Search Results

EPIQ: The intended use of EPIQ Ultrasound Diagnostic System is diagnostic ultrasound imaging and fluid flow analysis of the human body, with the following indications for use: Abdominal, Cardiac Adult, Cardiac other (Fetal), Cardiac Pediatric, Cerebral Vascular, Cephalic (Adult), Cephalic (Neonatal), Fetal/Obstetric, Gynecological, Intraoperative (Vascular), Intraoperative (Cardiac), intra-luminal, intra-cardiac echo, Musculoskeletal (Conventional), Musculoskeletal (Superficial), Ophthalmic, Other: Urology, Pediatric, Peripheral Vessel, Small Organ (Breast, Thyroid, Testicle), Transesophageal (Cardiac), Transrectal, Transvaginal, Lung.
Affiniti: The intended use of Affiniti Series Diagnostic Ultrasound Systems is diagnostic ultrasound imaging and fluid flow analysis of the human body, with the following indications for use: Abdominal, Cardiac Adult, Cardiac Other (Fetal), Cardiac Pediatric, Cerebral Vascular, Cephalic (Adult), Cephalic (Neonatal), Fetal/Obstetric, Gynecological, Intraoperative (Vascular), Intraoperative (Cardiac), Musculoskeletal (Conventional), Musculoskeletal (Superficial), Other: Urology, Pediatric, Peripheral Vessel, Small Organ (Breast, Thyroid, Testicle), Transesophageal (Cardiac), Transrectal, Transvaginal, Lung.

The R-Trigger algorithm software feature on Philips EPIQ and Affiniti Ultrasound System is intended to support detection of R-wave peak (R-trigger) as an input to certain TTE clinical applications, initially including AutoStrain LV, AutoEF, 2D Auto LV (collectively referred to as "AutoStrain"), and AutoMeasure applications. The R-trigger algorithm is planned to be implemented as workflow enhancement for transthoracic clinical applications on EPIQ and Affiniti Ultrasound Systems in the VM13 software release. The Auto-Measure and AutoStrain features support users during B-mode (2D), CW-, PW- and TDI-Doppler measurements by automating some of the measurements needed to complete a routine transthoracic echo (TTE) exam for adult patients. The R-trigger feature (non-ECG-based) has been developed to enable clinical users to use AutoMeasure and AutoStrain application without the R-trigger (ECG based) input, which is currently required. There are no hardware changes to the EPIQ and Affiniti systems due to change to the introduction of the R-Trigger software application. The software application is supported by all EPIQ and Affiniti models running software version 13.0 or higher.

The provided FDA 510(k) clearance letter describes the R-Trigger software application on Philips EPIQ and Affiniti Ultrasound Systems, which aims to provide an alternative method for detecting R-wave peaks (R-triggers) for cardiac clinical applications like AutoStrain and AutoMeasure, especially when the ECG signal is unavailable or unusable.

Here's an analysis of the acceptance criteria and the study proving the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance:

The primary acceptance criteria for the R-Trigger algorithm are related to the agreement of its R-wave time stamp detection with the ground truth (ECG-based R-trigger) and the subsequent impact on the clinical outputs of AutoMeasure and AutoStrain. These are evaluated using Bland-Altman analysis (for agreement, specifically the Upper and Lower Limits of Agreement, LoA) and Pearson's correlation (for correlation, specifically the Lower Confidence Bound, LCB).

Ask a specific question about this device

The intended use of the 5000 Compact series ultrasound systems is diagnostic ultrasound imaging and fluid flow analysis of the human body with the following Indications for Use:Abdominal, Cardiac Adult, Cardiac Pediatric, Carotid, Cerebral Vascular, Cephalic (Adult),Cephalic (Neonatal), Fetal Echo, Fetal/Obstetric, Gynecological, Intraoperative (Vascular), Lung, Musculoskeletal (Conventional), Musculoskeletal (Superficial), Ophthalmic, Pediatric, Peripheral Vessel, Small Parts, Transesophageal (Cardiac), Transrectal, Transvaginal, and Urology.

The purpose of this Traditional 510(k) Pre-Market Notification is to introduce the Auto Measure Artificial Intelligence-Machine Learning software feature onto the 5000 Compact Series Ultrasound Systems.

The Auto Measure feature utilizes machine learning to provide a subset of semi-automated and editable measures during an echocardiography or when reviewing an already acquired echocardiography. When Auto Measure Version 2 is enabled, the healthcare professional performs an echocardiography with a workflow that provides the user with a semi-automated measurement that can be edited, accepted, or rejected.

Philips has designed Auto Measure as a "locked" algorithm prior to marketing. As defined by FDA in the discussion paper Proposed Requlatory Framework for Modifications to AI/ML Based Software as a Medical Device (SaMD) published April 2, 2019, this "locked" algorithm provides the same result each time the same input is applied to it and does not change with use.

The Auto Measure software feature does not introduce new modes, presets, measurements, or system components (e.g. transducers) to the Philips 5000 Compact Series Ultrasound Systems K222648.

No hardware changes to the 5000 Compact Series Ultrasound Systems K222648 are required when using the Auto Measure feature, and existing, commercialized Philips transducers are used for the Auto Measure feature.

5000 Compact Series Ultrasound Systems are part of the VM platform product family, The Auto Measure Version 1 feature was originally cleared (K211597) on EPIQ and Affiniti models running software version 9.0 (VM9.0). The Auto Measure feature is also available to all software releases following VM9.0.

Since the initial Auto Measure feature initial clearance (Version 1.0), a subset of integrated measurement detectors has been trained with additional training data in the Auto Measure feature (Version 2) which is scope of this submission

Auto Measure for this premarket notification utilizes the same software version platform VM as the Reference Device, Affiniti Diagnostic Ultrasound Systems K211597.

This document describes the Philips 5000 Compact Series Ultrasound Systems with the new Auto Measure Version 2 AI-Machine Learning software feature. The information provided focuses on the performance data to demonstrate substantial equivalence.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for each measurement detector are based on the Checked Limits of Agreement (LoAs), derived from clinical literature on human interobserver variability. The reported device performance is the Measured Limits of Agreement (LoA) between the detector's prediction and the manual ground truth. Each detector met its specified acceptance criteria.

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

The test set consisted of a total of 500 studies, with one study per subject. This included 200 known normal subjects and 300 patients with a confirmed pathology.

The data provenance is described as:

• Anonymized transthoracic echocardiography DICOM data and metadata.
• Data recorded by multiple sonographers and physicians qualified in echocardiography.
• Large sample of adults of various ethnicities.
• Echocardiographic recordings acquired according to guideline-standard echocardiographic procedures between 2010 and 2021.
• Data collected from 20 centers and 16 countries.

The number of studies used for validation (N) per specific detector varied, as not all measurements are performed in all studies, ranging from 53 to 469 as shown in the table above.

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

The ground truth for the test set was established by clinical experts during routine care. While the exact number is not specified for the test set's ground truth, it is stated that "All detector models were trained from the ground up... Training of the detectors was based on manual measurements performed by human experts done for diagnostic purpose." This implies the ground truth for both training and testing was established by qualified human experts performing manual measurements for diagnostic purposes. The data was collected by "multiple sonographers and physicians qualified in echocardiography."

4. Adjudication Method for the Test Set

The adjudication method is not explicitly described as a formal 'X+Y' consensus method. The document states that the Auto Measure results were "compared to ground truth measurements established by clinical experts during routine care." This suggests that the established clinical measurements served as the ground truth directly for comparison, rather than a separate adjudication process of multiple readers for the test set.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not explicitly detailed in this summary. The study focuses on evaluating the standalone performance of the Auto Measure feature against expert manual measurements, using human-human interobserver variability data from clinical studies as a benchmark for acceptance criteria, not for direct comparison of human performance with and without AI assistance. The document mentions that the feature is a "workflow improvement" and that "the operator is responsible for the final result and has to apply manual edits to the automated output whenever required."

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, a standalone performance study was done. The "Detector performance results" table directly compares the "detector prediction" against "manual ground truth." The text further clarifies: "Auto Measure analyzed a set of new, previously images, and its automated results were compared to ground truth measurements established by clinical experts during routine care. Both the manual and automated measurements were performed on the same images without any adjustments to the software's output or the clinical data used as ground truth." This indicates a direct evaluation of the algorithm's output without human intervention in the Auto Measure readings during the test phase.

7. The Type of Ground Truth Used

The ground truth used was expert consensus / manual measurements performed by clinical experts during routine care for diagnostic purposes. These manual measurements are considered the "current gold standard" against which the Auto Measure output is compared.

8. The Sample Size for the Training Set

The total training pool comprises more than 6000 studies.

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

The ground truth for the training set was established through manual measurements performed by human experts for diagnostic purposes. The data was collected by "qualified medical [personnel] in echocardiography using TTA software (versions TTA2.31.00 and TTA2.50.00) or various Philips ultrasound systems annotation and following established guidelines (Mitchel et al 2019)." These data were then used for training and tuning in a cross-validation framework.

Ask a specific question about this device

Indications for use of TOMTEC-ARENA software are quantification and reporting of cardiovascular, fetal, and abdominal structures and function of patients with suspected disease to support the physician in the diagnosis.

TOMTEC-ARENA is a clinical software package for reviewing, quantifying and reporting digital medical data. The software can be integrated into third party platforms. Platforms enhance the workflow by providing the database, import, export and other services. All analyzed data and images will be transferred to the platform for archiving, reporting and statistical quantification purposes.
TTA2 consists of the following optional modules:

• IMAGE-COM
• I REPORTING
• AutoStrain LV / SAX / RV / LA I
• 2D CPA
• FETAL 2D CPA ■
• 4D LV-ANALYSIS
• . 4D RV-FUNCTION
• I 4D CARDIO-VIEW
• I 4D MV-ASSESSMENT
• I 4D SONO-SCAN
• TOMTEC DATACENTER (incl. STUDY LIST, DATA MAINTENANCE, WEB ■ REVIEW)
  The purpose of this traditional 510(k) pre-market notification is to introduce semi-automated cardiac measurements based on an artificial intelligence and machine learning (AI/ML) algorithm. The Al/ML algorithm is a Convolutional Network (CNN) developed using a Supervised Learning approach. This Al/ML algorithm enables TOMTEC-ARENA to produce semi-automated and editable echocardiographic measurements on BMODE and DOPPLER datasets. The algorithm was developed using a controlled internal process that defines activities from the inspection of input data to the training and deployment of the algorithm: The training process begins with the model observing, and optimizing its parameters based on the training pool data. The model's prediction and performance are then evaluated against the test pool. The test pool data is set aside at the beginning of the project. During the training process, the Al/ML algorithm learned to predict measurements by being presented with a large number of echocardiographic data manually generated by qualified healthcare professionals. The echocardiographic studies were randomly assigned to be either used for training (approx. 2,800 studies) or testing (approx. 500 studies). A semi-automated measurement consists of a cascade of detection steps. It starts with a rough geometric estimate, which is subsequently refined more and more: The user selects a frame on which the semi-automated measurements shall be performed in TOMTEC-ARENA. Image- & metadata, e.g. pixel spacing, are transferred to the semi-automated measurement detector. The semi-automated measurement detector predicts the position of start and end caliper in the pixel coordinate system. These co-coordinates are transferred back to the CalcEngine, which converts the received data back into real world coordinates (e.g. mm) and creates the graphical overlay. This superimposed line can be edited by the user. The end user can edit, accept, or reject the measurement(s). This feature does not introduce any new measurements, but allows the end user to perform semi-automated measurements. The end user can also still perform manual measurements and it is not mandatory to use the semi-automated measurements. The semi-automated measurements are licensed separately.

Here's an analysis of the acceptance criteria and study details for the TOMTEC-ARENA device, based on the provided FDA 510(k) summary:

The 510(k) summary describes the TOMTEC-ARENA software, which introduces semi-automated cardiac measurements based on an AI/ML algorithm. The primary focus of the non-clinical performance data is on software verification, risk analysis, and usability evaluation, as no clinical testing was conducted.

1. Table of Acceptance Criteria and Reported Device Performance

The provided document does not explicitly list quantitative acceptance criteria for the AI/ML algorithm's performance in terms of accuracy or precision of the semi-automated measurements. Instead, it states that "Completion of all verification activities demonstrated that the subject device meets all design and performance requirements." and "Testing performed demonstrated that the proposed TOMTEC-ARENA (TTA2.50) meets defined requirements and performance claims." These are general statements rather than specific, measurable performance metrics.

Similarly, there are no reported quantitative device performance metrics (e.g., accuracy, sensitivity, specificity, or error rates) for the AI/ML algorithm's measurements mentioned in this summary. The summary focuses on the functional equivalence and safety of the AI-powered feature compared to existing manual measurements and predicate devices.

However, the document does imply a core "acceptance criterion":

Without specific quantitative metrics for the AI's measurement accuracy, it's challenging to provide a detailed performance table. The provided information focuses on the design validation process rather than specific benchmark results for the AI's performance.

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

• Sample Size for Test Set: Approximately 500 studies.
• Data Provenance: The document does not specify the country of origin of the data. It states, "The echocardiographic studies were randomly assigned to be either used for training (approx. 2,800 studies) or testing (approx. 500 studies)." It does not explicitly state if the data was retrospective or prospective. Given that these are "studies" used for training and testing an algorithm, it is highly probable that they are retrospective data sets, collected prior to the algorithm's deployment.

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

• Number of Experts: Not specified. The document states "a large number of echocardiographic data manually generated by qualified healthcare professionals." This implies multiple professionals but does not quantify them.
• Qualifications of Experts: "qualified healthcare professionals." Specific qualifications (e.g., radiologist with X years of experience, sonographer, cardiologist) are not provided.

4. Adjudication Method for the Test Set

• Adjudication Method: Not specified. The ground truth was "manually generated by qualified healthcare professionals," but the process for resolving discrepancies among multiple professionals (if multiple were involved per case) is not described.

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

• Was an MRMC study done? No. The summary explicitly states: "No clinical testing conducted in support of substantial equivalence when compared to the predicate devices." The nature of the AI algorithm as providing semi-automated, editable measurements, rather than a diagnostic output, likely informed this decision. The user is always in the loop and can accept, edit, or reject the AI's suggestions.
• Effect size of human readers improvement with AI vs. without AI assistance: Not applicable, as no MRMC study was performed.

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

• Was a standalone study done? Not explicitly detailed in terms of quantitative performance metrics. While the algorithm "predicts the position of start and end caliper in the pixel coordinate system" and this prediction is mentioned as being evaluated against the test pool, the results are not presented as a standalone performance metric. The nature of the device, where the user can "edit, accept, or reject the measurement(s)", strongly implies that standalone performance is not the primary focus for regulatory purposes, as it is always intended to be used with human oversight. The comparison is generally with the predicate device's manual measurement workflow and a reference device's semi-automated features.

7. Type of Ground Truth Used

• Type of Ground Truth: "manually generated by qualified healthcare professionals." This suggests expert consensus or expert-derived measurements serving as the reference standard for the algorithm's training and testing.

8. Sample Size for the Training Set

• Sample Size for Training Set: Approximately 2,800 studies.

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

• How Ground Truth Was Established: "The Al/ML algorithm learned to predict measurements by being presented with a large number of echocardiographic data manually generated by qualified healthcare professionals." This indicates that human experts manually performed the measurements on the training data, and these manual measurements served as the ground truth for the supervised learning model.

Ask a specific question about this device