Search Results

The EndoRotor® is intended for use in endoscopic procedures by a trained gastroenterologist to resect and remove tissue, not intended for biopsy, of the gastrointestinal (GI) system including post-endoscopic mucosal resection (EMR) tissue persistence with a scarred base and residual tissue from the peripheral margins following EMR.

The EndoRotor®, with the design modifications and expanded indications for use described in the current 510(k) premarket notification, will continue to be marketed as the EndoRotor®. The new EndoRotor® will be referred to as the "modified EndoRotor" in this 510(k) Summary.

The overall device description remains unchanged since initial FDA clearance. Both the predicate and modified EndoRotor® are powered resection tools 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 design modifications to the EndoRotor® described in the current 510(k) premarket notification include enhancements to improve the ease of use of the console and catheter components as well as provide flexibility to the gastroenterologist.

The provided text describes modifications to an existing device, the EndoRotor®, and does not present an AI/ML device or its associated performance studies. Therefore, I cannot extract information related to AI/ML specific acceptance criteria, and study details like sample size for test sets, data provenance, expert qualifications, adjudication methods, MRMC studies, standalone performance, training set details, or ground truth establishment.

Instead, the document details non-clinical and clinical performance testing conducted to support the substantial equivalence of the modified EndoRotor® to its predicate device (K170120) and to support expanded indications for use.

Here's the relevant information about the acceptance criteria and the studies that prove the device meets them:

1. A table of acceptance criteria and the reported device performance:

The document doesn't explicitly list acceptance criteria in a quantitative table format with corresponding numerical performance for the modified device in the same way an AI/ML device submission might. Instead, it states that:

2. Sample size used for the test set and the data provenance:

• Non-clinical (Animal Testing for Predicate Device, applicable to modified):

  • Sample Size: 6 porcine animals (4 recovery, 2 acute) for mucosal resections, totaling 124 mucosal resections. Additionally, 4 animals for design validation and usability assessment.
  • Data Provenance: Porcine animal model.

• Clinical (Investigator-led study in UK):

  • Sample Size: 19 patients.
  • Data Provenance: Retrospective, from Queens Alexandra Hospital in Portsmouth, United Kingdom.

• Clinical (Commercial procedures in Western Europe):

  • Sample Size: 78 patients.
  • Data Provenance: Real-world commercial procedures in Western Europe (Austria, Germany, The Netherlands, Switzerland, and the United Kingdom). The document implies this data is also retrospective "compiled data provides real world clinical evidence".

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• Clinical (Investigator-led study in UK):
  • Number of Experts: Not explicitly stated, but the study was conducted by "Gastroenterologists" in Queens Alexandra Hospital.
  • Qualifications: "trained gastroenterologist" as per the Indications for Use. The study authors are listed as Khandiah K, Subramanian S, Thayalasekaran S, Chedgy F, and Bhandari P, implying multiple gastroenterologists were involved.
• Clinical (Commercial procedures in Western Europe):
  • Number of Experts: Not explicitly stated, but implies "Physicians" performing the procedures.
  • Qualifications: "trained gastroenterologist" as per the Indications for Use.

4. Adjudication method for the test set:

Not explicitly mentioned for the clinical studies. Decisions appear to have been made by the treating gastroenterologists.

5. If a multi-reader, multi-case (MRMC) comparative effectiveness study was done:

No, an MRMC comparative effectiveness study was not done. This is a medical device submission, not specifically an AI/ML device, and the studies performed focused on the safety and effectiveness of the device itself rather than human reader improvement with AI assistance.

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

Not applicable, as this is a medical device (EndoRotor®), not an algorithm. The device is always under the control of a trained physician.

7. The type of ground truth used:

• Non-clinical (Animal Testing): Histopathological assessment of tissue response (favorable and clinically acceptable tissue response).
• Clinical (Investigator-led study in UK & Commercial procedures): Physician assessment of lesion presence/scarring and procedure outcomes (disease eradication, luminal preservation, avoidance of surgery, no incidences of perforation or delayed bleeding). This can be considered expert consensus/clinical outcomes data.

8. The sample size for the training set:

Not applicable. This is not an AI/ML device; therefore, there isn't a "training set" in that context. The device's design improvements were based on feedback and engineering enhancements.

9. How the ground truth for the training set was established:

Not applicable for the same reason as above.

Ask a specific question about this device

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