Search Results

Ask a specific question about this device

The iotaSOFT™ Insertion System is intended to aid the surgeon in placement of cochlear implant electrode arrays into a radiographically normal cochlea by controlling the speed of implant insertion. The iotaSOFT Insertion System is intended for use in cochlear implant patients ages 12 years and older during cochlear implant procedures using either a round window or cochleostomy approach.

The iotaSOFT™ Insertion System is designed to assist the surgeon during cochlear implantation by controlling electrode array insertion speed. The system consists of a single-use, sterile drive unit (iotaSOFT™ DRIVE Unit) connected to a reusable, non-sterile touch screen control console and footpedal (iotaSOFT Controller and Accessories) (Figure 1).

Device Acceptance Criteria and Performance Study: iotaSOFT™ Insertion System

The iotaSOFT™ Insertion System is a powered insertion device designed to aid surgeons in placing cochlear implant electrode arrays. The acceptance criteria for this device and the studies conducted to prove it meets these criteria are detailed below, synthesizing information from the provided document.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the iotaSOFT™ Insertion System are primarily outlined in the "SPECIAL CONTROLS" section and implicitly through the "SUMMARY OF NONCLINICAL/BENCH STUDIES" and "SUMMARY OF CLINICAL TESTING" sections. The reported performance is extracted from the respective study summaries.

2. Sample Sizes and Data Provenance

The document describes several studies with varying sample sizes and data provenance:

• Bench Performance Testing (Device Characteristics):

  • Sample Size: "All samples tested." (Specific number of devices/arrays not quantified, but implied to be sufficient for statistical confidence based on sterilization/accelerated aging protocols).
  • Data Provenance: Non-clinical, bench testing.

• Synthetic Cochlea Insertion Force Comparison (Surgeon Bench Comparison Testing):

  • Sample Size: Devices "having undergone sterilization." (Specific number not quantified). Involves "multiple surgeons" (number not specified).
  • Data Provenance: Non-clinical, bench testing using synthetic cochlea models.

• Cadaveric Comparison Testing:

  • Sample Size: 16 cadaver heads (1 cadaver per surgeon).
  • Data Provenance: Non-clinical, cadaveric study.
  • Country of Origin: Not specified but common for such studies to be conducted in the US.
  • Retrospective/Prospective: Prospective, as it involves active experimentation.

• Usability Testing:

  • Sample Size: 16 Surgeon/Nurse teams, 16 cadavers (1 cadaver per team).
  • Data Provenance: Non-clinical, simulated use environment in cadavers.
  • Country of Origin: Not specified.
  • Retrospective/Prospective: Prospective.

• Clinical Testing:

  • Sample Size: 21 subjects.
  • Data Provenance: Clinical study.
  • Country of Origin: Not specified, but generally US-based for FDA de novo.
  • Retrospective/Prospective: Prospective.

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

For the cadaveric comparison testing and usability testing, the "experts" are the surgeons involved.

• Cadaveric Comparison Testing:

  • Number of Experts: 16 surgeons.
  • Qualifications: Not explicitly stated for each of the 16, but generally implied to be experienced CI surgeons. The usability testing section mentions "Surgeon CI experience level ranged from 1 to 30+ years: approximately equal numbers of surgeons with less than 10 years, 10 to 20 years, and greater than 20 years. Between 10 and 180 implantations were performed manually in the year preceding this study, across all surgeons." This provides a strong indication of their expertise.
  • Ground Truth Establishment: Array intracochlear position determined via 3D x-ray. This serves as objective measurement for array placement.

• Usability Testing:

  • Number of Experts: 16 Surgeon/Nurse teams.
  • Qualifications: Same as above for surgeons. ("Surgeon CI experience level ranged from 1 to 30+ years: approximately equal numbers of surgeons with less than 10 years, 10 to 20 years, and greater than 20 years. Between 10 and 180 implantations were performed manually in the year preceding this study, across all surgeons.")
  • Ground Truth Establishment: "Was task successful? Y/N responses," use errors noted, user comments, Likert scale responses. This is a subjective assessment of usability and safety by the users themselves.

For the clinical testing, the assessment is based on clinical outcomes relevant to device safety and function.

• Number of Experts: 3 surgeons.
• Qualifications: "The principal investigator (PI) of the study... a neurotologist with nearly 40 years experience who implanted ~75 CIs manually in the year prior to this study." The other two surgeons "implanted 25 to 30 CIs each in their careers prior to commencement of this study." This establishes high levels of expertise.
• Ground Truth Establishment: Objective clinical measurements (electrode impedance, NRT measurements, x-ray imaging) and observed adverse events.

For bench performance testing, the ground truth is established by objective engineering measurements (e.g., speed, force values) using calibrated equipment.

4. Adjudication Method for the Test Set

• Cadaveric Comparison Testing: The method for 3D x-ray determination of array position is not detailed in terms of adjudication. It appears to be an objective measurement, not requiring expert consensus adjudication in the typical sense of radiological review.
• Usability Testing: Adjudication is implicitly through observation of tasks, noting of use errors, and direct feedback from the surgeon/nurse teams. No explicit multi-rater adjudication process is described for this feedback.
• Clinical Testing: Clinical outcomes are observed and reported. While medical staff would monitor these, a formal adjudication panel (e.g., for adverse events) is not explicitly detailed but standard practice for clinical trials. The outcome (normal impedance, NRT presence, satisfactory x-ray, expected AEs) suggests general medical consensus for interpreting these results.

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

No explicit MRMC comparative effectiveness study, as typically seen in AI diagnostic imaging (comparing human readers with vs. without AI assistance), was performed. The studies are more focused on validating the device's performance compared to manual insertion or evaluating its safe and effective use.

• Cadaveric Comparison: This study compares device-assisted insertions to manual insertions, which is a form of comparative effectiveness, but it's human + device vs. human, not human reader with AI vs. human reader without AI assistance in interpretation. The outcomes are about insertion quality, not diagnostic accuracy.
  • Effect Size: Not applicable in the context of an MRMC diagnostic study. The "effect" here is the device's impact on insertion parameters and placement. The device assisted in achieving similar array placement outcomes as manual insertion, while showing lower maximum insertion force and lower insertion force variation.

6. Standalone Performance

The "Bench Performance Testing" section details the standalone performance of the algorithm/device (meaning, the device's mechanical and software functions) without human-in-the-loop performance influencing the intrinsic numerical output of speed, force, etc. This shows whether the device itself can meet its operational specifications.

• Insertion speed verification within 20% of set rates (0.1 and 1 mm/s).
• Maximum insertion force <300mN.
• Slip force >80mN.
• Maximum drive wheel pinch force <5.6N.
• Drive head removal force <100mN.

These tests evaluate the device's inherent mechanical and control capabilities.

7. Type of Ground Truth Used

Different types of ground truth were used depending on the study:

• Bench Performance Testing: Objective engineering measurements (e.g., physical force transducers, speed sensors) and device specifications.
• Synthetic Cochlea Insertion Force Comparison: Objective measurements of force (mN) and force variation (mN/sec) from the synthetic models.
• Cadaveric Comparison Testing: Objective anatomical imaging (3D x-ray) to determine intracochlear position, translocations, tip fold-over, and insertion angle.
• Usability Testing: User-reported success/failure, observed use errors, and subjective Likert scale responses.
• Clinical Testing: Objective clinical outcomes (electrode impedance, neural response telemetry (NRT), post-insertion x-ray imaging findings) and observed adverse events.

8. Sample Size for the Training Set

The document does not describe a separate "training set" in the context of machine learning. The iotaSOFT Insertion System is a robotic/mechanical device with sophisticated software, but there is no indication that it uses machine learning algorithms that require a specific "training set" in the way a diagnostic AI model would. The software and firmware are verified and validated, which is a different process from training a machine learning model.

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

Since there is no explicit machine learning training set mentioned, this question is not applicable. The device's operational parameters are based on engineering design and validated through bench, cadaveric, and clinical testing, not "training" in the ML sense.

Ask a specific question about this device