Most patients who undergo a combined upper and lower GI endoscopy think of it as one procedure. In clinical terms, it is two: an esophagogastroduodenoscopy examining the upper tract, followed by a colonoscopy examining the lower. The appeal of doing both in a single session is obvious. One preparation, one sedation event, and less time away from work or family. For health systems managing a specialist shortage that is projected to reach nearly 1,400 gastroenterologists by 2037, consolidating procedures wherever possible is moving from a convenience to a strategic necessity.
But combined GI endoscopy is rarely framed as a deliberate strategy for upper GI care. In most cases, it is driven by the colonoscopy. The patient is already there for colorectal cancer screening; the upper GI exam gets added on. That sequencing reflects something deeper: lower GI has decades of established quality infrastructure, with standardized metrics, outcome benchmarks, and a now-proven track record with AI-assisted detection. Upper GI less so.
When a procedure is treated as a secondary task rather than a primary clinical event, there is the risk that its quality standards could suffer. But that needn’t be the case.
The stomach is not the colon
To understand why upper GI quality lags, it helps to understand what each procedure is actually asking the clinician to do.
In colonoscopy, the task is focused: find a polyp on a relatively uniform surface. The colon has a defined path, and performance metrics reflect that. Adenoma detection rate, cecum detection, and withdrawal time are precise, measurable, and widely adopted benchmarks. AI in lower GI was built on top of this foundation, and the results are well documented. A 2024 meta-analysis of 28 randomized controlled trials involving nearly 24,000 patients found a 20% increase in adenoma detection rate and a 55% decrease in adenoma miss rate with AI-assisted colonoscopy.
Upper GI is different. Rather than hunting for a target, the clinical task is ensuring complete coverage across a complex landscape of folds, recesses, and anatomical landmarks. That gap in scrutiny has a documented cost. In 69% of missed upper GI cancer cases, the endoscopist had already recorded an abnormality at the exact site where cancer was later diagnosed. The cancer was not invisible. The area simply was not examined carefully enough.
This is why upper GI cancer miss rates remain above 8% for esophageal and gastric cancers, with some estimates reaching 11.3% across a three-year window. It is also why AI in upper GI starts with landmark verification, confirming in real time that all anatomically required areas were examined before the scope is withdrawn. Think less “spotter” and more as a sophisticated GPS: tracking whether the endoscopist passed through every required checkpoint and viewed it at the appropriate image quality, not just flagging anything unusual along the way.
The gastroenterology community is currently working to define quality indicators for upper GI that match the rigor applied to lower GI. That work is necessary and even overdue, and the societies have begun publishing guidelines for such indicators.
When lower GI casts a shadow
In combined procedures, the big risk is cognitive displacement. Lower GI tends to dominate clinical attention in a combined session, and for understandable reasons. Colonoscopy carries the weight of cancer screening guidelines, the most developed quality metrics in the specialty, and the highest procedural volume. Upper GI frequently gets completed first, quickly, before the procedure that both the specialist and patient are often primarily there for.
The data supports that concern. Research into bidirectional endoscopy sequencing has found that procedural fatigue and endoscopist preference significantly affect upper GI findings in combined sessions. Upper GI examinations are at times performed faster than clinical best practice recommends, and the time pressure of a combined session gives that tendency room to grow.
Operator specialization adds another layer. Endoscopy training in the US emphasizes colonoscopy. Many practitioners develop deeper competency in lower GI over their careers, yet combined procedures require the same specialist to perform both, regardless of where their experience is concentrated.
And the reimbursement structure reinforces rather than corrects this. US endoscopy billing is largely binary: a base payment for the diagnostic procedure, with an additional increment if at least one biopsy or resection occurs, regardless of how many findings there are or how thorough the examination was. There is no financial signal encouraging a specialist to spend extra time on careful landmark coverage in the oesophagus, the stomach or the duodenum. The system, at present, leaves real-time AI coverage flagging (and offline report) as the main objective quality check the workflow could contain.
Applying what we learned in the colon
The credibility argument for AI-assisted endoscopy no longer needs to be made from scratch. Lower GI did that work. The colonoscopy experience demonstrated that real-time clinical support, including defined quality indicators and systematic landmark coverage verification, can work together to produce documented improvements in outcomes, and that the clinical community will follow when the evidence accumulates. The institutional reluctance that made early lower GI AI adoption slow no longer applies with the same force to upper GI.
The development roadmap for upper GI AI must follow a deliberate sequence, one shaped by where the quality gaps actually are. Landmark verification comes first, establishing that all required areas were examined and creating the quality baseline that detection work depends on. From there, the focus shifts to early detection of precancerous conditions: dysplasia in Barrett’s esophagus and gastric intestinal metaplasia. These are the upper GI counterparts to adenoma detection in colonoscopy, with similarly high clinical stakes and similarly significant room to improve on current unassisted performance.
That sequence extends into operational efficiency. When AI findings are injected automatically into procedure reports rather than documented manually after each case, time between procedures shrinks and throughput increases. In a combined session, where documentation requirements are effectively doubled, those gains compound. For health system operators weighing the cost of AI adoption, this is the ROI argument: quality improvement and operational efficiency run on the same roadmap. One funds the other.
Getting the infrastructure right
Combined GI screening is already a reality for a growing share of endoscopy programs in the US and Europe. The efficiency rationale is sound. What has not kept pace, yet, is the quality framework for upper GI, and the pressure to consolidate procedures is only accelerating that gap.
The lower GI experience showed what is achievable when evidence-based metrics and AI-assisted detection are built together rather than bolted on after the fact. Upper GI is now at the beginning of that same process. The indicators are being defined, the technology is being built to them, and the clinical community has enough experience with AI to evaluate the evidence on its merits.
The Upper GI has been treated as an afterthought for quite a long time, not out of negligence, but because colonoscopy’s prevalence and infrastructure commanded the attention. The tools, the framework, and the clinical precedent now exist to change that. AI can help ensure that the quality standards are applied with sufficient discipline, and that the adoption curve does not outpace the quality framework that underpins it.
Photo: sorbetto, Getty Images
Dr. Dror Zur is a technology executive with more than 30 years of experience in the tech industry, including over 25 years in medical devices. He is the founder and CEO of Magentiq Eye, where he leads the development of AI-driven diagnostic tools for endoscopic procedures, helping physicians improve detection rates during colonoscopy screenings.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
