(29 days)
syngo.via molecular imaging (MI) workflows comprise medical diagnostic applications for viewing, manipulation, quantification, analysis and comparison of medical images from single or multiple imaging modalities with one or more time-points. These workflows support functional data, such as positron emission tomography (PET) or nuclear medicine (NM), as well as anatomical datasets, such as computed tomography (CT) or magnetic resonance (MR), syngo.via MI workflows can perform harmonization of SUV (PET) across different PET systems or different PET reconstruction methods.
syngo.via MI workflows are intended to be utilized by appropriately trained health care professionals to aid in the management of diseases, including those associated with oncology, cardiology, neurology, and organ function. The images and results produced by the syngo.via MI workflows can also be used by the physician to aid in radiotherapy treatment planning.
syngo.via MI Workflows (including Scenium and syngo MBF applications) is a multi-modality postprocessing software only medical device intended to aid in the management of diseases, including those associated with oncology, cardiology, neurology, and organ function. The syngo.via MI Workflows applications are part of a larger syngo.via client/server system which is intended to be installed on common IT hardware. The hardware itself is not seen as part of the syngo.via MI Workflows medical device.
The syngo.via MI Workflows software addresses the needs of the following typical users of the product:
- Reading Physician / Radiologist – Reading physicians are doctors who are trained in interpreting patient scans from PET, SPECT and other modality scanners. They are highly detail oriented and analyze the acquired images for abnormalities, enabling ordering physicians to accurately diagnose and treat scanned patients. Reading physicians serve as a liaison between the ordering physician and the technologists, working closely with both.
- Technologist – Nuclear medicine technologists operate nuclear medicine scanners such as PET and SPECT to produce images of specific areas and states of a patient's anatomy by administering radiopharmaceuticals to patients orally or via injection. In addition to administering the scan, the technologist must properly select the scan protocol, keep the patient calm and relaxed, monitor the patient's physical health during the protocol and evaluate the quality of the images. Technologists work very closely with physicians, providing them with quality-checked scan images.
The software has been designed to integrate the clinical workflow for the above users into a serverbased system that is consistent in design and look with the base syngo.via platform and other syngo.via software applications. This ensures a similar look and feel for radiologists that may review multiple types of studies from imaging modalities other than Molecular Imaging, such as MR.
The syngo.via MI workflows software supports integration through DICOM transfers of positron emission tomography (PET) or nuclear medicine (NM) data, as well as anatomical datasets, such as computed tomography (CT) or magnetic resonance (MR).
Although data is automatically imported into the server based on predefined configurations through the hospital IT system, data can also be manually imported from external media, including CD, external mass storage devices, etc.
The Siemens syngo.via platform and the applications that reside on it, including syngo.via MI Workflows, are distributed via electronic medium. The Instructions for Use is also delivered via electronic medium.
syngo.via MI Workflows includes 2 workflows (syngo.MM Oncology and syngo.MI General) as the Scenium neurology software application and the syngo MBF cardiology software application which are launched from the OpenApps framework within the MI General workflow.
Here's a breakdown of the acceptance criteria and study information for the Siemens syngo.via MI Workflows, including Scenium, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
| Feature | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| Scenium Centiloid Score Calibration (Florbetapir) | Strong agreement with standard method | R² = 0.97 |
| Scenium Centiloid Score Calibration (Florbetaben) | Strong agreement with standard method | R² = 0.98 |
| Scenium Centiloid Score Calibration (Flutemetamol) | Strong agreement with standard method | R² = 0.95 |
| Scenium Centiloid Score Validation (Amyvid™) vs. ADNI CL | Strong agreement with ADNI CL values | SceniumCL = 1.044 × ADNI CL – 0.712; R² = 0.97 |
| Scenium Centiloid Score Validation (Neuraceq™) vs. ADNI CL | Strong agreement with ADNI CL values | SceniumCL = 1.095 × ADNI CL – 7.241; R² = 0.98 |
| Scenium Centiloid Score (Amyloid PET) Agreement with Visual Reading | Excellent agreement with visual-based classification | Area Under ROC Curve = 0.9872 (optimal CL cut-off value of 26, sensitivity 92.0%, specificity 96.3%) |
2. Sample Size and Data Provenance for Test Set
- Calibration Data:
- Sample Size: Not explicitly stated, but "calibration of PET images and their corresponding SUVr and CL reference data were obtained from the GAAIN website." This implies a sufficiently large dataset for method calibration.
- Provenance: GAAIN website (Global Alzheimer's Association Interactive Network) – likely a multinational, retrospective dataset of clinical trial data.
- Validation Data (ADNI):
- Sample Size: Not explicitly stated, but "two independent datasets" were used for validation against ADNI CL values for florbetaben. ADNI (Alzheimer's Disease Neuroimaging Initiative) is a large, multi-center, prospective observational study primarily conducted in North America.
- Provenance: ADNI (Alzheimer's Disease Neuroimaging Initiative), likely primarily from the USA and Canada. Prospective given the nature of ADNI.
- Validation Data (Visual Reading Agreement):
- Sample Size: 162 patients (69 females, 93 males)
- Provenance: Retrospective review of patients with Mild Cognitive Impairment (MCI) who underwent A-PET. The specific country of origin is not mentioned.
3. Number of Experts and Qualifications for Ground Truth (Test Set)
- Calibration Data (GAAIN): The "standard method" for Centiloid scale calculation (Klunk et al.²) implies a consensus-derived or established expert-validated process. The number and specific qualifications of experts involved in the original GAAIN data curation are not detailed but are assumed to be highly qualified specialists in PET imaging and Alzheimer's research.
- Validation Data (ADNI): The ADNI Centiloid values are established through rigorous, expert-driven protocols. The text states "ADNI CL values," implying the ground truth was derived from the ADNI project's established methods, which involve numerous qualified experts in neurology, radiology, and nuclear medicine.
- Validation Data (Visual Reading Agreement): Patients were classified as "negative" by consensus. The number and specific qualifications of experts involved in this consensus are not explicitly stated, but it would typically involve experienced nuclear medicine physicians or radiologists specializing in neuroimaging.
4. Adjudication Method for the Test Set
- Calibration Data (GAAIN): Not explicitly stated, but the "standard method" for Centiloid score calculation suggests an established, perhaps algorithmic, adjudication or consensus process applied to the raw data.
- Validation Data (ADNI): Not explicitly stated, but the ADNI's established protocols for data analysis and Centiloid score determination would inherently involve robust, multi-expert consensus or adjudicated methods.
- Validation Data (Visual Reading Agreement): Patients were "classified as 'negative' by consensus." This indicates that multiple experts reviewed the images and reached an agreement on the classification. The specific method (e.g., 2+1, 3+1) is not provided.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No MRMC comparative effectiveness study was done comparing human readers with AI assistance vs. without AI assistance. The study primarily focuses on validating the device's output (Centiloid scores) against established standards and visual interpretations, not on human workflow improvement with AI.
6. Standalone (Algorithm Only) Performance
- Yes, standalone performance was done for the Scenium Centiloid scoring feature. The studies directly compare Scenium's calculated Centiloid scores (algorithm output) against "standard method" values (from GAAIN) and ADNI CL values. The agreement with visual reading also assesses the algorithm's standalone diagnostic accuracy in classifying patients.
7. Type of Ground Truth Used
- Expert Consensus / Established Methodology:
- For the calibration, the ground truth was the "standard method" of Centiloid estimation as prescribed in Klunk et al.², using reference data from GAAIN, which is an established, expert-driven consortium.
- For validation, it involved "ADNI CL values," which are considered an established ground truth in Alzheimer's research.
- For the visual reading agreement, the ground truth was "visual-based classification" determined by expert consensus.
8. Sample Size for the Training Set
- The text does not explicitly mention a "training set" for the Scenium Centiloid scoring algorithm. The process described is a "calibration" using data from GAAIN to derive transformation equations, and then "validation" on independent datasets. It's possible the calibration data acts as a form of training/development set.
- Calibration Data (GAAIN): Sample size not explicitly stated for the "calibration analysis" dataset.
9. How the Ground Truth for the Training Set was Established
- As a "training set" isn't explicitly defined, we refer to the calibration process. The ground truth for the calibration (or equations derivation) was established using "calibration of PET images and their corresponding SUVr and CL reference data obtained from the GAAIN website." This reference data itself would have been established through rigorous scientific methods and likely expert consensus within the GAAIN consortium, adhering to the "level-2 calibration analysis prescribed in Klunk et al.²" to ensure a standardized and reliable ground truth.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).