GI Genius

DEN200055

Cosmo Artificial Intelligence - AI, LTD · granted 2021-04-09 · product code QNP · Gastroenterology-Urology

Premarket evidence — what FDA accepted

Device typehardware with ml
source quote (p.2)
The GI GENIUS™™ is an artificial intelligence/machine learning (AI/ML) device system comprised of software, hardware, and accessories that is intended for polyp detection during standard white-light colonoscopy.
Algorithmconvolutional neural network (CNN)
source quote (p.4)
AI/Detection Module: This module is responsible for identifying potential mucosal lesions. The main component consists of a convoluted neural network. ... This includes the convolutional neural network (CNN) of the AI/ML algorithm.
Adaptive (vs locked)FDA source did not state this
PCCPNo
Cybersecurity addressedYes
source quote (p.5)
Regarding the cybersecurity, the documentation included all the recommended information from the FDA guidance document “Content of Premarket Submissions for Management of Cybersecurity in Medical Devices.” This includes a threat model, cybersecurity mitigation information, a malware-free shipping plan, an upgrade plan, and other information for safeguarding the algorithms.

Validation studies (2)

Bench

n=150 cases

endpoints: object-level sensitivity; false positive rate (object-level); receiver operating characteristic (ROC); frame-level True Positive Rate (TPR); frame-level False Positive Rate (FPR)

standards: ANSI/AAMI/IEC 60601-1-2:2014, IEC 60601-1:2005 + A1:2012 (Ed. 3.1)

Prospective clinical

n=263 patients · 3 site(s)

endpoints: Adenoma Detection Rate* (ADR*); Adenomas per Colonoscopy* (APC*); Positive Percent Agreement (PPA); Adenoma Detection Rate (ADR); Adenomas per Colonoscopy (APC); Polyps per Colonoscopy (PPC); Polyp Detection Rate (PDR); Serrated Lesions per Colonoscopy (SLPC); Serrated Lesions Detection Rate (SLDR); Advanced Adenoma Detection Rate (aADR); Small Adenoma Detection Rate (sADR); Flat Adenoma Detection Rate (fADR); Proximal Adenoma Detection Rate (pADR); False Positive Rate (FPR)

Reported performance (6 observations)

sensitivity0.8196CI [77.35%; 85.97%]
source quote (p.10)
Sensitivity > 0 ms: 81.96% [77.35%; 85.97%]
aurocas written: “auc0.787CI [0.755-0.817]
source quote (p.14)
AUC=0.787 [95% C.I.:0.755-0.817]
detection_rateas written: “Adenoma Detection Rate* (ADR*) (GI Genius)0.551CI [44.0; 65.8]
source quote (p.19)
Adenoma Detection Rate (ADR*) (adjusted estimate, %) [95% C.I.] GI Genius (136 subjects): 55.1 [44.0; 65.8]
detection_rateas written: “Adenoma Detection Rate* (ADR*) (Standard Colonoscopy)0.42CI [31.3; 53.4]
source quote (p.19)
Adenoma Detection Rate (ADR*) (adjusted estimate, %) [95% C.I.] Standard colonoscopy (127 subjects): 42.0 [31.3; 53.4]
agreement_kappaas written: “Positive Percent Agreement (PPA) (GI Genius)0.621CI [43.4; 77.8]
source quote (p.21)
Positive Percent Agreement (PPA) (%) [95% C.I.] GI Genius: 62.1 [43.4; 77.8]
agreement_kappaas written: “Positive Percent Agreement (PPA) (Standard Colonoscopy)0.652CI [46.0; 80.4]
source quote (p.21)
Positive Percent Agreement (PPA) (%) [95% C.I.] Standard Colonoscopy: 65.2 [46.0; 80.4]

Each value carries its own analysis unit and task — never compare or pool across devices. Source: De Novo decision summary PDF.

Predicate network

Postmarket — what happened after clearance

Not yet tracked — the weekly postmarket refresh hasn't snapshotted this device.

Reimbursement — how devices like this got paid

Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).

Applicable FDA guidance — what the submission is measured against

FDA guidance documents and guiding principles applicable to De Novo AI/ML devices in the Gastroenterology-Urology panel. A curated reference index, not legal or regulatory advice — each item states its own status, and a draft is never binding.

Applicability is derived from the device's FDA advisory panel and pathway — cross-cutting guidances apply to every AI/ML device; panel-specific ones are flagged. Titles, dates, and links verified against fda.gov as of July 2026.

Constat Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: constat.dev/precedent/device/DEN200055