(8 days)
The device is a diagnostic imaging system for fixed or mobile installations that combines Positron Emission Tomography (PET) and X-ray Computed Tomography (CT) systems. The CT subsystem produces cross-sectional images of the body by computer reconstruction of x-ray transmission data. The PET subsystem produces images of the distribution of PET radiopharmaceuticals in the patient body (specific radiopharmaceuticals are used for whole body, head, heart and other organ imaging). Attenuation correction is accomplished by either CTAC or Cs-137. The device also provides for list mode, dynamic, and gated acquisitions.
Image processing and display workstations provide software applications to process, analyze, display, quantify and interpret medical images/data. The PET and CT images may be registered and displayed in a "fused" (overlaid in the same spatial orientation) format to provide combined metabolic and anatomical data at different angles. Trained professionals use the images in:
- The evaluation, detection and diagnosis of lesions, disease and organ function such as but not limited to cancer, cardiovascular disease, and neurological disorders.
- The detection, localization, and staging of tumors and diagnosing cancer patients.
- Treatment planning and interventional radiology procedures.
Cardiac imaging software provides functionality for the quantification of cardiology images and datasets including but not limited to myocardial perfusion for the display of wall motion and quantification of left-ventricular function parameters from gated myocardial perfusion studies and for the 3D alignment of coronary artery images from CT coronary angiography onto the myocardium.
Both subsystems (PET and CT) can also be operated independently as fully functional, diagnostic imaging systems including application of the CT scanner as a radiation therapy simulation scanner.
The device is a hybrid diagnostic imaging system that combines Positron Emission Tomography (PET) and X-ray Computed Tomography (CT) scanners that can be utilized in fixed installations or mobile environments. The device is comprised of the following system components/subsystems: Positron Emission Tomography (PET), X-ray Computed Tomography (CT), a patient table, gantry separation unit, and the acquisition and processing workstations. The 16-slice configuration is also referred to as Gemini 16 Power.
This document describes the Gemini PET/CT Imaging System and its clearance via a 510(k) premarket notification. The information provided heavily references existing standards and comparisons to predicate devices, rather than a standalone study with explicit acceptance criteria and performance metrics for the device itself as a new, distinct technology with novel AI components.
Based on the provided text, the device is an imaging system (hardware and associated software for display/analysis), not an AI algorithm performing a specific diagnostic task like lesion detection that would typically have quantitative acceptance criteria on metrics like sensitivity/specificity.
Therefore, the requested information, which is focused on AI algorithm performance, cannot be fully extracted from this document in the typical sense.
Here's an attempt to answer the questions based on the available information, noting where specific AI-related metrics or studies are not present:
1. A table of acceptance criteria and the reported device performance
Based on the document, the "acceptance criteria" are compliance with established regulatory and industry standards for medical imaging systems, and "performance" is demonstrated by adherence to these standards and equivalence to predicate devices. There are no specific quantitative acceptance criteria or reported device performance metrics in the sense of accuracy, sensitivity, or specificity for a diagnostic task that an AI algorithm would typically demonstrate.
| Acceptance Criteria (based on regulatory and industry standards) | Reported Device Performance |
|---|---|
| Safety: | |
| - 21 CFR 1020.30/1020.33 Performance Standards for Ionizing | "Philips Medical System adheres to recognized and established industry practice. Radiation safety is assured by compliance and certification to the performance standards for ionizing radiation emitting product, 21CFR 1020.30 and 21CFR 1020.33." Additionally, "Electrical and mechanical safety is assured by adherence and certification to the applicable standards in the IEC 60601-1 series." |
| Radiation Emitting Products, Computed Tomography Equipment | |
| - IEC 60601-1 series (Electrical and Mechanical Safety) | |
| Imaging Performance: | |
| - NEMA NU-2 standard | "The device performance was measured in accordance with the NEMA-NU2 standard." |
| Substantial Equivalence: | |
| - Similar intended use to predicate devices | "The device is substantially equivalent to the predicate devices based upon similar intended use, technological comparison, and system performance." |
| - Similar technological characteristics to predicate devices |
2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
The document describes a system performance test measured in accordance with the NEMA-NU2 standard. This standard typically involves phantom studies or specified tests on the physical imaging system itself, not studies on a specific patient data test set for AI algorithm validation. Therefore, information about patient sample size, data provenance, retrospective/prospective nature is not applicable/not provided in this document for the type of evaluation described.
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)
Since the document describes a system performance test (hardware/physics), not an AI algorithm's diagnostic performance on patient data, there is no mention of experts establishing ground truth for a test set of patient cases.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
Not applicable as there is no patient test set with ground truth established by experts.
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
There is no mention of an MRMC comparative effectiveness study or any study involving human readers with or without AI assistance. The submission is for an imaging system, not an AI diagnostic tool.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
The "device" being described is a combined PET/CT imaging system with associated software for processing, analysis, and display. There is no mention of a standalone AI algorithm (without human-in-the-loop) performance study for a specific diagnostic task. The software components are for image registration, display, quantification of cardiology images, and general image interpretation, presumably by trained professionals.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
For the system performance tests (NEMA-NU2), ground truth relates to the physical accuracy, resolution, and uniformity measurements of the imaging system, typically using phantoms with known properties. It does not involve expert consensus, pathology, or outcomes data related to patient diagnosis for an AI algorithm.
8. The sample size for the training set
The document describes system performance tests and substantial equivalence to predicate devices. It does not mention a training set for an AI algorithm. The device's software applications (e.g., for cardiac phantom quantification) are likely based on established algorithms and mathematical models, rather than deep learning requiring large training datasets as understood in current AI contexts.
9. How the ground truth for the training set was established
Not applicable, as there is no mention of a training set for an AI algorithm.
§ 892.1200 Emission computed tomography system.
(a)
Identification. An emission computed tomography system is a device intended to detect the location and distribution of gamma ray- and positron-emitting radionuclides in the body and produce cross-sectional images through computer reconstruction of the data. This generic type of device may include signal analysis and display equipment, patient and equipment supports, radionuclide anatomical markers, component parts, and accessories.(b)
Classification. Class II.