Search Results

The HERMES® O.R. Control Center and Port Expander is indicated for use with Stryker Endoscopy 882 Camera, Stryker Quantum 5000 Light Source, Stryker SE5 Shaver, WOM 20L Insufflator System, WOM 2.0L Arthroscopy Pump, Stryker Total Performance System, Berchtold Surgical Lights, Steris Amsco Table Model SP3085, Steris Amsco Table Model SP3085, AESOP®R (HERMES-Ready™), Valleylab Force FX™ Electrosurgical Unit, Smith & Nephew Dyonics® Access 15 Arthroscopic Fluid Irrigation System, Smith & Nephew Dyonics® Vision 635 Digital Image Management System, Skytron Stellar Series O.R. Lights, S&N Dyonics® Power Shaver (K030240) and the Smith & Nephew InteliJET Arthroscopic Fluid Management System. It can be used in general laparoscopy. nasopharyngoscopy, ear endoscopy, and sinuscopy where a laparoscope/endoscope is indicated for use. A few examples of the more common endoscopic surgeries are laparoscopic cholecystectomy, laparoscopic hernia repair, appendectorny, laparoscopic pelvic lymph node laparoscopic dissection. laparoscopically assisted hysterectomy, laparoscopic & thoracoscopic anterior spinal fusion, decompression fixation, wedge resection, lung biopsy, pleural biopsy, dorsal sympathectomy, pleurodesis, internal mammary artery dissection for coronary artery bypass, coronary artery bypass grafting where endoscopic visualization in indicated and examination of the evacuated cardiac chamber during performance of valve replacement.

The HERMES O.R. Control Center is a computer-driven system whose basic function is to offer the surgeon the additional option of voice control for ancillary devices. The intent of the HERMES O.R. Control Center is to allow for simplified and more direct control of medical device settings by the physician, thereby eliminating the necessity of using the various interfaces existing on ancillary devices, or relying on verbal communications between the surgeon and other personnel in the operating room in order to adjust the surgical equipment.

The HERMES® O.R. Control Center is a device that allows for voice control of ancillary medical devices during surgical procedures. The 510(k) summary provides information on the device's adherence to various engineering and safety standards, but it does not include a study proving device performance against specific clinical acceptance criteria in terms of AI accuracy or effectiveness.

Here's a breakdown of the requested information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The document lists various standards the device was tested against. These are primarily safety, electrical, and software validation standards, not performance metrics related to diagnostic accuracy or clinical outcomes often seen in AI/ML medical devices. Therefore, the "reported device performance" in relation to clinical acceptance criteria is not applicable or provided in this context.

Note: This submission primarily focuses on electrical safety, electromagnetic compatibility, and software validation for a device that provides voice control functionality. It is not an AI/ML device in the modern sense that interprets data for diagnostic or predictive purposes, and therefore, typical AI acceptance criteria like sensitivity, specificity, or AUC based on clinical data are not present.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

The document does not describe a clinical "test set" for performance evaluation in the context of diagnostic or predictive AI. The testing mentioned refers to technical standards compliance (functional, environmental, software). Therefore, information on sample size, data provenance, or retrospective/prospective nature of a clinical test set is not applicable here.

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

As there is no clinical "test set" for evaluating diagnostic or predictive performance, there is no information on experts establishing ground truth. The device's function is voice control, and ground truth would relate to the accuracy of voice command recognition and execution, which would be covered under functional testing (VA-24772).

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Not applicable, as no clinical test set for diagnostic/predictive performance is described.

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

Not applicable. The HERMES O.R. Control Center is a voice-controlled interface for surgical equipment, not an AI system designed to assist human "readers" (e.g., radiologists interpreting images). The purpose is to streamline control, not to improve diagnostic interpretation.

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

The device itself is a "human-in-the-loop" device by its very nature, as it responds to voice commands from a surgeon. Therefore, a standalone (algorithm only) performance, in the context of interpretation, is not applicable. Its standalone performance would be its ability to accurately recognize and execute commanded actions, which would fall under functional testing.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

For the functional aspects of the device, the ground truth would likely be the correct understanding and execution of voice commands by the system, as verified during "CMI System Functional Testing" (VA-24772) and "CMI Software Verification and Validation" (CP-15345). This is not a clinical ground truth like pathology or outcomes data, but rather a technical ground truth related to functional correctness.

8. The sample size for the training set

This device, in this submission, is not described as an AI/ML device that undergoes "training" on a dataset in the modern sense (e.g., for image recognition or predictive analytics). It's a control system with voice recognition. The "training" for such a system typically involves developing and refining the voice recognition algorithms against a collection of voice commands, but the document does not specify a training set size or methodology.

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

Given the absence of a described "training set" as understood in current AI/ML contexts, there is no information on how ground truth for a training set was established. For voice recognition, ground truth would usually involve correctly transcribed speech paired with the associated intended action, but these details are not provided in the 510(k) summary.

Ask a specific question about this device