Search Results

ClariPulmo is a non-invasive image analysis software for use with CT images which is intended to support the quantification of lung CT images. The software is designed to support the physician in the diagnosis and documentation of pulmonary tissue images (e.g., abnormalities) from the CT thoracic datasets. (The software is not intended for the diagnosis of pneumonia or COVID-19). The software provides automated segmentation of the lungs and quantification of low-attenuation and high-attenuation areas within the segmented lungs by using predefined Hounsfield unit thresholds. The software displays by color the segmented lungs and analysis results. ClariPulmo provides optional denoising and kernel normalization functions for improved quantification of lung CT images in cases when CT images were taken at low-dose conditions or with sharp reconstruction kernels.

ClariPulmo is a standalone software application analyzing lung CT images that can be used to support the physician in the quantification of lung CT image when examining pulmonary tissues. ClariPulmo provides two main and optional functions: LAA Analysis provides quantitative measurement of pulmonary tissue image with low attenuation areas (LAA). LAA are measured by counting those voxels with low attenuation values under the user-predefined thresholds within the segmented lungs. This feature supports the physician in quantifying lung tissue image with low attenuation area. HAA Analysis provides quantitative measurement of pulmonary tissue image with high attenuation areas (HAA). HAA are measured by counting those voxels with high attenuation values using the user-predefined thresholds within the segmented lungs. This feature supports the physician in quantifying lung tissue image with high attenuation area. Lungs are automatically segmented using a pre-trained deep learning model. The optional Kernel Normalization function provides an image-to-image translation from a sharp kernel image to a smooth kernel image for improved quantification of lung CT images. The Kernel Normalization algorithm was constructed based on the U-Net architecture. The optional Denoising function provides an image-to-image translation from a noisy low-dose image to a noise-reduced enhanced quality image of LDCT for improved quantification of lung LDCT images. The Denoising algorithm was constructed based on the U-Net architecture. The ClariPulmo software provides summary reports for measurement results that contains color overlay images for the lungs tissues as well as table and charts displaying analysis results.

The provided text is specifically from a 510(k) summary for the ClariPulmo device. It details non-clinical performance testing rather than a large-scale clinical study with human-in-the-loop performance. Therefore, some of the requested information (e.g., MRMC study, effect size of human reader improvement) may not be present in this document because it was not a requirement for this specific type of submission.

Here's the breakdown of the information based on the provided text:

Acceptance Criteria and Device Performance Study for ClariPulmo

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by the "excellent agreement" reported for the different functions, measured by Pearson Correlation Coefficient (PCC) and Dice Coefficient.

2. Sample Sizes Used for the Test Set and Data Provenance

• HAA Analysis Test Set: The specific number of cases for the test set is not explicitly stated. It mentions "patients with pneumonia and COVID-19."
• LAA Analysis Test Set: The specific number of cases for the test set is not explicitly stated. It mentions "both health and diseased patients."
• AI-based Lung Segmentation Test Set: The specific number of cases is not explicitly stated. It used "one internal and two external datasets."
• Data Provenance: The document does not specify the country of origin for the data. The studies were retrospective as they involved testing against existing datasets with established ground truth.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of Experts

• The document states "expert-established segmentations" and "expert radiologist's imageJ based segmentation."
• The number of experts is not explicitly stated, nor are their specific qualifications (e.g., years of experience), beyond being referred to as "expert."

4. Adjudication Method for the Test Set

• The document does not specify an explicit adjudication method (e.g., 2+1, 3+1). It implies a single "expert-established" ground truth was used for comparison, suggesting either a single expert or a pre-adjudicated consensus.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and its effect size

• No, an MRMC comparative effectiveness study was not done or reported in this document. The performance testing described is focused on the standalone agreement of the AI with expert-established ground truth, not how human readers improve with AI assistance. The document explicitly states: "ClariPulmo does not require clinical studies to demonstrate substantial equivalence to the predicate device."

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

• Yes, standalone performance testing was done. The entire "Performance Testing" section describes the algorithm's performance against expert-established ground truth ("Al-based lung segmentation demonstrated excellent agreements with that by expert radiologist's imageJ based segmentation").

7. The Type of Ground Truth Used

• The ground truth used was expert consensus / expert-established segmentations. Specifically, for lung segmentation, it explicitly mentions "expert radiologist's imageJ based segmentation." For HAA and LAA, it refers to "expert-established segmentations."

8. The Sample Size for the Training Set

• The document does not specify the sample size for the training set. It only mentions that the lung segmentation used a "pre-trained deep learning model" and the Kernel Normalization and Denoising algorithms were "constructed based on the U-Net architecture."

9. How the Ground Truth for the Training Set Was Established

• The document does not explicitly describe how the ground truth for the training set was established. It only refers to the test set ground truth as "expert-established segmentations." It is implied that the training data would also have been expertly annotated, given the nature of deep learning model training.

Ask a specific question about this device

AVIEW provides CT values for pulmonary tissue from CT thoracic and cardiac datasets. This software could be used to support the physician quantitatively in the diagnosis, follow up evaluation of CT lung tissue images by providing image segmentation of sub-structures in lung, lobe, airways and cardiac, registration and expiration which could analyze quantitative information such as air trapped index, and inspiration/ expiration ratio. And also, volumetric and structure analysis, density evaluation and reporting tools. AVIEW is also used to store, transfer, inquire and display CT data set on premise and as cloud environment as well to allow users to connect by various environment such as mobile devices and chrome browser. Characterizing nodules in the lung in a single study, or over the time course of several thoracic studies. Characterizations include nodule type, location of the nodule and measurements such as size (major axis), estimated effective diameter from the volume of the nodule, volume of the nodule, Mean HU(the average value of the CT pixel inside the nodule in HU), Minimum HU, Max HU, mass(mass calculated from the CT pixel value), and volumetric measures(Solid major; length of the longest diameter measured in 3D for solid portion of the nodule, Solid 2nd Major: The longest diameter of the solid part, measured in sections perpendicular to the Major axis of the solid portion of the nodule), VDT (Volume doubling time), and Lung-RADS (classification proposed to aid with findings). The system automatically performs the measurement, allowing lung nodules and measurements to be displayed and, integrate with FDA certified Mevis CAD (Computer aided detection) (K043617). It also provides CAC analysis by segmentation of four main artery (right coronary artery, left main coronary, left anterior descending and left circumflex artery then extracts calcium on coronary artery to provide Agatston score, volume score and mass score by whole and each segmented artery type. Based on the score, provides CAC risk based on age and gender.

The AVIEW is a software product which can be installed on a PC. It shows images taken with the interface from various storage devices using DICOM 3.0 which is the digital image and communication standard in medicine. It also offers functions such as reading, manipulation, analyzing, post-processing, saving and sending images by using the software tools. And is intended for use as diagnostic patient imaging which is intended for the review and analysis of CT scanning. Provides following features as semi-automatic nodule management, maximal plane measure, 3D measures and columetric measures, automatic nodule detection by integration with 3rd party CAD. Also provides Brocks model which calculated the malignancy score based on numerical or Boolean inputs. Follow up support with automated nodule matching and automatically categorize Lung-RADS score which is a quality assurance tool designed to standardize lung cancer screening CT reporting and management recommendations that is based on type, size, size change and other findings that is reported. It also automatically analyzes coronary artery calcification which support user to detect cardiovascular disease in early stage and reduce the burden of medical.

The provided FDA 510(k) summary for the AVIEW 2.0 device (K200714) primarily focuses on establishing substantial equivalence to a predicate device (AVIEW K171199, among others) rather than presenting a detailed clinical study demonstrating its performance against specific acceptance criteria.

However, based on the nonclinical performance testing section and the overall description, we can infer some aspects and present the available information regarding the device's capabilities and how it was tested. It is important to note that explicit acceptance criteria and detailed clinical study results are not fully elaborated in the provided text. The document states: "This Medical device is not new; therefore, a clinical study was not considered necessary prior to release. Additionally, there was no clinical testing required to support the medical device as the indications for use is equivalent to the predicate device."

Here's a breakdown of the requested information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Note: The document does not explicitly state "acceptance criteria" with numerical or performance targets. Instead, it describes general validation methods and "performance tests" that were conducted to ensure functionality and reliability. The "Reported Device Performance" here refers to the successful completion or validation of these functions.

2. Sample Size Used for the Test Set and Data Provenance

The 510(k) summary does not specify the sample size used for the test set(s) used in the performance evaluation, nor does it detail the data provenance (e.g., country of origin, retrospective or prospective). It simply mentions "software verification and validation" and "nonclinical performance testing."

3. Number of Experts Used to Establish Ground Truth and Qualifications

The document does not provide information on the number of experts used to establish ground truth or their specific qualifications for any of the nonclinical or performance tests mentioned. Given that no clinical study was performed, it is unlikely that medical experts were involved in establishing ground truth for a test set in the conventional sense for clinical performance.

4. Adjudication Method

No information is provided regarding an adjudication method. Since the document states no clinical study was conducted, adjudication by multiple experts would not have been applicable for a clinical performance evaluation.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not reported. The document explicitly states: "This Medical device is not new; therefore, a clinical study was not considered necessary prior to release. Additionally, there was no clinical testing required to support the medical device as the indications for use is equivalent to the predicate device." Therefore, there is no mention of an effect size for human readers with or without AI assistance.

6. Standalone Performance Study

Yes, a standalone (algorithm only without human-in-the-loop) performance evaluation was conducted, implied by the "Nonclinical Performance Testing" and "Software Verification and Validation" sections. The "Performance Test" section specifically lists several automatic and semi-automatic functions (e.g., "Auto Lung & Lobe Segmentation," "Airway Segmentation," "CAC Performance") that were tested for the device's inherent capability.

7. Type of Ground Truth Used

The document does not explicitly state the type of ground truth used for each specific performance test. For software components involving segmentation, it is common to use expert-annotated images (manual segmentation by experts) as ground truth for a quantitative comparison. For calculations like Agatston score, or VDT, the ground truth would likely be mathematical computations based on established formulas or reference standards applied to the segmented regions. However, this is inferred, not explicitly stated.

8. Sample Size for the Training Set

The document does not specify the sample size for any training set. It mentions the use of "deep-learning algorithms" for segmentation, which implies a training phase, but details about the training data are absent.

9. How Ground Truth for the Training Set Was Established

The document does not specify how the ground truth for any training set was established. While deep learning is mentioned for certain segmentation tasks, the methodology for creating the labeled training data is not detailed.

Ask a specific question about this device

The Thirona LungQ software provides CT values for pulmonary tissue which is essential for providing quantitative support for diagnosis and follow up examination. The Lung Q software can be used to support physician in the diagnosis and documentation of pulmonary tissues images (e.g., abnormalities) from CT thoracic datasets. Three-D segmentation and isolation of sub-compartments, volumetric analysis, density evaluation, fissure evaluation, and reporting tools are provided.

The LungQ software is designed to aid in the interpretation of Computed Tomography (CT) scans of the thorax that may contain pulmonary abnormalities. LungQ is standalone command-line software which must be run from a command-line interpreter and does not have a graphical user interface.

The provided text describes two non-clinical performance studies conducted for the LungQ software: an "Equivalence study" and "Additional performance testing of Thirona LungQ fissure analysis."

Here's a breakdown of the acceptance criteria and study details for the Equivalence Study, as it is the primary study proving the device meets general acceptance criteria by demonstrating equivalence to a predicate device:

1. A table of acceptance criteria and the reported device performance

Note: The document states "The results showed that outputs from Thirona LungQ 1.1.0 are equivalent to the predicate device, VIDA PW2," implying that all defined acceptance criteria were met. Specific numerical results for each metric proving equivalence are not provided beyond this general statement within the text.

2. Sample size used for the test set and the data provenance

• Sample Size: 250 CT scans.
• Data Provenance:
  • Country of Origin: Not explicitly stated, but the data is from the COPDGene study (http://www.copdgene.org), which is a U.S. based study.
  • Retrospective or Prospective: Retrospective, as scans were "randomly selected from the entire COPDGene cohort."

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

The "ground truth" for this equivalence study was established by the predicate device, VIDA PW2. Human experts were not used to establish a separate ground truth for comparison in this study. The study aims to show that LungQ's outputs are equivalent to those of an already FDA-cleared device.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Not applicable, as ground truth was established by the predicate device, not through human consensus or adjudication. The comparison was direct between the two software outputs.

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, a MRMC comparative effectiveness study involving human readers and AI assistance was not conducted for this equivalence study. This study focused on software-to-software equivalence.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, an algorithm-only standalone performance study was done for both the LungQ software and the predicate device (VIDA PW2). The outputs of LungQ were directly compared to the outputs of VIDA PW2 to determine equivalence.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

The ground truth for this equivalence study was the output of a legally marketed predicate device (VIDA PW2).

8. The sample size for the training set

The document does not explicitly state the sample size used for the training set of the LungQ software. The "Equivalence Study" describes the test set used for validation.

9. How the ground truth for the training set was established

The document does not provide details on how the ground truth for the training set was established. The focus of this submission is on the non-clinical validation of the device against a predicate, not on the specifics of its development or training data.

Ask a specific question about this device