(99 days)
CT:V software is a non-invasive image processing technology that measures volume changes from paired inspiration-expiration CTs to quantify and visualize regional and global ventilation. These regional measures are derived entirely from the lung tissue displacement and lung volume change between the paired inspiration-expiration chest CTs.
CT:V is for use in adult patients. Quantification and visualizations are provided in the form of a report.
CT:V may be used when physicians need a better understanding of a patient's lung function and/or respiratory condition.
CT:V Software (i.e. the "Device"), also known as "CT:V", is a software-based image processing technology that analyzes two non-contrast Computed Tomography (CT) images of the lungs to quantify the reqional ventilation of pulmonary tissue to support clinicians in their assessment of patient lung conditions and diseases.
CT:V is provided on a 'Software as a Service' basis. The paired CT series, in DICOM format, are transferred electronically to 4DMedical. The images are input into the Software Device that operates in a cloud environment. The Device is fully automated and assesses the CT series and identifies and segments the lungs. CT:V measures ventilation captured in the CTs by directly measuring the motion field of the lung tissue at thousands of points using a form of three-dimensional motion tracking. This motion tracking quantifies the expansion of lung tissue, and hence airflow (i.e. ventilation) is determined. These regional lung motion and ventilation measurements are used to provide quantitative outputs and color maps showing ventilation within a segmented region. An Analysis Report is generated for each completed workflow which is returned to the requesting physician. The ventilation data presented in the CT:V Report provides the user with additional information that may assist in characterizing the patient.
The provided document, an FDA 510(k) summary, describes the CT Lung Ventilation Analysis Software (CT:V) developed by 4DMedical Limited. It details the device's indications for use, its comparison to a predicate device, and the performance testing conducted to demonstrate its safety and effectiveness.
Here's an analysis 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 does not explicitly present a table of "acceptance criteria" with specific numerical thresholds for accuracy, sensitivity, specificity, etc., as one might typically see for a diagnostic device. Instead, the performance testing focuses on demonstrating the device's ability to consistently measure and visualize regional and global ventilation, and its correlation and consistency with established "gold standard" or "best practice" measures (PFT, SPECT, PET).
The implicit acceptance criteria are framed around the demonstration of correlation, consistency, and substantial equivalence to existing methods for assessing lung function and ventilation.
| Implicit Acceptance Criteria (Demonstrated via) | Reported Device Performance (Summary of Findings) |
|---|---|
| Correlation with PFTs (Quantitative analysis of CT:V metrics vs. gold standard PFTs) | - CT:V Inspiratory Volume, Expiratory Volume, and Volume Change correlated with Total Lung Capacity (TLC), Functional Residual Capacity (FRC), and Vital Capacity (VC) respectively. |
- Functional metrics (VH, VHSS, VHLS, VDP) correlated with FEV1 (% predicted) and FEV1/FVC. |
| Consistency with SPECT ventilation images (Qualitative and quantitative comparison of CT:V visualizations and outputs vs. SPECT) | - CT:V visualizations were consistent overall with SPECT ventilation images, providing additional clarity for regional distribution. - "Substantial equivalence between CT:V and SPECT in the assessment of regional distribution of ventilation." |
| Consistency with PET ventilation fields (Quantitative comparison of lobar and voxel-level ventilation between CT:V and PET) | - No systematic differences in lobar ventilation between CT:V and PET. - Absence of mean difference, systematic bias, or heteroscedasticity, indicating similar detection capabilities at high/low ranges of lung ventilation.
- Strong association between CT:V and Nuclear Medicine Imaging (PET) spatial ventilation data via voxel-wise Spearman correlations. |
| Robustness across diverse inputs (Verification testing across varying pixel/slice spacings, SNR, and clinical data) | - Device was robust within acceptable performance limits across varying pixel/slice spacing (min. 2.5mm x 2.5mm pixels, 2.5mm slice spacing) and SNR. - Tested with data from various models, manufacturers, institutions, and diverse patients. |
| Safety and Effectiveness Profile Similar to Predicate (Overall conclusion based on all studies) | - CT:V Software was found to have a safety and effectiveness profile similar to the predicate device, but without the need for contrast agents. |
| Software Conformance to User Needs (Analytical Validation) | - Workflow testing demonstrated system requirements and features were implemented, reviewed, and met user needs. |
2. Sample Size Used for the Test Set and Data Provenance
-
Clinical Study 1 ( Observational Comparison Study):
- Sample Size: 32 participants (19 male, 13 female).
- Data Provenance: Data acquired in the USA.
- Nature of Data: Clinical data, including healthy participants and those with previously diagnosed lung diseases (15 patients with COPD, remainder healthy). Participants presented with symptoms like shortness of breath, coughing, phlegm, chest tightness. Diverse racial demographic groups.
- Retrospective or Prospective: Not explicitly stated as prospective, but the description "observational comparison study using data acquired in the USA" suggests it could be either pre-existing data or newly collected for the study. However, the use of "participants were included" leans towards a prospective or newly assembled cohort for the study.
-
Clinical Study 2:
- Sample Size: 17 lung cancer patients.
- Data Provenance: Publicly available dataset, collected from a single institution in Australia.
- Nature of Data: Clinical data of lung cancer patients undergoing radiotherapy, with varying lung function.
- Retrospective or Prospective: Described as "publicly available dataset," implying it was retrospective analysis of pre-existing data.
-
Non-Clinical Tests (Benchtop Verification and Validation):
- Sample Size: Not specified in terms of "cases," but refers to "a range of input parameters covering a spectrum of patient anatomies and breathing physiologies" via synthetically generated phantom image data.
- Data Provenance: Synthetically generated data.
- Nature of Data: Synthetic CT image pairs simulating human breath-holds, with known simulated lung physiologies and ventilation volumes ("ground truth").
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
The document does not specify the number or qualifications of experts used to establish the ground truth for the test set (clinical studies).
- For Clinical Study 1, the comparison was made against Pulmonary Function Testing (PFT) and Nuclear Medicine Imaging (SPECT) outcomes. PFTs are objective measurements, and SPECT interpretation would typically be done by nuclear medicine physicians or radiologists, but the document doesn't detail the process or number of readers.
- For Clinical Study 2, the comparison was made against PET (positron emission tomography) outputs. PET interpretations would similarly involve trained specialists, but no specifics are given.
- For the non-clinical tests, the "ground truth" was derived from known simulated lung physiologies and ventilation volumes in synthetically generated phantom images. This ground truth was inherent to the synthetic data generation, not established by human experts.
4. Adjudication Method for the Test Set
The document does not describe any specific adjudication method (e.g., 2+1, 3+1 consensus) for establishing ground truth or for interpreting the comparison modalities (PFT, SPECT, PET) within the clinical studies. The results are presented as quantitative correlations and qualitative consistencies with these established methods.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done
No, the document does not describe a Multi-Reader Multi-Case (MRMC) comparative effectiveness study where human readers' performance with and without AI assistance was evaluated. The studies focused on comparing the AI device's outputs directly with established gold standards (PFT, SPECT, PET) and demonstrating its consistency and correlation.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Study was Done
Yes, the performance testing described for CT:V appears to be primarily standalone (algorithm only). The device is described as "fully automated" and assesses the CT series to generate quantitative outputs and color maps, with an "Analysis Report" returned to the requesting physician. The clinical studies directly compare these automated outputs to PFT, SPECT, and PET, demonstrating the algorithm's performance independent of human interpretation improvements due to AI assistance.
7. The Type of Ground Truth Used
- Clinical Study 1:
- Quantitative aspects: Pulmonary Function Testing (PFT) metrics (TLC, FRC, VC, FEV1, FEV1/FVC) were used as ground truth for correlation. PFTs are objective physiological measurements.
- Qualitative aspects: SPECT ventilation images were used as a comparative "gold standard" for assessing regional distribution of ventilation. SPECT is a nuclear medicine imaging modality.
- Clinical Study 2:
- Quantitative aspects: PET (positron emission tomography) ventilation fields were used as a comparative "gold standard" for assessing spatial ventilation distribution at lobar and voxel levels. PET is a nuclear medicine imaging modality.
- Non-Clinical Tests:
- Simulated ground truth: "Known simulated lung physiologies and ventilation volumes" from synthetically generated phantom images. This is a controlled, artificial ground truth.
8. The Sample Size for the Training Set
The document does not specify the sample size for the training set. It only describes the test sets used for performance validation. This is common in 510(k) summaries, which focus on the validation data rather than the development details.
9. How the Ground Truth for the Training Set was Established
Since the document does not describe the training set size or specific details, it also does not explain how the ground truth for the training set was established.
§ 892.1750 Computed tomography x-ray system.
(a)
Identification. A computed tomography x-ray system is a diagnostic x-ray system intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data from the same axial plane taken at different angles. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.