, and apply Theator calls it “surgical intelligence.”
“How do we ensure that no matter where a patient goes to seek treatment, they receive the best possible care for them?” says Wolf.
The story inside
Many surgeries that can be recorded, quite simply, are not. They happen and are only in the operating room – then whatever result the patient will have.
“We realized that even if there were millions – tens of millions – of actions performed under visual guidance, the vast majority of them were not captured,” says Wolf. “Not stored, not parsed.”
Left without using a wealth of valuable surgical data, it is a well-known currency among football coaches: ribbon.
Sometimes surgical videos may find their way to students as study aids, or – in individual cases especially – as presentations at conferences. But it’s not consumed, analyzed and acted upon the way a point guard, pilot or amateur golfer would, identifying missed details and trends, and figuring out how to improve performance.
What went right? what happened? why? What do I do better than my colleagues? What do I do worse? Why am I faster and slower. They have patients with better outcomes and worse outcomes – why? Did I hit the surgical procedure steps correctly? Am I doing things I don’t need to do? Am I figuring out things I’m supposed to do like safety checks? How long do my patients usually spend at the table? Can I reduce this time?
Key issue: The video is raw.
The process of explaining the procedure, and breaking it down into steps so that it can be followed more easily, is time consuming and impractical for working surgeons.
“Not only is it not practical, it is impossible,” says Ishai Levin, director of the department of gynecology at Ichilov Hospital Sourasky Medical Center in Tel Aviv.
Levine has calls day and night from his attendance for consultation and supervision of surgeries, and a list of operations to perform himself. Ichilov is a large hospital, and the Levin department alone performs about 6000 procedures per year, according to its estimates.
Without annotation, the tape is difficult to parse, and often is not.
Theator aims to allow surgeons to analyze this tape and improve their performance by automating this process.
Using AI-powered deep learning computer vision software, Theator automatically suspends the procedure and quickly provides a number of analytical measures.
“Within 30 minutes, I received a report,” says Matthew Tollefson, a Mayo Clinic urologist who uses the program. Mayo is a partner in Theator and an investor.
Tollefson receives a notification on his phone telling him that the surgery has been analyzed, outlining the steps, including unexpected events and important safety checks. He can measure this condition with his own rates or global averages.
Automated annotation is essentially pattern recognition—a place where computer vision algorithms shine, says Asselmann, co-founder and CTO.
After initially training the algorithm on huge libraries of human-tagged surgical videos, the algorithm “learns” to identify a specific object in the footage, and then can say, for a certain amount of time, that a specific action has occurred.
“In our case, it could be autopsy,” Asselmann says. “It could be something that has some critical meaning in context,” such as seeing a specific organ or unexpected bleeding.
The AI then names those actions; Part of the collaboration with Mayo, Tollefson says, is breaking down procedures into steps for the algorithm to learn and recognize. There are still some misnomers that the algorithm will mitigate with human assistance as well.
For example, when dealing with dermal cysts, AI can confuse the oily fluid inside them with pus when they are spilled, Levine says.
To help protect patient privacy, the algorithm ensures that nothing that happens on the camera is blurred outside the patient’s body, and personal details of both the patient and the surgical team are hidden. The data then deals with the users of Theator, not the company itself.
Currently, Theator is limited to urology, gynecology, bariatric and colorectal surgery, although the goal is to continue adding more types of surgeries as the algorithm is trained and improved itself.
From Intel to Insights
“We talk about technology, but technology for technology’s sake is useless,” says Wolf.
Wolff’s vision is not limited to making the Theator a surgical annotation tool – although it may certainly be aimed at teaching hospitals as a training aid first and foremost. Instead, it’s a file Analytics From the data facilitated by Theator which can be more valuable.
“Everything has to be really instant, everything has to be doable” for surgeons to be able to incorporate video analysis into their hectic schedules, says Wolf.
Levin, MD, director of gynecology in Tel Aviv, used the platform to improve surgical practices across the department.
Seeing both ureters, the plumbing that carries urine out of the kidneys, at the end of surgery is a key safety check that can help prevent ureter injuries, Levine says.
Although rare, these injuries are most commonly due to surgical procedures, and can be hidden in place, but with serious complications including urine pooling in other points of the body, kidney loss, and even death.
“Everything has to be really instant, and everything has to be doable.”
After Levine started using Theator to see if surgeons were completing a ureter integrity check, he noticed an improvement in the number of surgeons who made sure they took the step. The safety mark shifted from completion in 43% of surgeries to 67%.
“That’s a huge thing in terms of safety” — knowing that the surgery will now include a debriefing provides an important sense of accountability, Levine says.
Since starting to use the Theator, the department has also reduced hysterectomy time from an average of 100 minutes to 87 minutes, which is a significant improvement, given that less time spent on the operating table leads to better patient outcomes.
Levine compares this to the analysis that fighter pilots do. “Every time they take a flight, every time they take training flights, every time they take combat flights, they review that,” he says.
“She learns from your experiences and gets better all the time,” says Levine. “This is a very important thing if you look at the medical profession.”
Tollefson, at Mayo, envisions eventually being able to identify more and better safety measures that could improve patient outcomes by being able to analyze surgeries more easily.
“Right now, frankly, this is somewhat rudimentary,” Tolefson says. “Yes, you made it; no, you didn’t.”
Alternatively, surgeons may be able to receive feedback about the decisions they have made regarding the likelihood of outcomes; You missed the move xSo, your patient’s chance increases y.
Furthermore, Theator wants to be able to provide real-time decision support in surgeryusing all the surgical intelligence he’s gathered to provide critical feedback, input, and suggestions to surgeons while patients are at the table.
“To be able to say to the surgeons, ‘In order to improve this patient’s erectile dysfunction, you should think about cutting here, instead of here’—that’s like the big picture,” Tolefson says.
“Something that would be really cool to get to at some point.”
We love to hear from you! If you have a comment about this article or if you have advice regarding a future Freethink story, please email us at [email protected]
#train #surgeons #fighter #pilots