K Number
DEN230012
Date Cleared
2023-07-14

(148 days)

Product Code
Regulation Number
888.4560
Type
Direct
Panel
OR
Reference & Predicate Devices
N/A
Predicate For
N/A
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The Ruthless Spine RJB device is intended to measure the angle of surgical instruments in two planes relative to a vertical plumb line in line with gravity. It is indicated for use during lumbosacral pedicle screw implantation in conjunction with applicable spinal instrumentation and as an adjunct to fluoroscopy or intraoperative x-ray. The RJB device is not intended to replace a surgeon's judgment and has not undergone clinical evaluation. No clinical benefit has been demonstrated or is claimed.

Device Description

The Ruthless Spine RJB device ("RJB device") is an intraoperative surgical angle measurement guide that attaches to surgical instruments to measure the instrument relative to a vertical plumb line in line with gravity. The device can measure the axial and sagittal angles relative to gravity. The RJB system only provides measurements for angles in two planes relative to the vertical gravitational plumb line. As such. the RJB device does not provide surgical assistance, guidance, or navigation against patient anatomy. The RJB device is not intended to replace a surgeon's clinical judgement and has not undergone clinical evaluation. No clinical benefit has been demonstrated or is claimed.

The RJB device is provided sterile for single use and utilizes Bluetooth Low Energy (BLE) to connect to a tablet computer and display the angle measurements via the RJB Application (App). A set of handles and instruments compatible with the RJB are provided with the device for use in lumbosacral pedicle screw placement.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the studies conducted for the Ruthless Spine RJB device, based on the provided text:

Acceptance Criteria and Reported Device Performance

Test CategoryAcceptance CriteriaReported Device Performance
Angle Measurement AccuracySystem-Level Accuracy: Not explicitly stated as a strict numerical threshold in the "Angle Measurement Accuracy" section, but implied by passing all relevant portions and meeting "the acceptance criteria." The overall benefit/risk section mentions an acceptable system accuracy of ±1º.Axial Orientation: Average Difference Between RJB and Protractor Measurement: $0.36 \pm 0.32^\circ$
Sagittal Orientation: Average Difference Between RJB and Protractor Measurement: $0.74 \pm 0.47^\circ$
Battery Life TestingMinimum 8-hour usage life.All devices demonstrated a battery life of 8 hours, meeting the acceptance criterion.
Procedural Accuracy TestingComposite Error: None of the composite errors exceeded 2.6 degrees (acceptance criteria: ±3º).
Fluoroscopic Repeatability: Largest fluoroscopic image repeatability deviation was 2.7 degrees (acceptance criteria: ±3º).Composite Error: The histogram shows that composite errors are mostly concentrated between 0.2 and 1.4 degrees, with a peak around 0.8 degrees. None of the composite errors exceeded 2.6 degrees.
Fluoroscopic Repeatability: The largest fluoroscopic image repeatability deviation was 2.7 degrees.
Usability TestingAll responses to the Usability Test Questionnaire should indicate acceptable usability, with user needs met and no need for device design or labeling changes. Specific criteria like "Avg Score/Summary" of 5 for certain items (e.g., secure fit, easy pairing) and "Yes" for a high percentage of participants for others (e.g., RJB displayed, Hold/Offset functional).Overall: No feedback warranted device design or labeling changes. User needs are met with the current device design and labeling.
Examples from Table 2:
  • RJB device fit securely: 5 (Avg Score)
  • Battery pull tab easy to remove: 5 (Avg Score)
  • Easy to pair: 5 (Avg Score)
  • Four RJB devices able to pair: Yes: 13, No: 1
  • Angles updated in real-time: Yes: 14, No: 1
  • Accuracy of RJB is ±3.5 degrees: 4.9 (Avg Score)
  • Well-integrated features: 4.9 (Avg Score)
  • Most people would learn to use quickly: 4.7 (Avg Score) |
    | General Special Controls | - Non-clinical performance data demonstrates device performs as intended.
  • Usability testing demonstrates correct use based on IFU.
  • Biocompatibility demonstrated.
  • Sterility and shelf life supported by performance testing.
  • Software V&V and hazard analysis performed.
  • Electrical safety, EMC, wireless coexistence demonstrated.
  • Labeling includes required information. | The document states that all these special controls were met through various tests (Biocompatibility, EMC/Electrical Safety, Software V&V, Packaging/Sterilization/Shelf Life) and through the content of the device labeling. The "CONCLUSION" states that the De Novo request is granted. |

Study Details

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

  • Angle Measurement Accuracy:
    • Test Set Sample Size: Not explicitly stated how many individual measurements were taken, but the results ($0.36 \pm 0.32^\circ$ for axial, $0.74 \pm 0.47^\circ$ for sagittal) suggest multiple readings. The comparison was against a "NIST-traceable calibrated digital protractor."
    • Data Provenance: Bench test, likely conducted in a controlled laboratory environment. No country of origin mentioned, but implied to be internal testing by the manufacturer (Ruthless Spine, LLC, based in Irwindale, CA, USA). Retrospective.
  • Procedural Accuracy Testing:
    • Test Set Sample Size: 240 trials (3 measurements at each of 5 levels for 16 surgeon users).
    • Data Provenance: Simulated use conditions, described as the user positioning the RJB instrument against a cadaver pedicle. This indicates a controlled lab setting, not human clinical data. Retrospective.
  • Usability Testing:
    • Test Set Sample Size: 16 surgeon users.
    • Data Provenance: Simulated use conditions involving surgeon evaluation. Retrospective.

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

  • Angle Measurement Accuracy: The ground truth was established by a "NIST-traceable calibrated digital protractor." This is a metrological standard, not human experts.
  • Procedural Accuracy Testing:
    • Ground Truth: "Ground-truth measurement (gravity)" derived from fluoroscopic and photographic images taken when the instrument was positioned against a cadaver pedicle at a "pre-planned axial angle from MRI imaging and sagittal angle from fluoroscopy imaging." One could infer that the pre-planned angles themselves were established by experts interpreting MRI/fluoroscopy, but this is not explicitly stated. The final comparison is to the "gravity," implying the ideal sensor reading.
    • Experts: 16 "surgeon users" participated in positioning the instrument, but their role was as test subjects operating the device, not establishing the ground truth for the device's accuracy output.
  • Usability Testing:
    • Experts: 16 surgeons.
    • Qualifications: Referred to as "surgeons representing potential clinical users." No specific years of experience or subspecialty are given.

4. Adjudication Method for the Test Set

  • Angle Measurement Accuracy: No human adjudication method; direct comparison to a calibrated instrument.
  • Procedural Accuracy Testing: No explicit human adjudication method for the final composite angle error. The error was calculated by comparing the device's measured composite angle to the "ground-truth measurement (gravity)" derived from imaging.
  • Usability Testing: Surgeon feedback was captured via a questionnaire, with "Avg Score/Summary" indicating aggregation of individual surgeon responses, but not an adjudication process among experts to resolve discrepancies in truth.

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

  • No, an MRMC comparative effectiveness study was not explicitly described.
  • The studies focused on the intrinsic accuracy and usability of the device itself in simulated settings, not on how the device improves human reader performance in a clinical decision-making context.
  • The "Indications for Use" and "Limitations" sections explicitly state: "No clinical benefit has been demonstrated or is claimed," and "The RJB device is not intended to replace a surgeon's judgment and has not undergone clinical evaluation."

6. Standalone Performance (Algorithm Only Without Human-in-the-Loop Performance)

  • Yes, standalone performance was assessed. The "Angle Measurement Accuracy" test directly assessed the device's (RJB's) ability to measure angles against a calibrated protractor, independent of human interaction beyond initial setup.
  • The "procedural accuracy testing" involved surgeons in a simulated-use environment, but the measurement itself is the algorithm's output compared to a ground truth, rather than measuring human decision-making with or without the algorithm.

7. Type of Ground Truth Used

  • Angle Measurement Accuracy: NIST-traceable calibrated digital protractor (metrological standard).
  • Procedural Accuracy Testing: "Ground-truth measurement (gravity)" calculated from fluoroscopic and photographic images of the instrument positioned against a cadaver pedicle at pre-planned angles (derived from MRI/fluoroscopy).
  • Usability Testing: Surgeon subjective feedback via questionnaire.

8. Sample Size for the Training Set

  • The document describes the device's firmware and app software as "programmed firmware" and "software." It refers to "Software verification testing" and "Verification Testing" to confirm the system functions as designed.
  • No information is provided regarding a training set sample size. This suggests that the device's core functionality (measuring angles using an accelerometer and displaying them) is based on deterministic algorithms and physics, rather than machine learning models that require training data.

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

  • Since there's no indication of a machine learning-based training set, this question is not applicable to the information provided. The device's operation relies on physical principles (accelerometer, trigonometry) and programmed logic, not learned patterns from a training dataset.

N/A