AnX Robotica Highlights Documented ProScan AI Workflow Performance from Submission to FDA

Separate FDA submissions report mean small-bowel capsule review-time reductions of 88.6% with ProScan-assisted reading and 24.7% with a recently cleared comparator AI-assisted workflow.

PLANO, Texas, Oct. 11, 2026 /PRNewswire/ -- AnX Robotica today highlighted the documented results of clinical performance of NaviCam^® ProScan™, its AI-assisted capsule endoscopy reading technology, following a recent U.S. Food and Drug Administration clearance of another AI-assisted capsule endoscopy platform.

In its September 25, 2026, 510(k) clearance (K254230), the FDA identifies NaviCam ProScan (DEN230027) as the primary predicate device.¹

Both technologies are designed to assist clinicians reviewing small-bowel capsule endoscopy studies. In their respective FDA submissions, both products demonstrated reductions in mean review time, although the studies were conducted independently and should not be interpreted as a head-to-head comparison.

FDA Documented Reader Study Results

For ProScan, FDA's De Novo summary reports that the prospective multicenter ARTIC study included 133 patients and found mean small-bowel reading time decreased from 33 minutes 42 seconds with standard reading to 3 minutes 50 seconds with AI-assisted reading (p<0.001). Diagnostic yield was 62.4% with standard reading and 73.7% with AI-assisted reading.²

For the recently cleared comparator, FDA's K254230 summary reports that its reader study used 111 patients and found mean review time decreased from 45.7 minutes to 34.4 minutes with AI assistance, a mean difference of 11.3 minutes. The filing also reports AI-assisted patient-level sensitivity of 92.3% and diagnostic yield of 87.6% in that study.³

"The FDA documentation demonstrates that ProScan can materially reduce capsule review time while supporting a physician-led interpretation workflow," said Tim Thomas, Vice President, Regulatory and Clinical Affairs, AnX Robotica. "As AI adoption grows, performance claims should be considered in the context of each product's cleared indication and study design. We believe the clinical evidence supporting ProScan provides a strong foundation for efficient, AI-assisted capsule endoscopy review."

Important Comparison Context

The ProScan and comparator studies were separate clinical evaluations involving different capsule systems, patient populations, study designs, readers, reference standards and validation datasets. The figures above therefore describe results reported in each product's FDA submission. Both products are intended to assist, not replace, clinician interpretation. FDA's ProScan De Novo summary states that the clinician remains responsible for assessment and that ProScan is not intended as a stand-alone diagnostic device. The comparator's clearance similarly describes its AI functions as tools to assist reviewers.⁴

About AnX Robotica

AnX Robotica develops capsule endoscopy technologies designed to support efficient, high-quality gastrointestinal imaging and interpretation.

FDA source notes

1. FDA 510(k) Summary, K254230, decision Sept. 25, 2026. Identifies NaviCam ProScan (DEN230027) as the primary predicate. https://www.accessdata.fda.gov/cdrh_docs/pdf25/K254230.pdf

2. FDA De Novo Classification Request, NaviCam ProScan, DEN230027, decision Dec. 12, 2023. ARTIC study results include diagnostic yield of 62.4% standard vs. 73.7% AI+Physician and mean small-bowel reading time of 33:42 standard vs. 3:50 AI-assisted. https://www.accessdata.fda.gov/cdrh_docs/reviews/DEN230027.pdf

3. FDA 510(k) Summary, K254230, Reader Study. Mean reading time 45.7 minutes conventional vs. 34.4 minutes AI-assisted; AI-assisted patient-level sensitivity 92.3%; diagnostic yield 87.6%. https://www.accessdata.fda.gov/cdrh_docs/pdf25/K254230.pdf 

4. FDA DEN230027 indication/limitations and K254230 device description establish that these AI functions assist reviewers and do not replace clinician judgment. https://www.accessdata.fda.gov/cdrh_docs/reviews/DEN230027.pdf

SOURCE AnX Robotica