Search Results

The Sensis Vibe Hemo (VD15B) and Sensis Vibe Combo (VD15B) recording systems are intended to be used as a diagnostic and administrative tool supporting hemodynamic cardiac catheterizations and/or electrophysiology studies, for cardiac as well as interventional radiology and surgical studies. The system is equipped with modules, enabling various configurations ranging from a standalone acquisition unit with limited administrative functionality to multiunit installations with a common database and satellite workstations accessing the administrative tools.

The device is intended to be used on either or both of the following populations:

1. Adult and pediatric populations requiring electrophysiology examinations, typically when the patient is suffering from cardiac arrhythmias.
2. Adult and pediatric populations requiring hemodynamic examinations, typically when the patient has a heart or vascular disease resulting in insufficient hemodynamic functionality.

SIEMENS Medical Solutions USA. Inc. intends to market the Sensis Vibe Hemo (VD15B) and Sensis Vibe Combo (VD15B), a hemodynamic and electrophysiological recording system. This 510(k) submission describes modifications to the previously cleared Primary Predicate Device the Sensis (K150493). Sensis Vibe Hemo (VD15B) and Sensis Vibe Combo (VD15B) is a multi-channel computer-based stationary system for the measurement, display, and printout of bio-physiological events. There are two configurations for this device: Sensis Vibe-Hemo and Sensis Vibe Combo.

Hemodynamic and electrophysiologic signals such as intracardiac pressure, ECG signals, and intracardiac electrograms (ICEG) are measured and displayed by the system; several hemodynamic calculations are performed based on the measured values of the input signals. These data can be recorded in real-time and stored on removable media or in a digital DICOM archive.

The Sensis Vibe Hemo (VD15B) and Sensis Vibe Combo (VD15B) system is comprised of the following basic hardware components: a small cabinet (video distribution box), front-end electronics, a keyboard with a mouse, and master and slave monitor(s) for real-time presentation of ECG tracings and pressure and ICEG waveforms. The small cabinet (video distribution box) contains power distribution electronics, video drivers, and a separation device for electrical isolation between the small cabinet and the signal input box. The front-end electronics contain modules for the acquisition of invasive blood pressure, ECG, SpO2, CO, and optionally ICEG and NBP, and are normally stalled at the operating table.

The Siemens Medical Solutions USA Inc. Sensis Vibe Hemo (VD15B) and Sensis Vibe Combo (VD15B) are electrophysiological and hemodynamic recording systems. The acceptance criteria and the study proving the device meets these criteria are detailed below. It's important to note that the provided document outlines conformity to standards and non-clinical performance testing for specific modifications to an existing cleared device (Sensis K150493) rather than a comprehensive de novo clinical study for the entire system.

1. Table of Acceptance Criteria and Reported Device Performance:

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

The document primarily describes non-clinical bench testing for the modifications made to the device.

• DFR™ Bench Test Study: DFR indices obtained from Sensis Vibe Hemo (VD15B) and Sensis Vibe Combo (VD15B) were compared with DFR indices obtained from a previous bench test study performed and submitted for iLab Polaris-Modality Guidance System (K191008).
• Other Testing: General "verification and validation testing," "non-clinical tests," and "software documentation" were performed. No specific sample sizes for clinical data sets are mentioned, as the focus is on a substantial equivalence claim based on modifications and adherence to standards.
• Data Provenance: The data provenance for the DFR™ comparison is a prior bench test study. For other aspects, it is internal company testing and validation processes. No mention of country of origin for specific test data is provided. Given it's premarket notification, it's likely internal development and testing data.

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

The document does not specify the number or qualifications of experts used to establish ground truth for this submission, as it relies heavily on bench testing and reference to prior clearances. For the DFR™ comparison, the "ground truth" is essentially the established performance of the legally marketed predicate device (iLab Polaris).

4. Adjudication Method for the Test Set:

Given that the testing described is primarily non-clinical bench testing, there is no mention of an adjudication method (e.g., 2+1, 3+1). Such methods are typically used in clinical studies involving interpretation by multiple readers.

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 described in this submission. The device is a "programmable diagnostic computer" intended as a "diagnostic and administrative tool," not an AI-assisted diagnostic aid that directly improves human reader performance in interpreting medical images or signals. Its function is to measure, display, and record bio-physiological events and perform calculations like DFR™.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:

The description of the device and its testing suggests that it functions as a standalone system (algorithm/hardware only) for measuring, displaying, and recording physiological data, and performing calculations like DFR™. The DFR™ bench test compared the device's output numerically to that of a predicate device, indicating a standalone performance evaluation of this specific function.

7. The Type of Ground Truth Used:

For the DFR™ assessment modification, the ground truth relied upon was the numerical output and algorithm of a legally marketed predicate device (iLab Polaris-Modality Guidance System). For other aspects (temperature display, software, etc.), the "ground truth" is adherence to established engineering specifications, safety standards, and performance benchmarks as determined by internal verification and validation processes. There is no mention of pathology or outcomes data as ground truth for this submission.

8. The Sample Size for the Training Set:

The document does not provide information on the sample size for a training set. The submission focuses on modifications to an existing device, which implies that core algorithms have likely been developed and validated previously. The new DFR™ functionality appears to be a direct implementation of an existing, cleared algorithm from a predicate device, rather than a novel AI model requiring a new training set.

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

As no specific training set is mentioned for the modifications, the document does not describe how ground truth for any training set was established.

Ask a specific question about this device

This computed tomography system is intended to generate and process cross-sectional images of patients by computer reconstruction of x-ray transmission data.

The images delivered by the system can be used by trained staff as an aid in diagnosis, treatment and radiation therapy planning as well as for diagnostic and therapeutic interventions.

This CT system can be used for low dose lung cancer screening in high risk populations*.

*As defined by professional medical societies. Please refer to clinical literature, including the results of the National Lung Screening Trial (N Engl J Med 2011; 365:395-409) and subsequent literature, for further information.

The subject device SOMATOM CT Scanner Systems with SOMARIS/7 syngo CT VB30 are Computed Tomography X-ray Systems which feature one (single source) continuously rotating tube-detector system and function according to the fan beam principle. The SOMATOM CT Scanner Systems with Software SOMARIS/7 syngo CT VB30 produces CT images in DICOM format, which can be used by trained staff for post-processing applications commercially distributed by Siemens Healthcare and other vendors as an aid in diagnosis, treatment preparation and therapy planning support (including, but not limited to, Brachytherapy, Particle including Proton Therapy, External Beam Radiation Therapy, Surgery). The computer system delivered with the CT scanner is able to run optional post processing applications.

The platform software for the SOMATOM CT Scanner Systems, SOMARIS/7 syngo CT VB30, is a commandbased program used for patient management, data management, X-ray scan control, image reconstruction, and image archive/evaluation.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text.

1. Table of Acceptance Criteria and Reported Device Performance

The document primarily focuses on functional verification and validation testing rather than explicit, quantifiable acceptance criteria with corresponding performance metrics for each feature in a tabular format. Instead, it describes the objective of each test and then states that the results were found to be acceptable or passed.

However, we can extract the objectives and the documented outcomes for features where some quantifiable or descriptive performance is mentioned:

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

• FAST Bolus: The test describes using a "real contrast enhancement curve" determined by measurements with a dynamic scan mode. The subsequent supporting peer-reviewed studies provide more detail:

  • Korporaal et al. (2015): Not explicitly stated, but implies a cohort undergoing bolus tracking.
  • Hinzpeter et al. (2019): 108 patients received patient-specific trigger delay (subject), 108 patients received fixed trigger delay (reference). Prospective CT angiography scans of the aorta.
  • Gutjahr et al. (2019): 3 groups, 20, 20, and 40 patients respectively.
  • Yu et al. (2021): 104 patients (52 per group, implied) in abdominal multiphase CT, comparing individualized vs. fixed post-trigger delay.
  • Yuan et al. (2023): 204 consecutive participants randomly divided into two groups (102 patients each). A prospective study in coronary CT angiography (CCTA).
  • Schwartz et al. (2018): Not explicitly stated, but implied patient-specific data.
  • Data Provenance: The supporting studies imply a mix of retrospective analysis (e.g., Korporaal et al. simulating retrospectively differences) and prospective studies based on the descriptions provided. The locations of these studies are not explicitly mentioned in the excerpt, but given Siemens' global presence, it's likely multi-national.

• FAST 3D Camera, FAST Isocentering, FAST Range, FAST Direction, FAST Planning, Tin Filtration: For these features, the testing is described as "bench testing" using phantoms and internal validation. "Patient data" is mentioned for FAST Planning but without specific numbers.

  • Sample Size: Not specified for these internal bench tests; often involves phantom studies rather than patient-level data for performance metrics. For FAST Planning, it refers to "patient data" for validation, but the sample size is not indicated.
  • Data Provenance: Implied internal testing, likely at Siemens R&D facilities. No external patient data provenance details are given.

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

• For FAST Bolus, the "ground truth" for the internal bench test was defined as an "ideal post bolus delay" determined by measurements with a dynamic scan mode. This suggests an objective, data-driven approach rather than expert consensus on individual cases for the initial ground truth. However, the supporting studies mention:
  • Hinzpeter et al. (2019): Mentions subjective image quality and CNR, which would typically involve expert readers, but the number and qualifications are not provided.
  • Yuan et al. (2023): Mentions "Both readers rated better subjective image quality." suggesting at least two readers, but their qualifications are not provided.
• For other features (FAST 3D Camera, FAST Planning, etc.), the ground truth seems to be established through objective measurement against predefined targets (e.g., "calculated by FAST Planning algorithm that are correct and can be applied without change"). No specific expert involvement for ground truth establishment for these features is detailed.

4. Adjudication Method for the Test Set

• The document does not describe a formal adjudication method (e.g., 2+1, 3+1) for the establishment of ground truth or for reader studies. Where multiple readers are mentioned (e.g., Yuan et al. for FAST Bolus), it only states their findings without detailing an adjudication process. This suggests either independent readings or consensus where needed, but not a formal adjudication protocol.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done, and Effect Size

• Yes, implicitly for FAST Bolus: The supporting publications function as comparative effectiveness studies where human assessment (e.g., subjective image quality, diagnostic confidence) is evaluated with or without the aid of the FAST Bolus prototype.
  • Hinzpeter et al. (2019): "higher overall and more uniform attenuation in the individualized cohort compared to the fixed cohort. No difference between the cohorts for image noise was found, but a higher contrast-to-noise ratio (CNR) and higher subjective image quality in the individualized cohort compared to the fixed cohort." This indicates improvement with the AI-assisted timing.
  • Yu et al. (2021): "In the arterial phase, the images of group A with the individualized post-trigger delay provided higher attenuation for all organs... Furthermore, the contrast-to-noise ratio (CNR) of liver, pancreas and portal vein were significantly higher in the group with the individualized scan timing compared to the fixed scan delay. The overall subjective image quality and diagnostic confidence between the two groups were similar." This indicates improved quantitative metrics, with subjective similar.
  • Yuan et al. (2023): "Both readers rated better subjective image quality for Group B with the individualized scan timing. Also, the mean vessel enhancement was significantly higher in Group B in all coronary vessels. After adjusting for the patient variation, the FAST Bolus prototype was associated with an average of 33.5 HU higher enhancement compared to the fixed PTD." This provides a direct effect size for enhancement.
• For the other features, the description is focused on the device's inherent performance (e.g., accuracy of landmark detection, successful implementation) rather than human reader improvements. So, no MRMC study for those.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Study was Done

• Yes, for multiple features. The "Bench Testing" descriptions primarily evaluate the algorithm's performance in a standalone manner against a defined ground truth or objective:
  • FAST Bolus: "the post bolus delay as calculated by FAST Bolus to an ideal post bolus delay... was calculated. The objectives of the test were to investigate the deviation from the post bolus delay as determined by FAST Bolus to an ideal/ground truth delay..." This is standalone.
  • FAST 3D Camera, FAST Isocentering, FAST Range, FAST Direction: The tests "demonstrate that the FAST 3D Camera feature... achieves comparable or more accurate results," "lateral isocenter accuracy... comparable," "robustness of the groin landmark is improved," "comparable accuracy of the pose detection." These are assessments of the algorithm's direct performance.
  • FAST Planning: "assess the fraction (percentage) of ranges calculated by the FAST Planning algorithm that are correct and can be applied without change." This is a direct measurement of the algorithm's output quality.
  • Tin Filtration: Verifies "successful implementation" and investigates "improved contrast-to-noise ratio (CNR) in spectral results." This is standalone performance of the image reconstruction/processing.

7. The Type of Ground Truth Used

• Objective/Measured Data:
  • FAST Bolus: "ideal post bolus delay" determined by "measurements with a dynamic scan mode" and "averaged dynamic scans."
  • FAST 3D Camera, FAST Isocentering, FAST Range, FAST Direction: Implied ground truth based on objective measurements of spatial accuracy relative to predefined targets or phantoms.
  • FAST Planning: "correct" ranges are the ground truth, implying comparison to a predefined standard or ideal plan.
  • Tin Filtration: Objective image quality criteria and spectral property measurements are used as ground truth indicators.
• Expert Consensus/Subjective Assessment (as secondary metric in supporting studies): Some of the supporting publications for FAST Bolus also incorporate subjective image quality ratings by human readers, which would likely involve some form of expert consensus or individual expert assessment.

8. The Sample Size for the Training Set

• The document does not provide information on the sample size used for the training set for any of the AI/algorithm features. This information is typically proprietary and not usually disclosed in a 510(k) summary unless specifically requested or deemed critical for demonstrating substantial equivalence.

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

• The document does not provide information on how the ground truth for the training set was established. Given the nature of these features (automated bolus timing, patient positioning, scan range planning), the training data would likely involve large datasets of CT scans annotated with physiological events, anatomical landmarks, and optimal scan parameters. These annotations would typically be established by highly qualified medical professionals (e.g., radiologists, technologists) or through automated processes validated against gold standards. However, the specific methodology is not detailed in this excerpt.

Ask a specific question about this device

The MAGNETOM system is indicated for use as a magnetic resonance diagnostic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, spectroscopic images and/or spectra, and that displays the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images and/or spectra may also be produced. Depending on the region of interest, contrast agents may be used. These images and/or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

The MAGNETOM system may also be used for imaging during interventional procedures when performed with MR compatible devices such as in-room displays and MR Safe biopsy needles.

MAGNETOM Amira and MAGNETOM Sempra with syngo MR XA50M include new and modified features comparing to the predicate devices MAGNETOM Amira and MAGNETOM Sempra with syngo MR XA12M (K183221, cleared on February 14, 2019).

The provided document is a 510(k) summary for the Siemens MAGNETOM Amira and Sempra MR systems, detailing their substantial equivalence to predicate devices. It describes new and modified hardware and software features, including AI-powered "Deep Resolve Boost" and "Deep Resolve Sharp."

However, the document does not contain the detailed information necessary to fully answer the specific questions about acceptance criteria and a study proving the device meets those criteria, particularly in the context of AI performance. The provided text is a summary for regulatory clearance, not a clinical study report.

Specifically, it lacks:

• Concrete, quantifiable acceptance criteria for the AI features (e.g., a specific PSNR threshold that defines "acceptance").
• A comparative effectiveness study (MRMC) to show human reader improvement with AI assistance.
• Stand-alone algorithm performance metrics for the AI features (beyond general quality metrics like PSNR/SSIM, which are not explicitly presented as acceptance criteria).
• Details on expert involvement, adjudication, or ground truth establishment for a test set used for regulatory acceptance, as the "test statistics and test results" section refers to quality metrics and visual inspection, and "clinical settings with cooperation partners" rather than a formal test set for regulatory submission.

The "Test statistics and test results" section for Deep Resolve Boost mentions "After successful passing of the quality metrics tests, work-in-progress packages of the network were delivered and evaluated in clinical settings with cooperation partners." It also mentions "seven peer-reviewed publications" covering 427 patients which "concluded that the work-in-progress package and the reconstruction algorithm can be beneficially used for clinical routine imaging." This indicates real-world evaluation but does not provide specific acceptance criteria or detailed study results for the regulatory submission itself.

Based on the provided text, here's what can be extracted and what is missing:

1. Table of acceptance criteria and reported device performance:

The document does not explicitly state quantifiable "acceptance criteria" for the AI features (Deep Resolve Boost and Deep Resolve Sharp) that were used for regulatory submission. Instead, it describes general successful evaluation methods:

2. Sample sized used for the test set and the data provenance:

The document mixes "training" and "validation" datasets. It doesn't explicitly refer to a separate "test set" for regulatory evaluation with clear sample sizes for that purpose. The "Test statistics and test results" section refers to general evaluations and published studies.

• "Validation" Datasets (internal validation, not explicitly a regulatory test set):
  • Deep Resolve Boost: 1,874 2D slices
  • Deep Resolve Sharp: 2,057 2D slices
• Data Provenance (Training/Validation):
  • Source: For Deep Resolve Boost: "in-house measurements and collaboration partners." For Deep Resolve Sharp: "in-house measurements."
  • Origin: Not specified by country.
  • Retrospective/Prospective: "Input data was retrospectively created from the ground truth by data manipulation and augmentation" (for Boost) and "retrospectively created from the ground truth by data manipulation" (for Sharp). This implies the underlying acquired datasets were retrospective.
• "Clinical Settings" / Publications (Implied real-world evaluation, not a regulatory test set):
  • Deep Resolve Boost: "a total of seven peer-reviewed publications 427 patients"
  • Data Provenance: Not specified by origin or retrospective/prospective for these external evaluations.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

This information is not provided in the document. It mentions "visual inspection" and "visual rating," but does not detail the number or qualifications of experts involved in these processes for the "validation" sets or any dedicated regulatory "test set." For the "seven peer-reviewed publications," the expertise of the authors is implied but not detailed as part of the regulatory submission.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

This information is not provided in the document.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

A formal MRMC comparative effectiveness study demonstrating human reader improvement with AI assistance is not described in this document. The document focuses on the technical performance of the AI features themselves and their general clinical utility as reported in external publications (e.g., faster imaging, no misinterpretation), but not a comparative study of human performance with and without the AI.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

Yes, the sections on "Test statistics and test results" for both Deep Resolve Boost and Deep Resolve Sharp describe evaluation of the algorithm's performance using quality metrics (PSNR, SSIM, perceptual loss) and visual/intensity profile comparisons. This implies standalone algorithm evaluation. No specific quantifiable results for these metrics are provided as acceptance criteria, only that tests were successfully passed and showed increased sharpness for Deep Resolve Sharp.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):

The ground truth for the AI training and validation datasets is described as:

• Deep Resolve Boost: "The acquired datasets represent the ground truth for the training and validation. Input data was retrospectively created from the ground truth by data manipulation and augmentation." This implies that the original, full-quality MR images serve as the ground truth.
• Deep Resolve Sharp: "The acquired datasets represent the ground truth for the training and validation. Input data was retrospectively created from the ground truth by data manipulation." Similarly, the original, high-resolution MR images are the ground truth.

This indicates the ground truth is derived directly from the originally acquired (presumably high-quality/standard) MRI data, rather than an independent clinical assessment like pathology or expert consensus. The AI's purpose is to reconstruct a high-quality image from manipulated or undersampled input, so the "truth" is the original high-quality image.

8. The sample size for the training set:

• Deep Resolve Boost: 24,599 2D slices
• Deep Resolve Sharp: 11,920 2D slices

Note that the document states: "due to reasons of data privacy, we did not record how many individuals the datasets belong to. Gender, age and ethnicity distribution was also not recorded during data collection."

9. How the ground truth for the training set was established:

As described in point 7:

• Deep Resolve Boost: The "acquired datasets" (original, full-quality MR images) served as the ground truth. Input data for the AI model was then "retrospectively created from the ground truth by data manipulation and augmentation," including undersampling, adding noise, and mirroring k-space data.
• Deep Resolve Sharp: The "acquired datasets" (original MR images) served as the ground truth. Input data was "retrospectively created from the ground truth by data manipulation," specifically by cropping k-space data so only the center part was used as low-resolution input, with the original full data as the high-resolution output/ground truth.

Ask a specific question about this device

MAMMOVISTA B.smart is a dedicated softcopy review environment for both screening and diagnostic mammography as well as digital breast tomosynthesis. Its user interface and workflow have been optimized to support experienced mammography and tomosynthesis reviewers in both screening and diagnostic reading. Efficiency and reading quality are supported by various specialized features. MAMMOVISTA B.smart provides visualization and image enhancement tools to aid a qualified radiologist in the review of digital mammography and digital breast tomosynthesis datasets, as well as other modalities of breast images.

MAMMOVISTA B.smart is an optional software application for the Siemens Healthineers syngo.via platform (K191040). MAMMOVISTA B.smart is an image viewing and processing software environment dedicated to breast image display. It is designed to provide the performance required for the high data volume of digital tomosynthesis and the display of multi-modality breast images, such as those from MRI and ultrasound. Individual workflows can be adapted to either screening or diaqnostic purposes.

MAMMOVISTA B.smart runs on a PC and can be used for Mammography image review together with monitors cleared for Mammography diagnostics. The software solution provides for the display of DICOM compatible information, such as breast density and CAD (Computer Aided Diagnostics) markers.

The provided text describes MAMMOVISTA B.smart (VB70), a software device for mammography image review. However, it does not explicitly state acceptance criteria or a dedicated study proving performance against such criteria. The submission is a 510(k) premarket notification for substantial equivalence, comparing the new VB70 version to a predicate device, MAMMOVISTA B.smart VB60 (K212621).

Here's an analysis based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance:

The document does not explicitly present a table of quantitative acceptance criteria or device performance metrics for the VB70 version beyond a feature-by-feature comparison to the predicate device. The performance is described in terms of functional equivalence and safety.

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

The document states: "Non-clinical tests (integration and functional) were conducted on the MAMMOVISTA B.smart during product development." It further notes, "Siemens did not conduct any clinical tests for the subject device." Therefore, the "test set" in this context refers to software testing and verification/validation, not a clinical data set for performance evaluation in a medical context.

• Sample Size for Test Set: Not specified, as it refers to internal software testing, not a clinical population.
• Data Provenance: Not applicable for a software-only 510(k) submission based on substantial equivalence and non-clinical testing. No patient data or clinical images are mentioned as being part of a "test set" for performance evaluation in the clinical sense.

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

Not applicable. This was a non-clinical software verification and validation study, not a clinical performance study requiring expert ground truth for medical diagnoses.

4. Adjudication Method for the Test Set:

Not applicable. This was a non-clinical software verification and validation study, not a clinical performance study involving human adjudication of medical findings.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

No. The document explicitly states: "Siemens did not conduct any clinical tests for the subject device." The device is a "softcopy review environment" that provides "visualization and image enhancement tools to aid a qualified radiologist," meaning it's a viewing workstation, not an AI or CAD device that provides interpretations or assists directly with diagnostic accuracy in a quantifiable way like an AI algorithm.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done:

No. The device is a viewing and processing software. It is not an algorithm designed to provide standalone diagnostic interpretations. Its purpose is to "aid a qualified radiologist in the review" of images.

7. The Type of Ground Truth Used:

Not applicable. As a software viewing platform, the concept of "ground truth" (pathology, expert consensus, outcomes data) for its own performance is not directly relevant. Its performance is related to its ability to display images correctly, adhere to DICOM standards, and provide tools as specified, which are verified through non-clinical software testing.

8. The Sample Size for the Training Set:

Not applicable. This device is a software viewing platform, not an AI/Machine Learning algorithm that requires a "training set" of data.

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

Not applicable, as there is no training set mentioned or implied for this type of device.

Ask a specific question about this device

syngo.via RT Image Suite is a 3D and 4D image visualization, multi-modality manipulation and contouring tool that helps the preparation of treatments such as, but not limited to those performed with radiation (for example, Brachytherapy, Particle Therapy, External Beam Radiation Therapy).

It provides tools to view existing contours, create, edit, modify, copy contours of regions of the body, such as but not limited to, skin outline, targets and organs-at-risk. It also provides functionalities to create simple geometric treatment plans. Contours, images and treatment plans can subsequently be exported to a Treatment Planning System.

The software combines the following digital image processing and visualization tools:

• . Multi-modality viewing and contouring of anatomical, and multi-parametric images such as but not limited to CT, PET, PET/CT, MRI, Linac CBCT images
• Multiplanar reconstruction (MPR) thin/thick, minimum intensity projection (MIP), volume rendering technique (VRT)
• . Freehand and semi-automatic contouring of regions-of-interest on any orientation including oblique
• Automated Contouring on CT images
• . Creation of contours on images supported by the application without prior assignment of a planning CT
• Manual and semi-automatic registration using rigid and deformable registration ●
• . Supports the user in comparing, contouring, and adapting contours based on datasets acquired with different imaging modalities and at different time points
• . Supports multi-modality image fusion
• . Visualization and contouring of moving tumors and organs
• Management of points of interest including but not limited to the isocenter ●
• . Creation of simple geometric treatment plans
• Generation of a synthetic CT based on multiple pre-define MR acquisitions ●

The subject device with the current software version SOMARIS/8 VB70 is an image analysis software for viewing, manipulation, 3D and 4D visualization, comparison of medical images from multiple imaging modalities and for the segmentation of tumors and organs-at-risk, prior to dosimetric planning in radiation therapy. syngo.via RT Image Suite combines routine and advanced digital image processing and visualization tools for manual and software assisted contouring of volumes of interest, identification of points of interest, sending isocenter points to an external laser system, registering images and exporting final results. syngo.via RT Image Suite supports the medical professional with tools to use during different steps in radiation therapy case preparation.

For the current software version SOMARIS/8 VB70 the following already cleared features have been modified:

• Patient Marking
• Contouring
• 4D Features ●
• Basic Features of the subject device ●

The provided text describes the acceptance criteria and a study demonstrating that the lobe-based lung ventilation algorithm within the syngo.via RT Image Suite meets these criteria.

Here's the breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

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

• AI-based Lung Lobe Segmentation:

  • Test Set Sample Size: 18 radiotherapy patients (6 female, 12 male).
  • Data Provenance: Acquired from external clinical collaborations with radiotherapy departments in Europe and the Americas. The clinics used standard radiotherapy equipment and protocols to acquire the CT images.

• Lobe-based Lung Ventilation Algorithm (overall validation):

  • Test Set Sample Size: 108 CT datasets from 74 individual lung radiotherapy patients (25 female, 49 male; median age: 66 yrs, range 42-87 yrs).
  • Data Provenance: Not explicitly stated beyond being "clinical validation," but implies patient data.

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

The document does not explicitly state the number or qualifications of experts used to establish the ground truth for the test set. It mentions "annotated ground truth" for the AI segmentation component, but doesn't specify how many experts performed the annotation or their specific qualifications (e.g., "radiologist with 10 years of experience").

4. Adjudication Method for the Test Set

The document does not describe a specific adjudication method (e.g., 2+1, 3+1). It simply refers to "annotated ground truth" for the AI segmentation.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No MRMC comparative effectiveness study was described where human readers improved with AI assistance versus without AI assistance. The study focuses on evaluating the standalone performance of the AI component and the overall algorithm against established ground truth or standard methods.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was Done

Yes, a standalone evaluation was done for the AI-based component and the overall lobe-based lung ventilation algorithm.

• The AI-based component for lung lobe segmentation was tested on an independent cohort, with its performance (DICE score) being measured directly against annotated ground truth.
• The entire lobe-based lung ventilation algorithm's output was compared to literature ground truth (first subgroup) and spirometry (second subgroup), indicating a standalone assessment of the algorithm's output.

7. The Type of Ground Truth Used

• AI-based Lung Lobe Segmentation: Expert (or human) annotated ground truth for CT scans.
• Lobe-based Lung Ventilation (4D-CT Normal Breathing): Expected values obtained from scientific literature.
• Lobe-based Lung Ventilation (Breathhold CT): Vital capacity measurements from spirometry (a clinical standard for pulmonary function assessment, acting as outcomes data).

8. The Sample Size for the Training Set

• AI-based Lung Lobe Segmentation:
  • Training Set Sample Size: 8721 thoracic CT scans.
  • Validation Set Sample Size (for AI training): 969 thoracic CT scans.

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

The document states that the segmentation model was trained and validated on "annotated thoracic CT scans." This implies that human experts (likely radiologists or other medical professionals) manually segmented the lung lobes to create the ground truth for both the training and internal validation datasets. The specific process or number of experts for this annotation is not detailed.

Ask a specific question about this device

The MAGNETOM system is indicated for use as a magnetic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, spectroscopic images and/or spectra, and that displays the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images and/or spectra may also be produced. Depending on the region of interest, contrast agents may be used. These inages and/ or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician vield information that may assist in diagnosis.

The MAGNETOM system may also be used for imaging during interventional procedures when performed with MR compatible devices such as in-room displays and MR Safe biopsy needles.

MAGNETOM Lumina and MAGNETOM Vida Fit with software syngo MR XA50A include new software compared to the predicate devices, MAGNETOM Vida Fit with software syngo MR XA20A (K192924) and MAGNETOM Lumina with syngo MR XA31A (K203443). This software and some hardware components are transferred from the reference device MAGNETOM Vida with software syngo MR XA50A (K213693) as well as an imaging feature from MAGNETOM Vida with software syngo MR XA11A (K181433). A high-level summary of the transferred hardware and software is provided below:

Hardware (Vida Fit only)
Transferred Hardware:

• The Nexaris Dockable Table is a new variant of the MR patient table which is used for intraoperative or interventional imaging. It enables the patient transfer between OR tables and the MR system without repositioning on the MR patient table and vice versa during interventional procedures and surgeries. Additionally, it can be used for diagnostic imaging.
• The Nexaris Head Frame holds up to two Ultra Flex Large 18 coils. It can be used for head imaging in combination with the Nexaris Dockable Table when the patient is positioned on the transfer board but not pinned in a head clamp.
• Transferred MaRS Computer
  Transferred Coil:
• The Nexaris Spine 36 is used in combination with and without transfer board for body imaging on the Nexaris Dockable Table.
  Transferred modifications for hardware:
• The Beat Sensor is a contact less method for generating cardiac triggers as an alternative to the already existing ECG or pulse triggers. It is based on a measurement of the modulation of a weak magnetic Pilot Tone, caused by conformation changes in conductive tissues.
  Software
  Transferred Features and Applications: Vida Fit only:
• SVS EDIT is a special variant of the SVS SE pulse sequence type, which acquires two different spectra (one with editing pulses on resonance, one with editing pulses off resonance) within a single sequence.
• BEAT FQ nav allows the user to make use of navigator echo based respiratory gating for flow imaging to acquire 4D flow data. Both navigator echo based respiratory gating as well as flow imaging are part of the predicate device already. New is merely the combination of both.
• The HASTE interactive pulse sequence type extends the existing HASTE pulse sequence type by offering the possibility to interactively change imaging parameters.
• GRE_WAVE is a special variant of the GRE pulse sequence type which allows larger acceleration factors, measuring one or two contrasts. GRE Wave results in higher signal-to-noise ratio for larger acceleration factors which can be leveraged to allow fast high-resolution 3D susceptibility-weighted imaging.
• The myExam Prostate Assist provides an assisted and quided workflow for prostate imaging. This automated workflow leads to higher reproducibility of slice angulation and coverage; this may support exams not having to be repeated.
• Iniector coupling is a software application that allows the connection of certain contrast agent injectors to the MR system for simplified, synchronized contrast injection and examination start.
  Lumina onlv:
• Compressed Sensing GRASP-VIBE is intended to be used in dynamic and/or non-contrast liver examinations to support patients who cannot reliably hold their breath for a conventional breath-hold measurement.
  Lumina and Vida Fit:
• Deep Resolve Swift Brain is a protocol for fast routine brain imaging primarily based on echo planar imaging (EPI) pulse sequences. Its main enablers are multi-shot (ms) EPI pulse sequence types and a deep learning-based image reconstruction.
• Deep Resolve Boost is a novel deep learning-based image reconstruction alqorithm for 2D TSE data, which reconstructs images from k-space raw-data.
• BLADE diffusion is a multi-shot imaging method based on TSE or TGSE (when EPI factor > 1) readout and a BLADE trajectory with diffusion preparation to enable diffusion weighted imaging with reduced sensitivity to B0 inhomogeneity and reduced T2 decay caused image blurring.
• HASTE diffusion (HASTE DIFF) is a single-shot imaging method based on TSE readout with diffusion preparation to enable diffusion weighted imaging with reduced sensitivity to B0 inhomogeneity.
  Transferred Modifications for Features and Applications:
  Vida Fit only:
• The AbsoluteShim mode is a shimming procedure based on a 3-echo gradient echo protocol.
• The 3D ASL sequence (tgse_asl) now provides relCBF maps, by implementing an additional M0 scan and performing the corresponding reconstruction method. It also provides BAT maps in multiple inversion time(multi-TI) imaging.
  Lumina and Vida Fit:
• Fast GRE RefScan: A speed-optimized reference scan for GRAPPA and SMS kernel calibration for echo planar imaging pulse sequence types.
• Static Field Correction is a reconstruction option reducing susceptibilityinduced distortions and intensity variations.
• Deep Resolve Sharp is an interpolation algorithm which increases the perceived sharpness of the interpolated images. Functionality is available for different pulse sequence types. (Newly transferred to Vida Fit)
• Deep Resolve Gain is a reconstruction option which improves the SNR of the scanned imaqes. Functionality is available for different pulse sequence types. (Newly transferred to Vida Fit)
• The myExam Angio Advanced Assist provides an assisted and quided workflow for peripheral angiography examination using care bolus. The main advantage of this new workflow is a simplified and improved planning procedure of multi-station peripherical angiography measurements.
  Other transferred Modifications and / or Minor Changes
  Vida Fit only:
• Elastography-AddIn synchronizes settings between the Elastography sequence and the active driver.
• HASTE MoCo is an image-based motion correction in the average-dimension for the HASTE pulse sequence type.
• Coil independent pulse sequences remove the coil information from the pulse sequences and generate this information during run-time from automatic coil detection and localization.
• The Needle Intervention AddIn provides a user interface for workflow improvement of MR-quided needle interventions under real-time imaging conditions. It supports planning a needle trajectory, laser-based localization of the entry point as well as automatic slice positioning.
• The PhaseRev Dot Addin/Component supports the measurement workflow of the user by automatically flipping the direction of the phase encoding gradient.
• The adjustment mode "offcenter" triggers a transmitter adjustment method that is specialized for offcenter imaging. The transmitter adjustment determines the RF voltage that is required to excite a certain B1 field.
  Lumina and Vida Fit:
• TSE MoCo is an image-based motion correction in the average-dimension for the TSE pulse sequence type.
• MR Breast Biopsy is improved with an automatic fiducial detection.

The provided text primarily focuses on the substantial equivalence of the MAGNETOM Lumina and MAGNETOM Vida Fit with syngo MR XA50A to predicate devices. It does not include detailed information regarding specific acceptance criteria, device performance metrics, or the study design (e.g., sample sizes, expert qualifications, ground truth methods) that would typically be found in a clinical or performance study report.

Therefore, I cannot extract the requested information about acceptance criteria and the study proving the device meets them from the given document.

The document states:

• "No additional clinical tests were conducted to support substantial equivalence for the subject devices." (Page 9)
• The primary testing conducted was "Verification and validation" of transferred hardware and software features against "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices / 21 CFR §820.30" (Page 9).
• The conclusion is that "the results from each set of tests demonstrate that the devices perform as intended and are thus substantially equivalent to the predicate devices to which they have been compared." (Page 9).

This indicates that the submission relies on demonstrating equivalence to previously cleared devices through non-clinical verification and validation, rather than presenting a de novo performance study with specific acceptance criteria.

Ask a specific question about this device

syngo.CT Dual Energy is designed to operate with CT images based on two different X-ray spectra.

The various materials of an anatomical region of interest have different attenuation coefficients, which depend on the used energy. These differences provide information on the chemical composition of the scanned body materials. syngo.CT Dual Energy combines images acquired with low and high energy spectra to visualize this information. Depending on the region of interest, contrast agents may be used.

The general functionality of the syngo.CT Dual Energy application is as follows:

• · Monoenergetic 1)
• · Brain Hemorrhage
• · Gout Evaluation
• · Lung Vessels
• · Heart PBV
• · Bone Removal
• · Lung Perfusion
• · Liver VNC
• · Monoenergetic Plus 1)
• · Virtual Unenhanced 1)
• Bone Marrow
• · Hard Plaques
• Rho/Z
• · Kidney Stones 2)
• · SPR (Stopping Power Ratio)
• · SPP (Spectral Post-Processing Format) 1)
• · Optimum Contrast 1)

The availability of each feature depends on the Dual Energy scan mode.

1. This functionality supports data from Photon-Counting CT scanners.

2. Kidney Stones is designed to support the visualization of the chemical composition of kidney stones and especially the differentiation between uric acid stones. For full identification of the kidney stone, additional clinical information should be considered such as patient history and urine testing. Only a well-trained radiologist can make the final diagnosis upon consideration of all available information. The accuracy of identification is decreased in obese patients.

Dual energy offers functions for qualitative and quantitative post-processing evaluations. syngo.CT Dual Energy is a post-processing application consisting of several post-processing application classes that can be used to improve the visualization of the chemical composition of various energy dependent materials in the human body when compared to single energy CT. Depending on the organ of interest, the user can select and modify different application classes or parameters and algorithms.

Different body regions require specific tools that allow the correct evaluation of data sets. syngo.CT Dual Energy provides a range of application classes that meet the requirements of each evaluation type. The different application classes for the subject device can be combined into one workflow.

Based on the provided text, the acceptance criteria and the study proving the device meets these criteria can be summarized as follows:

The document describes software verification and validation, non-clinical testing, and an evaluation of specific application classes for Photon Counting Data. However, it does not provide a quantitative table of acceptance criteria for specific performance metrics (e.g., sensitivity, specificity, accuracy) or detailed clinical study results with human readers (MRMC study). The testing described focuses on technical performance and consistency with expected phantom values and visual comparison with clinical data, rather than diagnostic accuracy or clinical effectiveness in a human-in-the-loop setting.

Here's a breakdown of the available information:

1. Acceptance Criteria and Reported Device Performance

The document states that "all software specifications have met the acceptance criteria" and "The testing results support that all the software specifications have met the acceptance criteria." However, the document does not explicitly list the specific acceptance criteria in a table format with corresponding reported device performance values for metrics like accuracy, sensitivity, or specificity.

Instead, the performance data provided focuses on:

• Software Verification and Validation: Conformance with "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices," risk analysis completion, and mitigation of identified hazards.
• Non-Clinical Testing: Integration and functional tests were conducted to demonstrate the ability of included features. "The results of these tests demonstrate that the subject device performs as intended."
• Evaluation of application classes for Photon Counting Data:
  • Monoenergetic Plus application class: "calculated values from phantom scans agreed well with the expected ones. Clinical data showed no artifacts. The iodine contrast clearly increased with lower keV settings and decreased with higher ones."
  • Virtual Unenhanced application class: "demonstrated that virtual non-contrast images and iodine concentration can be calculated from spectral data acquired at the NAEOTOM Alpha." In phantom scans, "the measured iodine concentration agrees well with the known iodine concentration. The VNC values are good approximations of the expected water value for all tested iodine concentrations." In clinical data, "the image impression of the virtual non-contrast images was compared with true non-contrast images. Measurements showed good agreement of CT values in the VNCs with the values in the TNCs."

No quantitative performance metrics (e.g., sensitivity, specificity, AUC) or a direct comparison to specific numerical acceptance criteria are provided in the document.

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

The document mentions "phantom scans" and "clinical data" for the evaluation of the Monoenergetic Plus and Virtual Unenhanced application classes.

• Phantom Scans: "Multi-Energy CT Phantom (Sun Nuclear Corporation, Melbourne, Florida, USA) was scanned at a NAETOM Alpha."
• Clinical Data: Used for visual comparison and measurement of CT values. The text refers to "clinical data" in general without specifying the sample size (number of patients/cases).
• Data Provenance: Not specified (e.g., country of origin). The data from the NAETOM Alpha appears to be prospectively acquired for testing purposes. It is not stated whether the clinical data used for comparison was retrospective or prospective.

3. Number of Experts Used to Establish Ground Truth and Qualifications

The document mentions that for the Kidney Stones feature, "Only a well-trained radiologist can make the final diagnosis upon consideration of all available information." However, it does not specify the number of experts used to establish ground truth for the test set or their specific qualifications (e.g., years of experience, subspecialty) for the evaluations described (phantom studies or clinical data comparisons).

4. Adjudication Method for the Test Set

The document does not describe any formal adjudication method (e.g., 2+1, 3+1 consensus) for establishing ground truth for the "clinical data" used. The evaluations seem to rely on technical comparisons for phantom data and general observation/measurement agreement for clinical data.

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

No MRMC comparative effectiveness study was conducted or reported. The submission focuses on technical validation and comparison of the device's outputs to expected values and impressions, rather than measuring human reader performance with and without AI assistance.

6. Standalone (Algorithm Only) Performance Study

The study appears to be an algorithm-only performance evaluation in terms of its ability to generate specific types of images/data (monoenergetic images, virtual non-contrast images, iodine concentrations) and the agreement of these outputs with expected or true values (for phantom data) and visual/measurement comparisons (for clinical data). However, no specific standalone diagnostic performance metrics (e.g., sensitivity, specificity for disease detection) are reported.

7. Type of Ground Truth Used

• Technical/Physical Ground Truth: For phantom studies, the "known iodine concentration" and "expected" values serve as ground truth.
• Reference Image Ground Truth: For the Virtual Unenhanced application, "true non-contrast images" are used as a reference for comparison.
• Expert Interpretive Ground Truth: While "well-trained radiologist" is mentioned in the Indications for Use for Kidney Stones, the actual methodology for establishing ground truth for the clinical data used in the evaluation is not detailed beyond "image impression" and "measurements." It's an implicit expert consensus by a "well-trained radiologist" who would interpret the images, but the methodology for establishing this is not formalized in the provided text.

8. Sample Size for the Training Set

The document does not specify the sample size for the training set used to develop the syngo.CT Dual Energy algorithms. The focus of this submission is on verification and validation of a device modification, not initial algorithm development.

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

The document does not describe how the ground truth for the training set was established, as it pertains to the validation of a device modification rather than the initial algorithm development.

Ask a specific question about this device

syngo.CT Applications is a set of software applications for advanced visualization, measurement, and evaluation for specific body regions.

This software package is designed to support the radiologists and physicians from emergency medicine, specialty care, urgent care, and general practice e.g. in the:

• · Evaluation of perfusion of organs and tumors and myocardial tissue perfusion
• · Evaluation of bone structures and detection of bone lesions
• · Evaluation of CT images of the heart
• · Evaluation of the coronary lesions
• · Evaluation of the mandible and maxilla
• · Evaluation of dynamic vessels and extended phase handling
• · Evaluation of the liver and its intrahepatic vessel structures to identify the vascular territories of sub-vessel systems in the liver
• · Evaluation of neurovascular structures
• Evaluation of the lung parenchyma
• · Evaluation of non-enhanced Head CT images
• · Evaluation of vascular lesions

The syngo.CT Applications are syngo based post-processing software applications to be used for viewing and evaluating CT images provided by a CT diagnostic device and enabling structured evaluation of CT images.

The syngo.CT Applications is a combination of thirteen (13) former separately cleared medical devices which are now handled as features / functionalities within syngo.CT Applications. These functionalities are combined unchanged compared to their former cleared descriptions; however, some minor enhancements and improvements are made for the application syngo.CT Pulmo 3D only.

The provided document is a 510(k) summary for syngo.CT Applications, which is a consolidation of thirteen previously cleared medical devices. The document explicitly states that "The testing supports that all software specifications have met the acceptance criteria" and "The result of all testing conducted was found acceptable to support the claim of substantial equivalence." However, it does not explicitly define specific acceptance criteria (e.g., target accuracy, sensitivity, specificity values) for the device's performance or detail the specific studies that prove these criteria are met. Instead, it relies on the premise that the functionalities remain unchanged from the previously cleared predicate devices, with only minor enhancements to one application (syngo.CT Pulmo 3D).

Therefore, based on the provided text, I cannot fill in precise quantitative values for acceptance criteria or specific study results for accuracy, sensitivity, or specificity. The information provided heavily emphasizes software verification and validation, risk analysis, and adherence to consensus standards, rather than detailing a comparative effectiveness study or standalone performance metrics against a defined ground truth.

Here's a breakdown of the available information and what is missing:

1. Table of acceptance criteria and the reported device performance:

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective):

• Sample Size for Test Set: Not specified in the document.
• Data Provenance: Not specified in the document.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):

• Number of Experts: Not specified in the document.
• Qualifications of Experts: Not specified in the document.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

• Adjudication Method: Not specified in the document.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:

• MRMC Study Done: No. The document does not mention any MRMC comparative effectiveness study where human readers' performance with and without AI assistance was evaluated. The submission focuses on the consolidation of existing, cleared applications.
• Effect Size of Improvement: Not applicable, as no MRMC study is reported.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• Standalone Study Done: Yes, implicitly. The document states, "The testing supports that all software specifications have met the acceptance criteria," suggesting that the software's performance was verified and validated independent of human interpretation to ensure its functionalities (visualization, measurement, evaluation) behave as intended. However, specific metrics (e.g., accuracy of a measurement tool compared to a gold standard) are not provided. The phrase "algorithm only" might not be fully accurate here given the device is a visualization and evaluation tool for human use, not an autonomous diagnostic AI.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

• Type of Ground Truth: Not explicitly specified. Given the nature of visualization and evaluation tools, it would likely involve comparisons to known values, measurements, or expert-reviewed datasets, but the document does not detail this.

8. The sample size for the training set:

• Training Set Sample Size: Not applicable/Not specified. The document describes the device as a consolidation of existing, cleared software applications with "minor enhancements and improvements" only to syngo.CT Pulmo 3D. It does not indicate that new machine learning models requiring large training sets were developed for this specific submission; rather, it refers to the performance of existing, cleared applications.

9. How the ground truth for the training set was established:

• How Ground Truth for Training Set was Established: Not applicable/Not specified, for the same reasons as point 8. The document does not describe a new AI model training process for this submission.

Summary of Device Rationale:

The core of this 510(k) submission is the consolidation of thirteen previously cleared syngo.CT applications into a single "syngo.CT Applications" product. The applicant, Siemens Medical Solutions USA, Inc., states that the functionalities within this combined product are "unchanged compared to their former cleared descriptions" with only "minor enhancements and improvements" in syngo.CT Pulmo 3D (specifically regarding color assignments for lobe borders).

The document asserts that "The performance data demonstrates continued conformance with special controls for medical devices containing software." It also states, "The risk analysis was completed, and risk control implemented to mitigate identified hazards. The testing results support that all the software specifications have met the acceptance criteria. Testing for verification and validation of the device was found acceptable to support the claims of substantial equivalence."

This implies that the "acceptance criteria" largely revolve around the continued functional performance and adherence to specifications of the already cleared individual applications, plus verification of the minor changes to syngo.CT Pulmo 3D, and the successful integration into a single software package. However, quantitative performance metrics for the device against specific clinical tasks are not provided in this 510(k) summary document, as the submission focuses on the substantial equivalence of the consolidated product to its predicate devices, rather than presenting new clinical efficacy data.

Ask a specific question about this device

syngo.via RT Image Suite is a 3D and 4D image visualization, multi-modality manipulation and contouring tool that helps the preparation of treatments such as, but not limited to those performed with radiation (for example, Brachytherapy, Particle Therapy, External Beam Radiation Therapy).

It provides tools to view existing contours, create, edit, modify, copy contours of regions of the body, such as but not limited to, skin outline, targets and organs-at-risk. It also provides functionalities to create simple geometric treatment plans. Contours, images and treatment plans can subsequently be exported to a Treatment Planning System.

The software combines the following digital image processing and visualization tools:

• . Multi-modality viewing and contouring of anatomical, and multi-parametric images such as but not limited to CT, PET, PET/CT, MRI, Linac CBCT images
• Multiplanar reconstruction (MPR) thin/thick, minimum intensity projection (MIP), volume ● rendering technique (VRT)
• . Freehand and semi-automatic contouring of regions-of-interest on any orientation including oblique
• Automated Contouring on CT images
• . Creation of contours on images supported by the application without prior assignment of a planning CT
• Manual and semi-automatic registration using rigid and deformable registration
• Supports the user in comparing, contouring, and adapting contours based on datasets acquired with different imaging modalities and at different time points
• . Supports multi-modality image fusion
• . Visualization and contouring of moving tumors and organs
• Management of points of interest including but not limited to the isocenter ●
• Creation of simple geometric treatment plans ●
• Generation of a synthetic CT based on multiple pre-define MR acquisitions ●

The subject device with the current software version SOMARIS/8 VB60 is an image analysis software for viewing, manipulation, 3D and 4D visualization, comparison of medical images from multiple imaging modalities and for the segmentation of tumors and organs-at-risk, prior to dosimetric planning in radiation therapy. syngo.via RT Image Suite combines routine and advanced digital image processing and visualization tools for manual and software assisted contouring of volumes of interest, identification of points of interest, sending isocenter points to an external laser system, registering images and exporting final results. syngo.via RT Image Suite supports the medical professional with tools to use during different steps in radiation therapy case preparation.

For the current software version SOMARIS/8 VB60 the following already cleared features have been modified:

• . Reference Point Management
• Patient Marking ●
• Contouring / Routine Contouring
• Structure Set Management ●
• Synthetic CT
• Basic Feature of syngo,via RT Image Suite

The provided documentation relates to the Siemens syngo.via RT Image Suite, specifically describing its 510(k) premarket notification for a new software version (SOMARIS/8 VB60) that includes an AI-based algorithm for synthetic CT generation.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Reported Device Performance

The document describes performance criteria for the AI-based algorithm for generating synthetic CT images from MR images. While not presented in a formal table with specific thresholds, the text outlines the key metrics evaluated and the results.

The document states that the geometric deviation of < 1 mm is "below the voxel resolution and therefore not clinically relevant," implying this meets the clinical relevance standard.

Study Details

The document refers to "performance tests (Non-clinical test reports)" and a "Summary of the Performance Evaluation of the Algorithm" to demonstrate meeting acceptance criteria.

1. Sample Size Used for the Test Set and Data Provenance:

• Sample Size: The document does not specify the exact sample size used for the test set for the AI algorithm. It only mentions "independent data."
• Data Provenance: The document does not specify the country of origin or whether the data was retrospective or prospective. It only states the AI algorithm was "tested on independent data."

2. Number of Experts Used to Establish Ground Truth and Qualifications:

• Number of Experts: Not specified.
• Qualifications: Not specified.
  • The ground truth methodology is not described in terms of expert consensus for the test set.

3. Adjudication Method for the Test Set:

• Not specified. The document states "automated bench tests" were used for geometric and HU accuracy, implying an objective, quantitative comparison rather than reader adjudication.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

• No MRMC study appears to have been performed or reported in this document for the AI-based synthetic CT algorithm.
• The evaluation focused on quantitative metrics (geometric and HU accuracy) of the algorithm's output, not on human reader performance with or without AI assistance.

5. Standalone (Algorithm Only) Performance:

• Yes, a standalone performance evaluation was conducted. The "Summary of the Performance Evaluation of the Algorithm" specifically details the AI-based algorithm's performance on "geometric fidelity and HU accuracy using automated bench tests." This implies an algorithm-only evaluation without human intervention in the performance measurement.

6. Type of Ground Truth Used:

• The ground truth for the synthetic CT evaluation appears to have been measured/reference CT images against which the synthetic CTs were compared for geometric and Hounsfield Unit (HU) accuracy. This falls under reference standard/objective measurement data, implying accuracy was determined by comparing the AI's output to a known, accurate CT.

7. Sample Size for the Training Set:

• Not specified. The document does not provide details about the training set size for the deep-learning algorithm.

8. How the Ground Truth for the Training Set Was Established:

• Not specified. The document indicates that the "algorithm for brain and pelvis synthetic CTs has been changed from Atlas based to a deep-learning algorithm." However, it does not describe how the ground truth for training this deep-learning algorithm was established (e.g., source of data, annotation methods, expert review, etc.).

Ask a specific question about this device

AI-Rad Companion Prostate MR is a post-processing image analysis software that assists clinicians in viewing, manipulating, analyzing and evaluating MR prostate images for US guided MR-US fusion biopsy support.

AI-Rad Companion Prostate MR aims to assist the radiologist in the preparation of MR prostate images for targeted biopsies of the prostate gland using MR-Ultrasound fusion biopsy. It allows the radiologist to communicate the location and spatial extent of lesions and the prostate volume in prostate MR images to a urologist in order to help perform biopsies.

AI-Rad Companion Prostate MR is a cloud-based image processing software that provides quantitative and qualitative information based on prostate MR DICOM images. More specifically, it provides information on the prostate volume which can be used to support the planning of prostate biopsies in the case of ultrasound guided MR-US fusion biopsies of the prostate gland. It is enabled via artificial intelligence algorithms and a cloud infrastructure.

The primary features of AI-Rad Companion Prostate MR include:

• Automatic prostate segmentation and volume estimation, with the possibility of manual adjustments
• Manual determination of location and size of lesions in a suitable user interface
• Calculation of the PSA density, based on the input of the PSA value of the patient by the clinical user
• Export in a suitable format for reading and archiving in PACS, as well as in a second format that can be imported by ultrasound systems (e.g. RTStruct), allowing the urologist to perform targeted MR-US fusion biopsy

The document provided refers to AI-Rad Companion Prostate MR and states that no clinical tests were conducted to test the performance and functionality of the modifications introduced within AI-Rad Companion Prostate MR (Page 8, Section 9. Clinical Tests). Therefore, the information requested regarding acceptance criteria and performance based on a clinical study cannot be fully provided from the given text.

However, based on the non-clinical tests and the comparison to the predicate device, here's what can be inferred and stated:

1. Table of acceptance criteria and the reported device performance

The document does not provide specific quantitative acceptance criteria or reported device performance metrics from a clinical study. It mentions non-clinical tests were conducted to assess performance claims and substantial equivalence. These tests included functionality, software validation, and bench testing (Unit, System, and Integration tests). All testable requirements in the Requirement Specifications and Risk Analysis were verified.

2. Sample size used for the test set and the data provenance

Since no clinical tests were conducted, details about a clinical test set are not available. For the non-clinical software "bench" testing, sample sizes for the test data are not explicitly stated in the provided text. The data provenance (e.g., country of origin, retrospective/prospective) for these internal tests is also not mentioned.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

The document does not mention the use of experts to establish ground truth for a clinical test set, as no clinical tests were performed. For internal software testing, the ground truth would typically be defined by design specifications and expected outputs.

4. Adjudication method for the test set

Not applicable, as no clinical test set requiring adjudication by experts is described.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. The document explicitly states: "No clinical tests were conducted to test the performance and functionality of the modifications introduced within AI-Rad Companion Prostate MR." (Page 8, Section 9. Clinical Tests).

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done

The document indicates that software "bench" testing was performed, which would be a form of standalone testing for the algorithm's functionality and performance against its requirements. However, specific results or detailed methodologies are not provided for "algorithm only" performance. The device's functionalities, such as "Automatic prostate segmentation and volume estimation," imply standalone algorithm components.

7. The type of ground truth used

For the non-clinical software testing, the ground truth would be based on the software's design specifications and expected outputs as defined by the developers (e.g., correct segmentation results against internal references, accurate volume calculations based on known inputs).

8. The sample size for the training set

The document does not mention the sample size for a training set. While the device utilizes "artificial intelligence algorithms" (Page 5), details about the training data used to develop these algorithms are not provided within this document.

9. How the ground truth for the training set was established

The document does not provide information on how the ground truth for any potential training set was established, as details about the AI algorithm's development and training are not included.

Ask a specific question about this device