Search Results

The EndoRotor is intended for use in endoscopic procedures by a trained gastroenterologist to resect and remove residual tissue from the peripheral margins following EMR (Endoscopic Mucosal Resection).

The EndoRotor is a powered resection tool consisting of a power console, foot control, specimen trap with pre-loaded filter, and a single-use resection catheter that is inserted into the working channel of a compatible endoscope. The catheter consists of an outer cylindrical cannula attached to a braided catheter and a second inner cylindrical cannula positioned inside the lumen of the outer cannula. The inner tube has blades cut into the distal end, which is oriented adjacent to the distal end of the outer tube. A torque coil, located inside the braid, is attached directly to the inner tube, and provides rotation to the cutting tool when actuated by a foot pedal controlled drive motor. The inner tube rotates relative to the outer tube along its longitudinal axis to simultaneously cut tissue at the distal end only when the user actuates suction aspirating tissue through the lumen. The specimen is collected by a micron filter that is positioned in the flow path (in the single use specimen trap) that is mounted in a dedicated slot on the console.

The provided text does not contain typical acceptance criteria for an AI/ML medical device, nor does it describe a study specifically designed to prove that the device meets such criteria in the context of AI/ML performance (e.g., diagnostic accuracy, sensitivity, specificity, or human-in-the-loop improvement). Instead, it describes a device for endoscopic tissue removal (EndoRotor) and the non-clinical and pre-clinical testing performed to establish its safety and effectiveness for FDA 510(k) clearance based on substantial equivalence to a predicate device.

The study presented is a pre-clinical animal study and non-clinical performance testing for a mechanical device, not an AI/ML algorithm. Therefore, many of the requested elements pertaining to AI/ML acceptance criteria, ground truth, expert adjudication, MRMC studies, and training sets are not applicable to the provided document.

However, I can extract information related to the device's performance as presented in the safety and effectiveness studies.

Here's an interpretation based on the provided document, addressing the closest relevant points:

Device: EndoRotor (Mechanical Tissue Resection Device)
Regulatory Submission: 510(k) Premarket Notification (K170120)

1. Table of Acceptance Criteria and Reported Device Performance

The document does not specify formal, quantitative acceptance criteria for device performance in the same way an AI/ML device would have for accuracy metrics. Instead, the "acceptance criteria" are implied by successful completion of mechanical performance tests, biocompatibility, sterilization, shelf-life testing, and acceptable safety outcomes in an animal study to demonstrate substantial equivalence to the predicate device.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set (Animal Study):
  • Sample Size: 6 porcine animals (4 recovery, 2 acute). A total of 124 mucosal resections created.
  • Data Provenance: Not explicitly stated, but typically pre-clinical animal studies are conducted in a controlled lab environment. No country of origin for the data is specified. The study was performed under Good Laboratory Practice (GLP), indicating a controlled, high-quality pre-clinical study.
  • Retrospective/Prospective: Prospective (experimental study in animals).

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

• Ground Truth Establishment: The ground truth for the animal study (e.g., bleeding scores, perforations, deepest affected layer) was established through direct observation during the procedures, necropsy, and microscopic evaluation of resected tissue.
• Number of Experts: Not specified. Pathologists would typically be involved in the microscopic evaluation, and veterinary surgeons in the animal procedures, but their number and specific qualifications are not detailed.

4. Adjudication Method for the Test Set

• Not specified. This is often not explicitly detailed for pre-clinical animal studies unless there are subjective outcome assessments that require multiple observers. The listed outcomes (bleeding scores, perforations, histological depth) typically follow defined protocols.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done

• No. An MRMC study is relevant for diagnostic imaging devices where human readers interpret medical images with or without AI assistance. This document describes a mechanical surgical device and its pre-clinical testing, not a diagnostic AI/ML algorithm.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

• Not applicable. This is not an AI/ML algorithm.

7. The Type of Ground Truth Used

• For Animal Study (Safety/Physical Performance): Direct physiological outcomes, surgical observations, and histological examination of tissue (e.g., bleeding, perforation, depth of resection).

8. The Sample Size for the Training Set

• Not applicable. This is a mechanical device, not an AI/ML algorithm that requires a "training set."

9. How the Ground Truth for the Training Set was Established

• Not applicable. (See #8)

Summary Points:

The provided document describes the regulatory submission for a mechanical medical device (EndoRotor) and its supporting safety and performance data. The "acceptance criteria" and "studies" are geared toward demonstrating mechanical safety, functionality, and biological compatibility, primarily through non-clinical bench testing and a pre-clinical animal study. It does not involve an AI/ML component, and thus concepts like training/test sets for AI, expert adjudication of AI outputs, or MRMC studies for AI assistance are not relevant to this specific premarket notification.

Ask a specific question about this device