K Number
K231875
Date Cleared
2024-02-22

(241 days)

Product Code
Regulation Number
884.1425
Panel
OB
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The Vaginal Tactile Ultrasound Imager obtains high-resolution mapping of pressures, ultrasound images and assesses the strength of pelvic floor muscles within the vagina. It is used in a medical setting to acquire the pressures, ultrasound images and store the corresponding data. It also provides visualization, analysis tools and information. The real time data as well as the analysis information can then be viewed with an intention of assisting in the diagnosis and evaluation. The device is intended for use by physicians, surgeons and medically trained personnel.

Device Description

The Vaginal Tactile Ultrasound Imager (TIUSv) acquires ultrasound images and tactile images (pressure maps) from within the vagina and assesses the strength of pelvic floor muscles within the vagina. The device provides data on the pressures applied to vaginal walls along with the probe location to visualize pelvic floor support structures, and to record pelvic floor muscle contraction. The TIUSv software provides measurement, imaging, and reporting tools. The acquired data and the analysis results provided by TIUSv software can be used by a physician for quantitative biomechanical assessment of the vagina and pelvic floor conditions. The TIUSv probe is equipped with 96 pressure (tactile) sensors, 192 ultrasound transducers, an orientation sensor (accelerometer), temperature sensors, and micro-heaters. During the clinical procedure, the probe is used to acquire dynamic ultrasound images and pressure responses from the vaginal walls under applied loads. The TIUSv examination procedure includes data collection from all segments of the vagina. Real-time data are sampled from the probe sensors via the interface electronics. The TIUSv software displays the acquired data in real time. The resulting pressure maps (tactile images) of the vagina integrate all the acquired pressure and positioning data for each of the pressure sensing elements. The TIUSv records the dynamic contraction for pelvic floor muscle(s). The probe surfaces that contact the vaginal walls are preheated to human body temperature. The TIUSv supports physician data analysis by means of a playback function, which replays a stored session using previously recorded data instead of the real time data. The TIUSv also provides data and graphs such as pressure applied to the probe and location, resting pressures, muscle contraction pressures and calculated pressure gradients.

AI/ML Overview

The FDA 510(k) clearance letter for the Vaginal Tactile Ultrasound Imager (TIUSv) (K231875) (pages 0-10) indicates that the device's performance was evaluated through various tests, but it does not provide a detailed study proving the device meets specific acceptance criteria for its diagnostic capabilities (e.g., accuracy in assessing pelvic floor muscle strength or identifying abnormalities using ultrasound).

Instead, the documentation focuses on:

  • Bench testing for measurement accuracy of pressure, temperature, and orientation.
  • Safety and engineering compliance (biocompatibility, electrical safety, EMC, cleaning/disinfection validation).
  • Software verification and validation.
  • Comparison to a predicate device (Vaginal Tactile Imager K142355) and a reference device (Acuson Ultrasound System) to establish substantial equivalence.

While the document states, "The real-time data as well as the analysis information can then be viewed with an intention of assisting in the diagnosis and evaluation," it does not present clinical study results from human subjects to establish the device's diagnostic performance relative to acceptance criteria in terms of sensitivity, specificity, accuracy, or its impact on human reader performance.

Therefore, the following sections will be based on the information provided in the document, acknowledging the absence of clinical diagnostic performance data.


Acceptance Criteria and Reported Device Performance

As specific diagnostic performance acceptance criteria from a clinical study are not provided, the table below focuses on the measurement performance acceptance criteria that were met through bench testing.

1. Table of Acceptance Criteria and Reported Device Performance (Measurement)

ParameterAcceptance Criteria (Bench Test)Reported Device Performance (Bench Test)
Pressure Measurement
Pressure Range0 – 30 kPa0 – 30 kPa
Accuracy (0-30 kPa)± 0.7 kPa (Standard Deviation for all sensors)± 0.7 kPa
Max Deviation (individual sensor)± 2.5 kPa± 2.5 kPa
Orientation Measurement
Orientation Range (rotation)0 ± 360 degrees0 ± 360 degrees
Orientation Range (elevation)0 ± 45 degrees0 ± 45 degrees
Accuracy (Orientation)± 1.0 degrees± 1.0 degrees
Temperature Measurement
Temperature Range+20 °C to +40 °C+20 °C to +40 °C
Accuracy (Temperature)± 0.5 °C± 0.5 °C
Ultrasound Imaging
Ultrasound Frequency4-7 MHz4-7 MHz
Lateral Resolution2.0 mm2.0 mm
Lateral Positioning Accuracy±1.0 mm±1.0 mm
Axial Resolution1.0 mm1.0 mm
Axial Positioning Accuracy±2.0 mm±2.0 mm

2. Sample Size for the Test Set and Data Provenance

The provided document details various engineering and bench tests, along with software verification and validation. It does not describe a 'test set' in the traditional sense of a clinical dataset of patients assessed by the device for diagnostic performance. The performance data presented (e.g., pressure accuracy, ultrasound resolution) are derived from bench testing as indicated in Section 6, "OVERVIEW OF THE PERFORMANCE DATA." Therefore, there is no patient-specific sample size, country of origin, or information on retrospective/prospective data collection for diagnostic efficacy evaluation.

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

Since the document focuses on bench testing and engineering verification, no experts were used to establish ground truth for a clinical test set. The "ground truth" for the reported measurement accuracies (pressure, temperature, orientation) would have been established using precisely calibrated instruments and phantom models during laboratory bench testing.

4. Adjudication Method for the Test Set

As there was no clinical "test set" requiring human interpretation or diagnosis, no adjudication method was used or described.

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

A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not performed or reported in this documentation. The clearance is based on substantial equivalence to a predicate device, and engineering/bench testing, not on clinical efficacy beyond what is needed to demonstrate equivalence.

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

The document does not describe a standalone performance study of an AI algorithm in the context of diagnostic accuracy. The device provides "visualization, analysis tools and information" to assist physicians, suggesting a human-in-the-loop use case. The reported performance metrics are for the device's physical measurement capabilities, not an AI algorithm's diagnostic output.

7. Type of Ground Truth Used

For the reported measurement performance (e.g., pressure, temperature, orientation accuracy, ultrasound resolution), the ground truth was established through calibrated reference instruments and defined test conditions during bench testing and engineering verification. There is no mention of expert consensus, pathology, or outcomes data being used as ground truth for diagnostic accuracy in patient populations.

8. Sample Size for the Training Set

The document does not mention any training set size, as it does not describe an AI/machine learning model whose diagnostic performance was evaluated from a clinical dataset. The software verification and validation refer to general software development practices (e.g., IEC/EN 62304), not the training of a predictive model.

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

As no training set is described for an AI/machine learning model, there is no information on how ground truth was established for such a set.

§ 884.1425 Perineometer.

(a)
Identification. A perineometer is a device consisting of a fluid-filled sack for intravaginal use that is attached to an external manometer. The devices measure the strength of the perineal muscles by offering resistence to a patient's voluntary contractions of these muscles and is used to diagnose and to correct, through exercise, uninary incontinence or sexual dysfunction.(b)
Classification. Class II (performance standards).