Search Results
Found 1 results
510(k) Data Aggregation
(86 days)
The Siemens Biograph Horizon PET/CT systems are combined X-Ray Computed Tomography (CT) and Positron Emission Tomography (PET) scanners that provide registration and fusion of high resolution physiologic and anatomic information.
The CT component produces cross-sectional images of the body by computer reconstruction of X-Ray transmission data from either the same axial plane taken at different angles or spiral planes taken at different angles. The PET subsystem images and measures the distribution of PET radiopharmaceuticals in humans for the purpose of determining various metabolic (molecular) and physiologic functions within the human body and utilizes the CT for fast attenuation correction maps for PET studies and precise anatomical reference for the fused PET and CT images.
The system maintains independent functionality of the CT and PET devices, allowing for single modality CT and / or PET diagnostic imaging. These systems are intended to be utilized by appropriately trained health care professionals to aid in detecting, localizing, diagnosing, staging and re staging of lesions, tumors, disease and organ function for the evaluation of diseases and disorders such as, but not limited to, cardiovascular disease, neurological disorders and cancer. The images produced by the system can also be used by the physician to aid in radiotherapy treatment planning and interventional radiology procedures.
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 Biograph Horizon PET/CT systems are combined multi-slice X-Ray Computed Tomography and Positron Emission Tomography scanners. These systems are designed for whole body oncology, neurology and cardiology examinations. The Biograph Horizon systems provide registration and fusion of high-resolution metabolic and anatomic information from the two major components of each system (PET and CT). Additional components of the system include a patient handling system and acquisition and processing workstations with associated software.
Biograph Horizon software is a command-based program used for patient management, data management, scan control, image reconstruction and image archival and evaluation. All images conform to DICOM imaging format requirements.
The software for the Biograph Horizon systems which is the subject of this application is substantially equivalent to the commercially available Biograph Horizon software. Modifications include, corrections to software anomalies and addition of new software features, including:
- OncoFreeze
- OncoFreeze AI (Data Driven Gating)
- CardioFreeze
- FlowMotion Multi-Parametric PET AI ●
- PET FAST Planning (FlowMotion Al)
- FAST PET Workflow
- QualityGuard
- . Updates to HD FoV
- Updates to PET DICOM dose Report
- Whole Body Scatter Correction
Additionally, minor modifications have been made to the computers due to obsolescence issues and to the controllers of the PHS for cost improvement. These changes do not affect system performance characteristics and have no impact on safety or effectiveness.
The Siemens Biograph Horizon PET/CT system (K193178) underwent performance testing against NEMA NU2-2018 standards for its PET subsystem. The device also had specific evaluations for new software features: OncoFreeze AI, OncoFreeze/CardioFreeze, FlowMotion Multi-Parametric PET AI, FlowMotion AI (PET FAST Planning), QualityGuard, and Whole Body Scatter Correction.
Here's the breakdown of the acceptance criteria and study information:
1. Table of Acceptance Criteria and Reported Device Performance:
| Performance Criteria | Acceptance | Reported Device Performance | Study Performed For |
|---|---|---|---|
| Resolution - Full Size | PET (NEMA NU2-2018) | ||
| Transverse Resolution FWHM @ 1 cm | ≤ 4.7 mm | Pass | PET (NEMA NU2-2018) |
| Transverse Resolution FWHM @ 10 cm | ≤ 5.5 mm | Pass | PET (NEMA NU2-2018) |
| Transverse Resolution FWHM @ 20 cm | ≤ 7.6 mm | Pass | PET (NEMA NU2-2018) |
| Axial Resolution FWHM @ 1 cm | ≤ 5.0 mm | Pass | PET (NEMA NU2-2018) |
| Axial Resolution FWHM @ 10 cm | ≤ 7.0 mm | Pass | PET (NEMA NU2-2018) |
| Axial Resolution FWHM @ 20 cm | ≤ 11.3 mm | Pass | PET (NEMA NU2-2018) |
| Resolution - 256 x 256 | PET (NEMA NU2-2018) | ||
| Transverse Resolution FWHM @ 1 cm | ≤ 7.3 mm | Pass | PET (NEMA NU2-2018) |
| Transverse Resolution FWHM @ 10 cm | ≤ 7.6 mm | Pass | PET (NEMA NU2-2018) |
| Transverse Resolution FWHM @ 20 cm | ≤ 8.9 mm | Pass | PET (NEMA NU2-2018) |
| Axial Resolution FWHM @ 1 cm | ≤ 6.1 mm | Pass | PET (NEMA NU2-2018) |
| Axial Resolution FWHM @ 10 cm | ≤ 7.3 mm | Pass | PET (NEMA NU2-2018) |
| Axial Resolution FWHM @ 20 cm | ≤ 11.9 mm | Pass | PET (NEMA NU2-2018) |
| Count Rate / Scatter / Sensitivity | PET (NEMA NU2-2018) | ||
| Sensitivity @435 keV LLD | ≥ 5.8 cps/kBq (≥ 10.9 cps/kBq for TrueV) | Pass | PET (NEMA NU2-2018) |
| Count Rate peak NECR | ≥ 78 kcps @ ≤ 26 kBq/cc (≥ 135 kcps @ ≤ 26 kBq/cc for TrueV) | Pass | PET (NEMA NU2-2018) |
| Count Rate peak trues | ≥285 kcps @ ≤ 53 kBq/cc (≥ 465 kcps @ ≤ 42 kBq/cc for TrueV) | Pass | PET (NEMA NU2-2018) |
| Scatter Fraction at peak NECR | ≤ 40% | Pass | PET (NEMA NU2-2018) |
| Mean bias (%) at peak NEC | ≤ +/- 6% | Pass | PET (NEMA NU2-2018) |
| Image Quality (4 to 1) - (% Contrast / Background Variability) | PET (NEMA NU2-2018) | ||
| 10mm sphere | ≥ 10% / ≤ 10% | Pass | PET (NEMA NU2-2018) |
| 13mm sphere | ≥ 25% / ≤ 10% | Pass | PET (NEMA NU2-2018) |
| 17mm sphere | ≥ 40% / ≤ 10% | Pass | PET (NEMA NU2-2018) |
| 22mm sphere | ≥ 55% / ≤ 10% | Pass | PET (NEMA NU2-2018) |
| 28mm sphere | ≥ 60% / ≤ 10% | Pass | PET (NEMA NU2-2018) |
| 37mm sphere | ≥ 65% / ≤ 10% | Pass | PET (NEMA NU2-2018) |
| Co-Registration Accuracy | System Performance | ||
| Max Error | ≤ 5 mm | Pass | System Performance |
| OncoFreeze AI (Data Driven Gating) | Change in SUVmax / SUVmean / Volume similar to Anzai-based gating | Pass | Comparison of motion-corrected images |
| ΔSUVmax (relative to static) | N/A (compared to Anzai) | 27% ± 22% (vs. 29% ± 22%) | OncoFreeze AI vs. Anzai-based OncoFreeze |
| ΔSUVmean (relative to static) | N/A (compared to Anzai) | 26% ± 22% (vs. 27% ± 22%) | OncoFreeze AI vs. Anzai-based OncoFreeze |
| ΔVolume (relative to static) | N/A (compared to Anzai) | -31% ± 19% (vs. -34% ± 23%) | OncoFreeze AI vs. Anzai-based OncoFreeze |
| FlowMotion AI (PET FAST Planning) | Successful identification of bed ranges at high percentage | 87.1% to 100% | Evaluation of accuracy in defining bed ranges |
| QualityGuard | Time savings and reduction in staff exposure | Up to 30 mins/day savings, 98% exposure reduction | Evaluation of time and exposure reduction |
| Whole Body Scatter Correction | Reduced difference from ground truth in ROI measurements compared to Single Bed Scatter Correction | Pass | Simulation study of phantom ROIs (ground truth) |
| Representative ROI close to phantom (high signal) | N/A (compared to Single Bed) | -2% (vs. +87%) | Whole Body Scatter Correction vs. Single Bed Scatter Correction |
| Representative ROI close to phantom (low signal) | N/A (compared to Single Bed) | -3% (vs. -42%) | Whole Body Scatter Correction vs. Single Bed Scatter Correction |
| Representative ROI inside phantom | N/A (compared to Single Bed) | -0.4% (vs. +0.5%) | Whole Body Scatter Correction vs. Single Bed Scatter Correction |
2. Sample size used for the test set and the data provenance:
- PET Testing (NEMA NU2-2018): The document implies physical phantom testing which is standard for NEMA NU2-2018. The number of samples (scans or measurements) from the phantom is not explicitly stated, but these tests are typically performed on two different configurations: a 3-ring version and a 4-ring version (TrueV). Data provenance is from physical phantom measurements on the device itself.
- OncoFreeze AI (Data Driven Gating): The document mentions "motion-corrected (OncoFreeze) images comparing Anzai based gating and deviceless gating." It does not specify the number of patient images or specific phantom studies. The provenance isn't explicitly stated as retrospective or prospective, nor a country of origin.
- FlowMotion AI (PET FAST Planning): No specific sample size (i.e., number of cases or scans) is given. The study evaluated "configuration set" effectiveness. The provenance is not explicitly stated.
- Whole Body Scatter Correction: This was a "simulation study of phantom." The sample size (number of ROIs or simulation runs) is not specified. Provenance is simulation data based on a phantom.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- For the NEMA NU2-2018 PET testing, the ground truth for performance metrics (resolution, sensitivity, scatter fraction, image quality) is established by the known physical properties and activity concentrations of the NEMA phantom according to the standard. No human experts are involved in establishing this ground truth.
- For OncoFreeze AI, the "ground truth" for motion correction comparison is implicitly the static image or the Anzai-based gated images acting as a reference for comparison of SUV and Volume changes. No explicit expert involvement for ground truth is mentioned.
- For FlowMotion Multi-Parametric PET AI, the "ground truth" for improved noise characteristics and delineation is a comparison against the Patlak transformation performed during reconstruction without the automatic feature. This implies qualitative assessment, but no expert details are given.
- For FlowMotion AI (PET FAST Planning), the "ground truth" for correctly defined bed ranges would be based on the known or intended bed ranges. No expert involvement in establishing this ground truth is mentioned.
- For QualityGuard, the "ground truth" for time savings and exposure reduction would be direct measurement or comparison to previous workflows. No expert involvement for ground truth is mentioned.
- For Whole Body Scatter Correction, "ground truth" was established by a "simulation study of phantom," implying that the true values for activity in different regions of interest were known from the simulation setup. No human experts are involved in establishing this ground truth; it's inherent to the simulation.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
No adjudication method for expert review of images or data is mentioned for any of the tests described. The tests are primarily technical performance evaluations (NEMA standards, quantitative comparisons, simulation studies) or workflow efficiency observations.
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 multi-reader multi-case (MRMC) comparative effectiveness study evaluating human reader improvement with AI assistance is mentioned in the provided text. The AI features mentioned (OncoFreeze AI, FlowMotion Multi-Parametric PET AI, FlowMotion AI (PET FAST Planning)) are related to image processing and workflow automation rather than direct diagnostic assistance for human readers in a comparative effectiveness setting.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
The device itself is a diagnostic imaging system (PET/CT), not an AI algorithm that provides a diagnostic output to a human. The AI features described within the device, such as OncoFreeze AI and FlowMotion AI, are designed to improve image quality, correct for motion, or automate planning. The performance metrics focus on the output of these features (e.g., changes in SUV, bed range identification accuracy, scatter correction accuracy) as part of the overall imaging system's functionality. Therefore, these features are evaluated in a standalone manner within the machine's operation, without explicit human-in-the-loop performance studies described here.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- NEMA NU2-2018 PET testing: Physical phantom with known activity concentrations and physical dimensions.
- OncoFreeze AI: Comparison to static images and Anzai-based gating, which implicitly serve as a reference or "ground truth" for the motion-corrected measurements.
- FlowMotion Multi-Parametric PET AI: Comparison to non-automated Patlak transformation results.
- FlowMotion AI (PET FAST Planning): Known or intended bed ranges from the configuration setup.
- QualityGuard: Measured time and staff exposure.
- Whole Body Scatter Correction: Known values from a "simulation study of phantom."
In summary, the ground truth types are primarily known physical properties of phantoms (or simulated phantoms), comparative measurements against established methods, or direct measurements of efficiency/safety.
8. The sample size for the training set:
The document does not provide any information about the training set size for the AI features (OncoFreeze AI, FlowMotion AI).
9. How the ground truth for the training set was established:
Since no information on training sets is provided, there is also no information on how ground truth for any training sets was established.
Ask a specific question about this device
Page 1 of 1