Search Results

The Breast Volume Navigator (BVN) is an add-on accessory for existing ultrasound imaging systems and is intended to control position and movement of ultrasound transducers for the systematic acquisition of 2-dimensional image slices throughout a volume of interest. The BVN is intended to acquire, analyze, store and retrieve digital ultrases for computerized 3-dimensional image processing.

The BVN is intended to assist health care providers trained in breast ultrasound with tools for electromagnetic tracking of instruments with respect of breast ultrasound images generated from FDA cleared hand-held ultrasound devices. The BVN is indicated for use as an adjunct to mammography for B-mode ultrasonic imaging of a patient's breast when used with a linear array transducer.

The BVN is indicated for use as an adjunct to hand-held breast ultrasound to assist the health care providers trained in breast ultrasound in their scanning process. The scanning paths are displayed on a route map and provide quality control of scanning to provide an overall observation of scanning process.

The BVN Software Application is intended as a standalone software device installed on a windows-based computer to assist physicians with manipulation and analysis tools in reviewing breast ultrasound images. Images and data are previously recorded from various imaging systems and other sources such as calibrated spatial positioning devices. The BVN Software Application provides the capability to visualize ultrasound images along with the scanning paths and position information of the probe that is stored in the DICOM file in advance.

The BVN will allow exporting to any third-party application that has the appropriate level of DICOM compliance. The BVN is intended as a general-purpose digital 3D breast ultrasound image processing tool for radiology and surgery.

The device is not intended to be used in the environment of strong magnetic fields, such as in Magnetic Resonance Imaging (MRI) room.

The device is not intended to be used as a replacement for screening mammography.

Device Description

The Breast Volume Navigator (BVN™) Model G-2000, hereinafter maybe referred to as the BVN™ G-2000 System or the BVN™ System, comprises hardware components and a software element, including the following components: a magnetic position tracking device, sensor attaching pieces used to attach magnetic sensors to the skin and ultrasound probe, a central control unit computer, and software for controlling the system, collecting, and processing ultrasound images and positional data, and performing automated annotations.

The BVN™ System has a touch-screen user interface and push-button for power on the system.

The Use Interface (touchscreen display and central control unit computer) are placed on a medical grade cart that can be positioned next to a standard adjustable examination table for ease of use and ergonomic adaptation.

The BVN™ System has a USB port available for transferring files via USB Memory Stick.

The BVN™ System has an Ethernet port for connection to a PACS system, using DICOM.

The BVN™ System has a VGA/DVI Input Ports for capturing images from an ultrasound imaging scanner.

The BVN™ System receives ultrasound DICOM images from the US machine via the network connection and telemetry data from a position tracking system. The BVN™ automatically detects when the image is being frozen on the US machine and takes a snapshot of the telemetry data at that time. Later, when the BVN™ receives the DICOM image, it associates the telemetry data to the image from the time when the image was frozen on the US machine.

A Completeness-of-Scan Assessment ensures that all breast tissue has been sufficiently covered, the BVN™ System will achieve this by tracking the real-time position of the probe relative to the breast and chest wall to ensure the entire breast volume has been covered with sufficient detail to allow the detection of sub centimeter lesions. As the probe is swept over the breast region, the technician will be provided a completeness map highlighting the portions of the breast that have not been sufficiently scanned. The BVN™ System will provide the operator a display of regions with insufficient coverage, prompting the operator to rescan these regions. The BVN™ System can retrieve previous examinations DICOM images allowing the user to enter positional data from another/previous ultrasound examination for follow-up purposes. Any anomalies discovered during the follow-up process can be evaluated using the localization and measurement tools included in the software.

The customer's existing ultrasound probe securely attaches to the BVN™ System probe sensor. During a scan, the operator applies constant pressure to the transducer against the patient's breast tissue and can rotate the transducer (pitch and roll) to accommodate for the physical characteristics of the breast.

Exam data is subsequently reviewed on standard radiological viewing stations.

AI/ML Overview

The provided text describes the acceptance criteria and study results for the MetriTrack Breast Volume Navigator (BVN™) Model G-2000.

Acceptance Criteria and Performance Study Table

Acceptance Criteria (Performance Metric)	Target/Threshold	Reported Device Performance
Linear Accuracy and Precision of Distance to Nipple	≤ 5 mm	Achieved ≤ 5 mm
Accuracy and Precision of Clock Face Angle	≤ 5 degrees	Achieved ≤ 5 degrees
Accuracy and Precision of Patient Body Angles	≤ 5 degrees for Coronal, Transverse, Sagittal planes	Achieved ≤ 5 degrees for all three anatomical planes
Accuracy and Precision of Linear Accuracy	≤ 1 mm at one standard deviation	Achieved ≤ 1 mm at one standard deviation
Completeness of Scanning Breast Contour Accuracy	+/- 5 mm maximum positional error in X-Y plane	Accurately captured and represented hand-drawn breast contour with a maximum positional error of +/- 5 mm in the X-Y plane.
Overlap between Adjacent Scanning Bands	Discern when overlaps are < set threshold (5-30 mm ± 2 mm)	Able to discern when overlaps of adjacent bands are less than a set threshold within a range of 5 to 30 mm ± 2 mm of overlap.
3D Point Coordinates Velocity	Detect velocities above/below threshold (15-50 mm/s +/- 10% at 1 SD)	Detected 3D Point coordinates of image corners velocity above and below a set threshold within a range of 15 to 50 mm/s +/- 10% at 1 standard deviation.

Study Details

Sample Size and Data Provenance:
- Test Set Sample Size: The document does not specify a distinct "test set" sample size in terms of number of cases or patients from a clinical study. Instead, it refers to "Extensive Performance, Verification and Validation testing" conducted "in-house by trained personnel in a simulated work-environment using phantoms."
- Data Provenance: The testing appears to be conducted in a simulated lab environment ("in-house") using "phantoms." The document does not indicate the country of origin for the data or whether it was retrospective or prospective. It suggests controlled lab studies rather than clinical data.
Number of Experts and Qualifications for Ground Truth:
- The document states that the testing was performed "by trained personnel." It does not specify the number of experts or their specific qualifications (e.g., radiologists, years of experience) used to establish ground truth for the test set. Given the phantom-based testing, the "ground truth" would be the known physical properties and measurements of the phantoms.
Adjudication Method for the Test Set:
- No information on adjudication methods is provided, as the testing was primarily against known phantom parameters rather than expert readings needing adjudication.
Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- No MRMC comparative effectiveness study involving human readers with or without AI assistance is mentioned. The study described focuses solely on the device's measurement accuracy and completeness of scan capabilities using phantoms.
Standalone (Algorithm Only) Performance:
- The performance data presented are for the BVN™ System itself, acting as an accessory to existing ultrasound systems, particularly its ability to track position, measure distances/angles, and ensure scan completeness. This can be considered the standalone performance of the device's core functionalities in a controlled setting, but not necessarily a standalone AI algorithm in a diagnostic context, as the device is not making diagnostic interpretations.
Type of Ground Truth Used:
- The ground truth was established by the known physical dimensions and properties of the phantoms used for testing. For example, distances, angles, and scan coverage could be precisely controlled and measured on the phantoms.
Training Set Sample Size:
- The document does not mention a separate "training set" or "training data" as the device is described as an accessory that performs measurements and tracking, rather than an AI/ML diagnostic algorithm that requires extensive training data. The "acceptance criteria" are based on accuracy against physical measurements, not a trained model's diagnostic output.
How Ground Truth for Training Set was Established:
- Not applicable, as no distinct training set for an AI/ML algorithm is described. The device's functionality appears to be based on established engineering and measurement principles applied to physical sensors and software logic, not a learned model from a dataset.

Ask a Question

Ask a specific question about this device

Page 1 of 1