Search Results
Found 2 results
510(k) Data Aggregation
(144 days)
The Ingenuity CT is a Computed Tomography X-Ray System intended to produce images of the head and body by computer reconstruction of x-ray transmission data taken at different angles and planes. These devices may include signal analysis and display equipment, patient and equipment supports, components and accessories. The Ingenuity CT is indicated for head, whole body, cardiac and vascular X-ray Computed Tomography applications in patients of all ages.
These scanners are intended to be used for diagnostic imaging and for low dose CT lung cancer screening for the early detection of lung nodules that may represent cancer*. The screening must be performed within the established inclusion criteria of programs / protocols that have been approved and published by either a governmental body or professional medical society.
*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 Philips Ingenuity CT consists of three system configurations, the Philips Ingenuity CT, the Philips Ingenuity Core and the Philips Ingenuity Core128. These systems are Computed Tomography X-Ray Systems intended to produce cross-sectional images of the body by computer reconstruction of X-ray transmission data taken at different angles and planes. These devices may include signal analysis and display equipment, patient, and equipment supports, components and accessories. These scanners are intended to be used for diagnostic imaging and for low dose CT lung cancer screening for the early detection of Jung nodules that may represent cancer*.
The main components (detection system, the reconstruction algorithm, and the x-ray system) that are used in the Philips Ingenuity CT have the same fundamental design characteristics and are based on comparable technologies as the predicate.
The main system modules and functionalities are:
-
- Gantry. The Gantry consists of 4 main internal units:
- a. Stator a fixed mechanical frame that carries HW and SW.
- b. Rotor A rotating circular stiff frame that is mounted in and supported by the stator.
- c. X-Ray Tube (XRT) and Generator fixed to the Rotor frame.
- d. Data Measurement System (DMS) a detectors array, fixed to the Rotor frame.
-
- Patient Support (Couch) carries the patient in and out through the Gantry bore synchronized with the scan.
-
- Console A two part subsystem containing a Host computer and display that is the primary user interface and the Common Image Reconstruction System (CIRS) - a dedicated powerful image reconstruction computer.
In addition to the above components and the software operating them, each system includes a workstation hardware and software for data acquisition, display, manipulation, storage and filming as well as post-processing into views other than the original axial images. Patient supports (positioning aids) are used to position the patient.
Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided text:
Important Note: The provided document is a 510(k) submission for a CT scanner (Philips Ingenuity CT), which focuses on demonstrating substantial equivalence to a predicate device (Philips Plus CT Scanner), rather than establishing new performance claims with specific acceptance criteria and clinical trial results typical for entirely novel AI/ML devices. Therefore, much of the requested information, particularly regarding AI-specific performance (like effect size of human reader improvement with AI, standalone AI performance, ground truth for training AI models) is not directly present. The clinical evaluation described is a comparative image quality assessment rather than a diagnostic accuracy clinical trial.
1. Table of Acceptance Criteria and Reported Device Performance
Given the nature of this 510(k) submission, the "acceptance criteria" are primarily established against international and FDA-recognized consensus standards for medical electrical equipment and CT systems, and against the performance of the predicate device. The "reported device performance" refers to the successful verification against these standards and equivalence to the predicate.
| Acceptance Criteria Category | Specific Criteria / Standard Met | Reported Device Performance |
|---|---|---|
| Safety and Essential Performance (General) | IEC 60601-1:2006 (Medical electrical equipment Part 1: General requirements for basic safety and essential performance) | All verification tests were executed and passed the specified requirements. |
| Electromagnetic Compatibility (EMC) | IEC 60601-1-2:2007 (Medical electrical equipment Part 1-2: General requirements for basic safety and essential performance - Collateral Standard: Electromagnetic disturbances -Requirements and tests) | All verification tests were executed and passed the specified requirements. |
| Radiation Protection | IEC 60601-1-3 Ed 2.0:2008 (Medical electrical equipment Part 1-3: General requirements for basic safety - Collateral standard: Radiation protection in diagnostic X-ray equipment) | All verification tests were executed and passed the specified requirements, including radiation metrics. |
| Usability | IEC 60601-1-6:2010 (Medical electrical equipment -- Part 1-6: General requirements for basic safety and essential performance - Collateral standard: Usability) | All verification tests were executed and passed the specified requirements. |
| Safety of X-ray Equipment (Specific) | IEC 60601-2-44:2009 (Medical electrical equipment Part 2-44: Particular requirements for the safety of X-ray equipment) | All verification tests were executed and passed the specified requirements. |
| Software Life Cycle Processes | IEC 62304:2006 (Medical device software Software life cycle processes) | Software Documentation for a Moderate Level of Concern (per FDA guidance) was included. All verification tests were executed and passed the specified requirements. |
| Risk Management | ISO 14971 (Medical devices Application of risk management to medical devices (Ed. 2.0, 2007)) | Traceability between requirements, hazard mitigations and test protocols described. Test results per requirement and per hazard mitigation show successful mitigation. |
| Image Quality Metrics (Comparative to Predicate) | CT number accuracy and uniformity, MTF, noise reduction performance (i.e., iDose4 vs. FBP), slice thickness, slice sensitivity profiles. Diagnostic image quality for brain, chest, abdomen, pelvis/orthopedic. | Bench tests included patient support/gantry positioning repeatability and accuracy, laser alignment accuracy, CT image quality metrics testing. Sample phantom images provided. Clinical evaluation found no difference in image quality between iDose4 and FBP, with iDose4 scoring higher in most cases, maintaining diagnostic quality. |
| Functional and Non-Functional Requirements (System Level) | System Requirements Specification, Subsystem Requirement Specifications, User Interface Verification | Functional and non-functional regression tests, as well as user interface verification, provided in the Traceability Matrix (successful). |
| Clinical Validation (Workflow & Features) | Covered requirements related to clinical workflows and features. | Validation test plan executed as planned, acceptance criteria met for each requirement. All validation tests demonstrate safety and effectiveness. |
| Serviceability Validation | Covered requirements related to upgrade, installation, servicing, and troubleshooting. | Validation test plan executed as planned, acceptance criteria met for each requirement. |
| Manufacturing Validation | Covered requirements related to operations and manufacturing. | Validation test plan executed as planned, acceptance criteria met for each requirement. |
2. Sample Size Used for the Test Set and Data Provenance
The document does not specify a distinct "test set" sample size in the sense of a number of clinical cases or patient images used for a diagnostic accuracy study. Instead, it refers to:
- Bench tests: These involved phantom images and physical testing of the system (e.g., patient support/gantry positioning repeatability and accuracy, laser alignment accuracy, CT image quality metrics testing). No sample size for these is given.
- Clinical Evaluation: An "image evaluation" was performed involving "images of the brain, chest, abdomen and pelvis/peripheral orthopedic body areas." The number of images or patient cases used for this evaluation is not specified.
- Data Provenance: Not explicitly stated, but given it's a Philips product, it's likely internal development and validation data. There is no mention of external datasets or specific countries of origin. The evaluation compares FBP and iDose4 reconstructions of the same images. The clinical evaluation implicitly relates to retrospective data as it compares reconstructed images.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
- Number of Experts: "a qualified radiologist". So, one expert.
- Qualifications of Experts: Described only as "a qualified radiologist." No specific experience (e.g., years of experience, subspecialty) is provided.
4. Adjudication Method for the Test Set
The evaluation was performed by a single radiologist using a 5-point Likert scale. Therefore, no adjudication method (like 2+1, 3+1 consensus) was used as there was only one reviewer.
5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study was Done
No, an MRMC comparative effectiveness study was not done. The document describes an image evaluation by a single radiologist, not multiple readers. It also describes a comparison of image quality between reconstruction techniques (FBP vs. iDose4), not a comparison of human reader diagnostic performance with vs. without AI assistance.
- Effect size of human readers improving with AI vs without AI assistance: This information is not applicable as this type of study was not performed. The study evaluated if iDose4-reconstructed images (which is an iterative reconstruction technique for image quality improvement and dose reduction, not an AI for diagnosis) maintained diagnostic quality compared to standard FBP.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was Done
Yes, in spirit, the primary evaluation is about the algorithm's output quality. The iDose4 iterative reconstruction algorithm directly produces images without human intervention, and these images were then evaluated by a radiologist. The core of this 510(k) is about the technical performance and safety of the CT scanner and its components, including its reconstruction algorithms. The evaluation described ("image evaluation...") is a standalone assessment of the image quality produced by the iDose4 algorithm compared to standard FBP. It is not an "AI diagnostic algorithm" standalone performance, but rather an "image reconstruction algorithm" standalone performance.
7. The Type of Ground Truth Used
For the clinical image evaluation, the "ground truth" was established by the evaluation of a qualified radiologist using a 5-point Likert scale to determine if images were of "diagnostic quality" and for comparing image quality between reconstruction methods. This could be considered a form of "expert consensus," albeit from a single expert in this case. There is no mention of pathology or outcomes data being used as ground truth for this specific image quality assessment.
8. The Sample Size for the Training Set
Not applicable in the context of this 510(k) as presented.
The device (Philips Ingenuity CT) is a hardware CT scanner with associated software, including image reconstruction algorithms (like iDose4). While iterative reconstruction algorithms might involve some form of "training" or optimization during their development, the document does not speak to a "training set" in the sense of a dataset used to train a machine learning model for a specific diagnostic task that would typically be described in an AI/ML device submission. The description focuses on technical modifications and adherence to engineering and safety standards, and performance against a predicate device.
9. How the Ground Truth for the Training Set was Established
Not applicable for the reasons stated above (no "training set" for an AI/ML diagnostic model described).
Ask a specific question about this device
(133 days)
The Philips Multislice CT Systems are Computed Tomography X-Ray Systems intended to produce cross-sectional images of the body by computer reconstruction of X-ray transmission data taken at different angles and planes. These devices may include signal analysis and display equipment supports, components and accessories. The scanners are intended to be used for diagnostic imaging and for low dose CT lung cancer screening for the early detection of lung nodules that may represent cancer*.
The screening must be performed within the established inclusion criteria of protocols that have been approved and published by either a governmental body or professional medical society.
*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.
Philips Low Dose CT Lung Cancer Screening option can be used with Philips whole body multi-slice CT X-Ray Systems installed in a healthcare facility (clinic / hospital). These systems provide a continuously rotating X-ray tube and detector array with multi-slice capability. The acquired x-ray transmission data is reconstructed by computer into cross-sectional images of the body taken at different angles and planes. Reconstruction algorithms available are standard reconstruction (filtered back projection), iDose 4 and IMR iterative reconstruction. These systems also include signal analysis and display equipment, patient and equipment supports, components and accessories.
There are no functional, performance, feature, or design changes required for the qualified CT systems onto which the LDCT LCS Option is applied. Because none of the CTs will require hardware or software modifications, the Philips Low Dose CT Lung Cancer Screening option and the currently marketed and predicate Philips Multislice CT System for qualified CT systems in the installed base consists of:
- A set of up to three reference LDCT LCS protocols: standard reconstruction, standard reconstruction with iDose 4, and with IMR iterative reconstruction (where applicable), for each qualified CT System on a per CT platform basis;
- Detailed instructions on how to create the protocols on the corresponding CT System; and
- A dedicated Instructions for Use for LDCT LCS that covers all qualified systems.
This document is a 510(k) premarket notification for the Philips Multislice CT System with Low Dose CT Lung Cancer Screening. It primarily focuses on demonstrating substantial equivalence to existing predicate devices, rather than presenting a standalone study proving the device meats specific acceptance criteria in a clinical setting with human readers.
However, based on the provided text, I can infer the acceptance criteria relate to maintaining image quality parameters for Low Dose CT Lung Cancer Screening (LDCT LCS) that are equivalent to or better than standard CT performance, especially given the reduced radiation dose. The "study" proving this largely relies on non-clinical bench testing and references to external clinical literature and trials.
Here's a breakdown of the information you requested:
1. Table of Acceptance Criteria and Reported Device Performance
The document doesn't explicitly state "acceptance criteria" in a quantitative, measurable sense for a clinical study with patients. Instead, it focuses on demonstrating that the LDCT LCS option does not degrade image quality compared to existing CT systems and that existing clinical evidence supports the efficacy of LDCT LCS itself.
Here's a table based on the image quality parameters evaluated in non-clinical testing and the reported outcome of that testing:
| Acceptance Criteria (Inferred from Image Quality Parameters Evaluated) | Reported Device Performance (Non-Clinical Bench Testing Outcomes) |
|---|---|
| Spatial Resolution (MTF): Ability to visualize fine anatomical details, preserved at lower dose. | "MTF is a measure of the high contrast spatial resolution performance of the system. Nodules in the lung are high contrast objects and therefore, MTF should be preserved at lower dose conditions." "Demontrates that the image quality metrics including MTF... are substantially equivalent among different family of scanners (Brilliance 16, Brilliance Big Bore, Brilliance 64/Ingenuity, Brilliance iCT and IQon)." |
| Contrast Resolution (CNR): Ability to differentiate tissues with subtle differences in attenuation, sufficient for nodule detection. | "Sufficient Contrast-to-Noise is needed to detect solid and non-solid nodules in the lung. This parameter accounts for the contrast between an object and the background. This could also be a parameter that could influence nodule detectability." "The CNR scans were completed using the LDCT LCS scan protocols for all scanners in the comparison." "The results of non-clinical bench testing demonstrate that the image quality metrics including... Contrast to Noise Ratio are substantially equivalent among different family of scanners." "The contrast of the lung nodules it high relative to this increased noise, demonstrated by the CNR results (section 18) and NLST study." |
| Image Noise (Standard Deviation): Acceptable background noise levels at reduced dose, not compromising nodule detectability/sizing. | "As dose is reduced, background noise in the image increases. If this noise becomes too large, nodule detectability and sizing measurement may be compromised." "The results of non-clinical bench testing demonstrate that the image quality metrics including... Image noise... are substantially equivalent among different family of scanners." "Noise goes up by the square root of the mAs. The contrast of the lung nodules it high relative to this increased noise, demonstrated by the CNR results (section 18) and NLST study." |
| Noise Power Spectrum (NPS): Acceptable noise texture, not influencing nodule detection capabilities. | "Similar to the noise, changes in texture of the noise may have an influence on the nodule detection capabilities." "The NPS scans were completed using the LDCT LCS scan protocol." "The results of non-clinical bench testing also demonstrate the image quality parameters for iDose4 and IMR reconstructions are equivalent to, or better than standard FBP reconstruction." |
| Slice Thickness: Accurate slice thickness for clear edges and boundaries of nodules. | "The ability to produce slice thicknesses (FWHM of the slice sensitivity profile) that are close to the nominal slice thickness is important in defining clear edges and boundaries of the nodule." "The results of non-clinical bench testing demonstrate that the image quality metrics including... Slice Thickness... are substantially equivalent among different family of scanners." |
| CT Number Uniformity: Sufficient uniformity in the lung for robust nodule detectability. | "In a low dose scanning protocols such as with lung cancer screening, maintaining sufficient CT number uniformity throughout the lung and its various structures is important for more robust detectability of the nodules." "The results of non-clinical bench testing demonstrate that the image quality metrics including... CT number linearity, CT number accuracy... are substantially equivalent among different family of scanners." |
| CT Number Linearity: Measured CT number in a nodule not significantly affected by low dose scanning. | "In a low dose scanning protocols such as with lung cancer screening, the CT number measured in a nodule may be affected and therefore measuring CT number linearity is important." "The results of non-clinical bench testing demonstrate that the image quality metrics including... CT number linearity... are substantially equivalent among different family of scanners." |
| Image Artifacts: No new or increased artifacts obscuring anatomical details or mimicking pathology. | (Implicit in overall image quality assessment, not explicitly detailed as a separate quantified metric but mentioned as an important image quality parameter). "Image artifacts can obscure anatomical details and mimic pathology." |
| Geometric Distortion: Accuracy of measurements and image correlation with other modalities. | (Implicit in overall image quality assessment, not explicitly detailed as a separate quantified metric but mentioned as an important image quality parameter). "Geometric distortion can affect the accuracy of measurements and the ability to correlate images with other modalities." |
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Test Set Sample Size: Not applicable in the context of a clinical patient test set for this specific submission. The "test set" for demonstrating substantial equivalence related to image quality was composed of various CT scanner models and reconstruction algorithms.
- Data Provenance: The primary data used to support the efficacy of LDCT LCS itself comes from externally referenced clinical literature, specifically the National Lung Screening Trial (NLST) (N Engl J Med 2011; 365:395-409) and the International Early Lung Cancer Action Program (I-ELCAP), along with "subsequent literature." These were large-scale prospective clinical trials, likely conducted across multiple centers, including within the US (NLST).
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Number of Experts: Not applicable for this specific submission's non-clinical testing. The non-clinical image quality phantom measurements do not involve expert interpretation for ground truth.
- Qualifications of Experts: For the external clinical trials (NLST, I-ELCAP) referenced, radiologists and other medical professionals were involved in establishing diagnoses and outcomes, but their specific numbers and qualifications are not detailed in this 510(k) document.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
- Adjudication Method: Not applicable for this specific submission's non-clinical testing. No human adjudication was performed for the image quality metrics tested.
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 Comparative Effectiveness Study: No, a multi-reader, multi-case comparative effectiveness study with human readers (with vs. without AI assistance) was not conducted or presented in this 510(k) submission. This submission is for the CT system itself with a low-dose protocol option, not for an AI-powered CAD (Computer-Aided Detection) or CADx (Computer-Aided Diagnosis) device. The device does not incorporate AI for interpretation.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Standalone Performance Study: No, a standalone algorithm-only performance study was not conducted or presented in this 510(k) submission. The device is a CT scanner, not an independent algorithm for diagnosis or detection.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- Type of Ground Truth:
- For the image quality bench testing: Ground truth was established by the physical phantoms used for measurements (e.g., wire phantom for MTF, ramp/disk phantom for slice thickness, water/density phantoms for CT numbers).
- For the clinical efficacy of LDCT LCS (referenced externally): The ground truth for the NLST and I-ELCAP studies would have included pathology reports (for confirmed cancers), clinical follow-up/outcomes data (for stable or resolving nodules), and potentially expert consensus reviews of indeterminate findings.
8. The sample size for the training set
- Training Set Sample Size: Not applicable. This submission describes a CT scanner with a new protocol, not a machine learning algorithm that requires a training set. The reconstruction algorithms (iDose4, IMR) themselves would have been developed using various data, but details are not provided here.
9. How the ground truth for the training set was established
- Ground Truth for Training Set: Not applicable, as this is not an AI/ML device requiring a training set with ground truth established through labeled data in the context of this submission.
Ask a specific question about this device
Page 1 of 1