Search Results

The Senhance Surgical System is intended to assist in the accurate control of laparoscopic instruments for visualization and endoscopic manipulation of tissue including grasping, cutting, blunt and sharp dissection, approximation, ligation, electrocautery, suturing, mobilization and retraction in laparoscopic colorectal surgery and laparoscopic gynecological surgery. The system is indicated for adult use. It is intended for use by trained physicians in an operating room environment in accordance with the Instructions for Use.

The Senhance Surgical System is a console-based, multi-arm surgical system which enables a surgeon to remotely control surgical instrumentation during minimally invasive surgery in the lower abdomen and pelvis. The capital equipment is comprised of three main sub-systems as follows:

• Cockpit The station where the surgeon inputs information through hand and eye movements to direct the motion of the arms in the surgical field.
• Manipulator Arms Independent mechanized support arms that interface with the endoscope ● and surgical instruments. The manipulator arms produce output movements based on the instructions from the surgeon at the cockpit. The system is configurable with up to three arms.
• Node A relay unit which connects the cockpit inputs to the manipulator arms in the system as configured, and converts and transmits the video signals to the 2D/3D monitor on the cockpit console.

The document is a 510(k) summary for the Senhance Surgical System (K171120), comparing it to the predicate device, the Intuitive Surgical da Vinci Si Surgical System IS3000 (K081137). The device is an endoscopic instrument control system.

Here's an analysis of the acceptance criteria and the studies performed:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state "acceptance criteria" in a quantitative manner for specific metrics like sensitivity, specificity, or AUC, as would be typical for diagnostic AI devices. Instead, the substantial equivalence is determined by comparing procedural endpoints of the Senhance System's clinical data to published literature data of the predicate device. The general acceptance criterion seems to be that the Senhance System is "as safe and effective as the predicate device for its intended use" based on these comparisons.

Here's a summary of the performance data presented, compared implicitly to the predicate device's published literature:

2. Sample Sizes and Data Provenance

• Test Set (Clinical Data):

  • Gynecological Laparoscopic Surgery: 150 patients. Prospective, non-randomized clinical trial. The country of origin is not explicitly stated, but clinical trials for FDA submissions are often multinational or US-based.
  • Colorectal Laparoscopic Surgery: 45 patients. Retrospective chart review (referred to as Real World Evidence or RWE). The country of origin is not explicitly stated.

• Training Set: The document does not mention separate "training sets" for an AI algorithm. The Senhance Surgical System is a robotic surgical system, not a diagnostic AI device requiring a machine learning training set in the typical sense. The development of the system itself involved various engineering tests, including software verification and validation, human factors/usability testing, bench testing, and animal testing.

3. Number of Experts and Qualifications for Ground Truth (Test Set)

• The clinical studies involved investigators/surgeons who performed the procedures and assessed adverse events. However, no specific number of experts used to establish a ground truth for a test set (e.g., for image annotations or diagnoses) is mentioned because this is a surgical system, not a diagnostic AI. The "ground truth" for the clinical studies was derived from the actual surgical outcomes, investigator assessments of device-related events, and standard clinical endpoints (e.g., blood loss, hospital stay, complications, reoperation).

4. Adjudication Method (Test Set)

• The document mentions that adverse events were considered "device related according to the investigator's assessment." This suggests that the investigators (surgeons/clinicians) made the primary determination. There is no mention of a formal multi-expert adjudication method (e.g., 2+1, 3+1 consensus) for the clinical outcomes or adverse events in the clinical data section.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No MRMC comparative effectiveness study was done. This type of study is typically for diagnostic aids or image interpretation AI where multiple human readers interpret cases with and without AI assistance. The Senhance System is a surgical tool, not a diagnostic AI.

6. Standalone (Algorithm Only) Performance Study

• No standalone algorithm performance study, in the sense of a diagnostic AI operating independently without human interaction, was performed. The Senhance System is designed to be a human-in-the-loop robotic surgical assistant, where the surgeon controls the system. Its performance is intrinsically tied to human interaction.
• However, various non-clinical tests essentially establish the "standalone" engineering performance of the system components, such as:
  • Bench testing (mechanical performance, system latency, motion accuracy, force feedback, eye sensing)
  • Electrical safety and EMC testing
  • Software Verification and Validation
  • Biocompatibility testing
  • Cleaning, Disinfection, and Sterilization validation

7. Type of Ground Truth Used (Test Set)

• The "ground truth" for the clinical studies was clinical outcomes data and investigator assessments. This includes:
  • Surgical success (successful completion of procedures)
  • Intraoperative and post-operative complications
  • Adverse events (and investigator's assessment of device relatedness)
  • Blood loss, operative time, hospital length of stay
  • Conversion rates
  • Reoperation rates, readmission rates, mortality
  • Surgical margins (for colorectal cancer cases)

8. Sample Size for the Training Set

• As explained in Point 2, there isn't a traditional "training set" for an AI algorithm in this context. The system development and validation involved various engineering, preclinical (animal), and human factors studies. The sample sizes for these are:
  • Human Factors/Usability Engineering: 16 teams of 3 users (surgeon, surgical assistant, nurse) in a porcine wet lab environment.
  • Animal Testing (Pre-clinical Design Validation): 4 teams of trained subjects (surgeon, surgical assistant) performed tasks on a live porcine model.

9. How Ground Truth for Training Set was Established

• Again, no traditional "training set" ground truth. For the human factors and animal testing, the "ground truth" was established by:
  • Successful completion of use tasks in the simulated (human factors) and live porcine (animal) environments.
  • Observation by independent usability investigators to ensure use tasks were completed without preventable errors causing harm (human factors).
  • Assessment of essential user requirements and successful completion of surgical procedures (animal testing).
  • For the engineering tests (bench, electrical, software), the "ground truth" was established by compliance with established standards, specifications, and verified functionality (e.g., IEC standards, FDA guidance for software, mechanical performance specs).

Ask a specific question about this device