FDA Clears Neptune Medical's Triton Robot, a 54.2% Detection Rate
Hardware / news
FDA Clears Neptune Medical's Triton Robot, a 54.2% Detection Rate
The Triton 1 system reached the cecum in every one of 50 trial procedures, but Neptune still won't say what it costs or when a hospital can actually buy one.
Neptune Medical said Sept. 16 that the Food and Drug Administration granted 510(k) clearance to its Triton 1 Robotic Endoscopy System, covering diagnostic and therapeutic procedures throughout the colon, including endoscopic mucosal resection and submucosal dissection, the harder removals a hand-held scope struggles to hold steady for. The clearance rests on a 50-patient trial rather than a marketing claim, and the number that matters sits in the third paragraph of Neptune's announcement, not the headline.
In the CARE 1 study, presented at Digestive Disease Week in May 2026 and cited in Neptune's clearance announcement, Triton reached the cecum in 100 percent of 50 procedures and detected adenomas at a 54.2 percent rate, above the 35 percent floor the American College of Gastroenterology sets as its quality benchmark. Zero patients had an adverse event, and none needed conversion to a standard manual colonoscope, according to Neptune and to MedTech Dive, which covered the clearance separately on Sept. 18.
The more interesting figure is aimed at the endoscopist rather than the patient. Neptune reported a 67 percent reduction in physician workload on the NASA Task Load Index, the standard instrument for measuring physical and mental strain during a procedure, against manual colonoscopy; no patient required repositioning during the trial, which is the detail that decides whether a robotic system saves a clinic time or just adds a setup step. Dr. Jason Samarasena, chief of the division of gastroenterology and hepatology at the University of California, Irvine and a CARE 1 investigator, said Triton "showed us the possibility of one system that can screen and treat anywhere in the colon, efficiently, thoroughly, and with minimal burden."
What Neptune has already sold once
Triton is not Neptune's first cleared device. Pathfinder, the company's steerable overtube for difficult colonoscopies, is its original FDA-cleared product and has been used in more than 15,000 patients across what Neptune calls a growing number of leading U.S. hospitals, giving Triton a distribution channel a first-time device maker would not have. Neptune, based in Burlingame, California, raised $97 million in a 2024 Series D round that included Olympus, the endoscopy incumbent whose scopes Triton is built to steer, and Sonder Capital, co-founded by Fred Moll, the surgeon credited as the father of medical robotics who built the da Vinci surgical robot and now chairs Neptune's board. That lineage matters more than the release does: the company Moll founded to build da Vinci has mostly left the lower gastrointestinal tract to newer entrants like Neptune.
What Neptune still hasn't said
As of Sept. 19, Neptune's own site still carries the standard pre-clearance line that Triton "is not available for sale within the United States, the European Union or the rest of the world," and the clearance announcement carried no price, no launch date and no count of the first hospitals that will install one. Every number in the CARE 1 results is Neptune's own; a 50-patient single-arm trial run by the device's own investigators is not the same evidence as a head-to-head trial against an Olympus or a Boston Scientific scope in an average gastroenterologist's hands, and no such trial exists yet.
Neptune's clearance follows two other autonomous-procedure robots into the market within a month: Vitestro's Aletta blood-draw device, which the FDA authorized Aug. 19 through its De Novo pathway, and Agility Robotics' Digit 5, a warehouse humanoid Agility priced at $200,000, or $8,500 a month. None of the three businesses are related, but all three now face the question Neptune has not answered: what a hospital or a warehouse actually pays once the pilot phase ends.
What would change that reading is a multi-center trial run against a standard scope rather than a historical detection-rate threshold, which Neptune has not announced. Until then, 54.2 percent describes what Triton did in the hands of the investigators who helped design it, not what it will do in a community endoscopy suite.
| CARE 1 result (n=50) | Triton 1 | Comparison |
|---|---|---|
| Cecal intubation rate | 100% | — |
| Adenoma detection rate | 54.2% | 35% ACG quality floor |
| Adverse events | 0 | — |
| Physician workload (NASA-TLX) | -67% | vs. manual colonoscopy |
Sources
More in Hardware
- 01Nvidia Ships a Billion RISC-V Cores, Names None of ThemThe chipmaker's own Linux driver documentation for the GPU System Processor, which runs on four RISC-V cores, never uses the term, even as Nvidia says it shipped about a billion such cores in 2024.
- 02Qualcomm's Adreno X2 Claims 2.3x Gains, Omits the BaselineIndependent testing by SemiAccurate and Chips and Cheese confirms real graphics gains but finds compute regressions and missing methodology that Qualcomm's own slides do not mention.
- 03PrismML's Ternary Bonsai 2 27B Hides a 75% ScoreThe 98.2% headline retention figure covers 20 benchmarks; on the two that test multi-step tool use, the compressed model keeps only three-quarters of its baseline score.
- 04OpenAI's Jalapeño Chip Claims Hide an Unequal TestEE Times found that OpenAI's efficiency numbers for its Broadcom-built inference chip skip an optimization rival systems are usually measured with, one worth another 3 to 5 times on its own.