(291 days)
EVIS EXERA III COLONOVIDEOSCOPE OLYMPUS PCF-H190TL/I
This instrument is intended to be used with an Olympus video system center, light source, documentation equipment, monitor. EndoTherapy accessories (such as a biopsy forceps), and other ancillary equipment for endoscopy and endoscopic surgery.
The EVIS EXERA III COLONOVIDEOSCOPE PCF-H190TL/I is indicated for use within the lower digestive tract (including the anus, rectum, sigmoid colon, colon, and ileocecal valve).
EVIS EXERA III COLONOVIDEOSCOPE OLYMPUS PCF-HQ190L/I
This instrument is intended to be used with an Olympus video system center, endoscope position detecting unit, light source, documentation equipment, monitor, EndoTherapy accessories (such as a biopsy forceps), and other ancillary equipment for endoscopy and endoscopic surgery.
The EVIS EXERA III COLONOVIDEOSCOPE PCF-HO190L/I is indicated for use within the lower digestive tract (including the anus, rectum, sigmoid colon, colon, and ileocecal valve).
The EVIS EXERA III COLONOVIDEOSCOPE OLYMPUS PCF-H190TL/I and PCF-HQ190L/I are intended to be used with an Olympus video system center, light source, endoscope position detecting unit (for PCF-HQ190L/I only), documentation equipment, monitor, EndoTherapy accessories (such as a biopsy forceps), and other ancillary equipment for endoscopy and endoscopic surgery. The PCF-H190TL/I and PCF-HQ190L/I are indicated for use within the lower digestive tract (including the anus, rectum, sigmoid colon, colon, and ileocecal valve).
The EVIS EXERA III COLONOVIDEOSCOPE OLYMPUS PCF-H190TL/I and PCF-HQ190L/I consist of three parts: the control section, the insertion section, and the connector section. The basic principle including user interface and operation for the procedure of the PCF-H190TL/I and PCF-HQ190L/I are identical to that of the predicate devices.
Available imaging modes are listed below:
WLI
NBI
This document pertains to the clearance of the EVIS EXERA III Colonovideoscope, specifically the PCF-H190TL/I and PCF-HQ190L/I models. It is a 510(k) premarket notification, which demonstrates substantial equivalence to previously cleared devices rather than requiring a full demonstration of safety and effectiveness as in a PMA. Therefore, the "acceptance criteria" and "study that proves the device meets the acceptance criteria" are not in the context of an AI/ML device evaluating specific metrics like sensitivity/specificity for a diagnostic claim, but rather demonstrating that the new colonoscopies are as safe and effective as existing ones, with the added adjunctive feature of Narrow Band Imaging (NBI).
The clinical performance section largely focuses on the NBI feature. Here's a breakdown of the information provided, specifically addressing the points related to performance data:
The document does not detail specific acceptance criteria for a "device performance" metric in the typical sense of a diagnostic algorithm (e.g., a specific sensitivity or specificity target). Instead, the performance testing is focused on demonstrating that the new device models perform as intended and meet design specifications, and that the NBI feature is an acceptable adjunctive tool. The "acceptance criteria" can be inferred as successful completion of the listed tests in accordance with relevant standards and guidance documents.
Device Performance Summary:
The document does not present a table of acceptance criteria and reported device performance with specific quantitative metrics directly comparable to an AI/ML algorithm's diagnostic performance (e.g., sensitivity, specificity, AUC). Instead, the performance data provided are primarily non-clinical and relate to the safety and functionality of the physical endoscope, along with clinical evidence for the adjunctive NBI feature.
The "acceptance" in this context is the FDA's determination of substantial equivalence (SE) based on these tests, which implies the device's acceptable performance for its intended use.
Inferred "Acceptance Criteria" for Substantial Equivalence:
- Reprocessing Validation: Compliance with guidance for reprocessing medical devices.
- Biocompatibility: Compliance with ISO 10993-1 and FDA guidance.
- Software Verification and Validation: Compliance with FDA guidance for software in medical devices and cybersecurity.
- Electrical Safety & EMC: Compliance with ANSVAAMI ES 60601-1, IEC 60601-2-18, and IEC 60601-1-2.
- Bench Testing: Device performs as intended and meets design specifications for thermal safety, composite durability, photobiological safety, and retroflexed withdrawal/detection of hidden polyps.
- Clinical Performance (for NBI): Real-time predictions of diminutive polyp histology using NBI provide "reasonable certainty" of neoplastic vs. non-neoplastic identity, and the NICE classification helps achieve similar performance. This is presented as an adjunctive tool, not a standalone diagnostic.
Reported Device Performance (from the document):
- Reprocessing Validation: Conducted and documentation provided as recommended. (Meets inferred criteria)
- Biocompatibility Testing: Conducted in accordance with guidance, including Cytotoxicity, Intracutaneous, and Guinea Pig Maximization Sensitization Tests. (Meets inferred criteria)
- Software V&V: Conducted and documentation provided as recommended. (Meets inferred criteria)
- Electrical Safety & EMC: System complies with relevant standards. (Meets inferred criteria)
- Bench Testing: Conducted to ensure the device performs as intended and meets design specifications. (Meets inferred criteria)
- Clinical Performance (Meta-analysis for NBI): Pooled data from NBI-experienced and inexperienced endoscopists making high-confidence predictions of diminutive polyp histology in real-time show "reasonable certainty" of neoplastic vs. non-neoplastic identity. The NICE classification helps achieve similar performance.
1. Table of Acceptance Criteria and Reported Device Performance
As explained above, specific quantitative acceptance criteria and reported performance metrics (like sensitivity/specificity targets) for a diagnostic AI are not provided in this 510(k) for a colonoscope. The "acceptance criteria" are implied by compliance with various testing standards and guidance, and the "reported performance" is that these tests were conducted and the device complied.
For the NBI feature, which is the closest to an "AI-assisted" clinical claim, the performance is described qualitatively:
| Feature/Test | Acceptance Criteria (Implicit) | Reported Performance |
|---|---|---|
| NBI Clinical Meta-analysis | Demonstrate "reasonable certainty" of neoplastic vs. non-neoplastic identity for diminutive polyps when used as an adjunctive tool. | Pooled data from NBI-experienced and inexperienced endoscopists making high confidence predictions of diminutive polyp histology in real-time provided "reasonable certainty" of neoplastic vs. non-neoplastic identity. |
2. Sample size used for the test set and the data provenance
The document specifies the clinical performance data for the NBI feature came from a meta-analysis of independent studies, with the exception of two Olympus-sponsored studies. It does not provide a specific sample size for a "test set" in the context of an algorithm evaluation, nor does it specify the country of origin of the data or whether the studies were retrospective or prospective. A meta-analysis implies aggregation of various study data.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
Not applicable in the typical sense for this submission. The clinical data reviewed is a meta-analysis on NBI, where the "ground truth" for polyps (neoplastic vs. non-neoplastic) would typically be established by histopathology, which is the gold standard, not by expert consensus on imaging. The document states: "This application does not seek to show superiority or non-inferiority of NBI to the gold standard histopathology." This confirms histopathology as the ground truth.
The meta-analysis included "NBI-experienced and inexperienced endoscopists," but their role was in making the predictions themselves, not establishing the ground truth from pathology.
4. Adjudication method for the test set
Not applicable in the typical sense for a meta-analysis focused on a medical device. The "ground truth" for the NBI feature's performance in differentiating polyps is histopathology, which does not require an adjudication method among experts on test set images.
5. If a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
An MRMC study was not explicitly described in the document as a primary method for this 510(k) submission. The clinical performance data relies on a meta-analysis of existing clinical literature regarding NBI. The description of the meta-analysis does not focus on an "AI vs. without AI assistance" paradigm, but rather on the predictive accuracy of endoscopists utilizing NBI. NBI itself is an imaging mode, not an AI, though it aids human interpretation. The application is for NBI as an "adjunctive tool," suggesting human-in-the-loop performance is inherently assumed. No specific effect size of improvement is quantified in the provided text.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
No standalone performance data was provided because NBI is an imaging mode used by an endoscopist, not an autonomous AI algorithm. The device itself (the colonovideoscope) is a medical instrument requiring human operation.
7. The type of ground truth used
For the clinical meta-analysis on NBI, the type of ground truth used for polyp differentiation was histopathology. This is explicitly stated: "This application does not seek to show superiority or non-inferiority of NBI to the gold standard histopathology."
8. The sample size for the training set
This information is not applicable as this is not an AI/ML algorithm that undergoes a "training" phase. The clinical data comes from independent studies and Olympus-sponsored studies for a meta-analysis of the NBI feature.
9. How the ground truth for the training set was established
This information is not applicable as there is no "training set" in the context of an AI/ML algorithm being cleared via this 510(k) notification. The ground truth for the meta-analysis of NBI's clinical performance was histopathology.
§ 876.1500 Endoscope and accessories.
(a)
Identification. An endoscope and accessories is a device used to provide access, illumination, and allow observation or manipulation of body cavities, hollow organs, and canals. The device consists of various rigid or flexible instruments that are inserted into body spaces and may include an optical system for conveying an image to the user's eye and their accessories may assist in gaining access or increase the versatility and augment the capabilities of the devices. Examples of devices that are within this generic type of device include cleaning accessories for endoscopes, photographic accessories for endoscopes, nonpowered anoscopes, binolcular attachments for endoscopes, pocket battery boxes, flexible or rigid choledochoscopes, colonoscopes, diagnostic cystoscopes, cystourethroscopes, enteroscopes, esophagogastroduodenoscopes, rigid esophagoscopes, fiberoptic illuminators for endoscopes, incandescent endoscope lamps, biliary pancreatoscopes, proctoscopes, resectoscopes, nephroscopes, sigmoidoscopes, ureteroscopes, urethroscopes, endomagnetic retrievers, cytology brushes for endoscopes, and lubricating jelly for transurethral surgical instruments. This section does not apply to endoscopes that have specialized uses in other medical specialty areas and that are covered by classification regulations in other parts of the device classification regulations.(b)
Classification —(1)Class II (special controls). The device, when it is an endoscope disinfectant basin, which consists solely of a container that holds disinfectant and endoscopes and accessories; an endoscopic magnetic retriever intended for single use; sterile scissors for cystoscope intended for single use; a disposable, non-powered endoscopic grasping/cutting instrument intended for single use; a diagnostic incandescent light source; a fiberoptic photographic light source; a routine fiberoptic light source; an endoscopic sponge carrier; a xenon arc endoscope light source; an endoscope transformer; an LED light source; or a gastroenterology-urology endoscopic guidewire, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 876.9.(2) Class I for the photographic accessories for endoscope, miscellaneous bulb adapter for endoscope, binocular attachment for endoscope, eyepiece attachment for prescription lens, teaching attachment, inflation bulb, measuring device for panendoscope, photographic equipment for physiologic function monitor, special lens instrument for endoscope, smoke removal tube, rechargeable battery box, pocket battery box, bite block for endoscope, and cleaning brush for endoscope. The devices subject to this paragraph (b)(2) are exempt from the premarket notification procedures in subpart E of part 807of this chapter, subject to the limitations in § 876.9.