Standardizing Surgical Healing: Building the Foundation for AI-Driven Care
Almost every surgical procedure starts with an incision. What happens to it afterward is one of surgery’s most common problems, and one of its least consistently measured. Infection, bleeding and wounds reopening are dangerous and costly, for patients and health systems alike. Yet despite decades of better techniques, devices and aftercare, these outcomes remain poorly reported. Definitions vary, surveillance is inconsistent, and results can’t be compared across specialties. Healthcare still lacks a common language to communicate and report surgical site outcomes effectively.
That gap matters more now than ever. AI and predictive analytics promise to show which care pathways lead to the best outcomes, but only if the outcomes themselves are recorded the same way. It matters for innovators too. As Juan Carlos “JC” Scott, Chief Strategy Officer and Executive Director, AdvaMed Accel noted at a recent AdvaMed Executive Roadshow event, three out of four emerging MedTech companies never see their ideas reach patients, held back by regulatory and coverage hurdles that come down to evidence. Without a common way to measure outcomes, that evidence is harder to build.
At the same event, held at EPAM’s Boston Seaport Innovation Space, EPAM’s Global Head of MedTech, Chris McCarthy, sat down with Nisha Johnson, Global President of Wound Closure & Healing at Johnson & Johnson MedTech. They discussed one effort to close that gap: the DISH Classification System, a clinician-built framework for grading surgical healing, and the work now underway to bring it into everyday care. The conversation offers lessons for any MedTech leader building data-driven digital solutions.
The Measurement Gap in Surgical Healing
The scale of the problem is hard to ignore. According to Johnson, one in four patients leaves the operating room with one of four unwanted outcomes: Dehiscence (a wound coming apart), Infection/Inflammation, Seroma (fluid buildup) or Hematoma (blood buildup). What’s more, she believes that figure is underreported. Part of the reason is definitional. What counts as an infection in North America may not on the other side of the world.
Johnson & Johnson has a long history with surgical innovation. In 1876, Robert Wood Johnson heard Joseph Lister make the case for sterile surgery at the World’s Fair in Philadelphia, and the company went on to mass-produce the first sterile sutures. Nearly a century and a half later, J&J MedTech is helping clinicians build a shared language for what happens after the incision is closed.
Standardized Outcomes Before AI
AI needs inputs it can trust. Once wound outcomes are classified consistently, Johnson explains, AI and predictive analytics can help define the best pathway for, say, a knee replacement: suture choice, operating room conditions, dressings and post-operative care.
Building the DISH standard took rigor. Around 100 clinicians, researchers and thought leaders (30 countries, 14 specialties) worked through multiple rounds of Delphi consensus, with input from health economics and market access experts. The authors recently published the framework in the British Journal of Surgery.
A standard clinicians build themselves is one they are far more likely to adopt, but the next test is everyday use – and that means building the technology around how hospitals already work. “We can’t create another system,” Johnson says. The goal is to capture DISH within existing workflows and electronic health records, not through a separate tool. Five pilot sites are signed up, with five more expected by the end of the year. They’ll show how the standard scales in practice and shape what the supporting technology needs to do, with the team observing real workflows and keeping the “humility to make changes as we go along.”
Beyond the Operating Room Asked about her biggest lesson, Johnson pointed to the patient rather than the technology. Picture going home after a C-section: a newborn, little sleep, and a wound that’s the last thing on your mind. When something feels off, a call to the hospital might end with “it’s normal for your incision to be sore,” and each follow-up means repeating your story to someone new. The technology to help already exists: a photo of the incision sent to a clinician through a secure portal, or a dressing that monitors temperature or electrical impedance. But even something as simple as speaking to the surgeon who treated you is hard today. What’s missing, Johnson says, is consistent communication and a shared understanding of what’s happening across the treatment pathway. Tools can only help if everyone describes healing the same way. |
Staying Focused on the Unmet Need Many of the clinicians who came to J&J as part of the DISH discovery process already had partial solutions. It’s tempting to start ideation with technology – especially those tools already showing promising results. But the team kept coming back to a harder question: who will use this beyond the few people already convinced, and how can it scale without a shared language? So J&J didn’t arrive with the answer. They brought physicians, surgeons and researchers together around a problem they shared. “It is absolutely not J&J’s work, but it is our role to convene,” Johnson says. The idea is a global standard, not something that works for 100 or even 1,000 people. |
Scaling the Standard A standard only becomes a standard if everyone can use it. DISH is vendor-neutral by design: it describes outcomes the same way regardless of whose sutures, dressings or antibiotics were used, including the many J&J doesn’t make. That mirrors J&J MedTech’s broader approach with its Polyphonic™ Digital Ecosystem, which Johnson describes as “open to all technology in the operating room,” not just J&J products. DISH also works anywhere on the body. A dehiscence behind the knee is as much a problem as one on the face, so DISH classifies the same four outcomes across every specialty and incision site. And healing matters more as care shifts. Johnson notes that hip and knee surgeons once focused mainly on choosing the prosthesis. Now, as more procedures move to ambulatory settings, they see healing as essential to getting patients moving quickly and to the prosthesis lasting. |
From Consensus to Infrastructure DISH now has what many digital health efforts lack: a clinically validated, published definition of what to measure. The next stage, as Johnson describes it, is still taking shape. The pilots start with whatever systems hospitals already use. From there, the team plans to “iterate on the technology, especially on the digital side,” with the longer-term ambition of a technology platform that can support the standard globally. That move from an agreed standard to everyday infrastructure is where many promising MedTech initiatives stall. A grading framework agreed by an expert panel still has to become something a busy clinician can record in seconds. An EHR has to store it as structured data, and an analytics team has to trust it across sites and countries. Making that happen is an engineering and design challenge as much as a clinical one. In our work across MedTech, a few things consistently separate digital tools that scale from those that stay in pilot:
Standards like DISH also lower a barrier for emerging companies. A shared outcome measure gives a new dressing, device or monitoring tool a common yardstick to prove its value, which is the kind of evidence regulators and payers look for. That’s one way the three-in-four failure figure starts to come down. What Better Data Makes Possible When J&J shared early consensus work with surgeons in Japan and South Korea, where meticulous technique is deeply embedded in training, the reaction surprised Johnson. Some surgeons had long said infection wasn’t a problem for them, but they weren’t measuring it. A common language focused on outcomes rather than blame changed that. Surgeons began talking openly about the cases that had stayed with them, like a cardiac surgeon who still remembered a sternotomy that wouldn’t heal and a patient who didn’t leave the hospital for four months. That candor is what makes better data possible. Once outcomes are recorded the same way, the data starts revealing patterns: where results vary between hospitals and regions, where the gaps in care are, and which early-stage innovations are making a difference. For the industry, that points to where new products, platforms and investment can make the most impact. The ultimate goal, as Johnson puts it, is an algorithm for best-in-class outcomes “regardless of where you are in the world.” That future depends on AI, but it starts with something far simpler: agreeing on what healing looks like. |
