Search Results

VelocityAIS (VelocityAI) is a stand-alone software product that provides the physician a means for comparison of medical imaging data from multiple DICOM conformant imaging modality sources. It allows the display, annotating, volume rendering, registration and fusing of medical images as an aid during use by diagnostic radiology, oncology, radiation therapy planning and other medical specialties. VelocityAIS (VelocityAI) is not intended for mammography diagnosis.

Version 2.0 of VelocityAIS, also marketed as VelocityAI with this version, is a stand-alone software product that provides medical image processing designed to facilitate the oncology or other clinical specialty work flow by allowing the comparison of medical imaging data from different modalities, points in time, and/or scanning protocols. The product provides users with the means to display, co-register and fuse medical images from multiple modalities including PET, SPECT, CT, and MR, draw Regions of Interest (ROI), calculate, and report relative differences in pixel intensities, Standardized Uptake Value (SUV) or other values within those regions, and import and export results to and from commercially available radiation treatment planning systems and PACS devices. Version 2.0 of the product contains the additional features of: Automated deformable registration. Anatomical atlas based segmentation tools. Operating systems compatibility with MicroSoft Windows XP Home and Professional (Service Pack 2), MicroSoft Vista Home and Professional, Linux (CentOS) and MacOSX (Leopard). VelocityAIS (VelocityAI) is used as a stand-alone application on recommended Off-The-Shelf (OTS) computers supplied by the company or by the end-user.

The provided document is a 510(k) Pre-Market Notification Summary for the VelocityAIS (also "VelocityAI") software, version 2.0. This document primarily focuses on establishing substantial equivalence to predicate devices for regulatory clearance, rather than detailing a specific clinical study with granular performance metrics and acceptance criteria for a novel device function.

Here's a breakdown of the requested information based on the provided text, and explicit statements about what is not available in this document:

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

Explanation: The document states that the software "meets its functional specifications and performance requirements" but does not provide a table of quantifiable acceptance criteria (e.g., sensitivity, specificity, accuracy, precision, processing speed) or specific numerical performance results. The core of this 510(k) is about demonstrating substantial equivalence to existing predicate devices, implying that its performance is equivalent or better without introducing new safety/effectiveness concerns.

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Sample Size for Test Set: Not mentioned in the provided text.
• Data Provenance: Not mentioned in the provided text. The document describes the software's capabilities with "medical imaging data from multiple DICOM conformant imaging modality sources" but does not specify the origin or nature of the data used for verification/validation.
• Retrospective/Prospective: Not mentioned in the provided text.

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)

• Not mentioned in the provided text. The document focuses on software functionality and comparison to predicate devices, not on the details of a clinical reader study or the establishment of ground truth by experts.

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

• Not mentioned in the provided text.

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

• Not mentioned in the provided text. This 510(k) summary does not describe an MRMC comparative effectiveness study or the impact of AI assistance on human readers. The device is described as "medical image processing designed to facilitate the oncology or other clinical specialty work flow" and providing "a means for comparison of medical imaging data," but it doesn't quantify improvements in human reader performance.

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

• The document describes the device as a "stand-alone software product" that provides image processing capabilities, implying its functionality without direct human intervention in its core processing. However, it's an "aid during use by diagnostic radiology, oncology, radiation therapy planning and other medical specialties," meaning human experts interpret the output.
• Performance of the algorithm only (e.g., specific metrics for deformable registration or segmentation accuracy): Not explicitly detailed in the provided text. The claim is general: "verified and validated that the VelocityAIS (VelocityAI) software meets its functional specifications and performance requirements."

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

• Not mentioned in the provided text.

8. The sample size for the training set

• Not mentioned in the provided text. The document does not describe the training of any AI/ML models, though it mentions "Automated deformable registration" and "Anatomical atlas based segmentation tools" which could imply the use of such models.

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

• Not mentioned in the provided text.

In summary, the provided 510(k) document is a regulatory submission focused on substantial equivalence. It confirms that the manufacturer verified and validated the software's functional specifications and performance requirements. However, it does not contain the detailed clinical study information (e.g., specific acceptance criteria, sample sizes, ground truth establishment, expert qualifications, or comparative effectiveness study results) that is typically found in a clinical validation report for a device with advanced diagnostic capabilities. This type of information is often provided in more detailed validation reports or during later clinical trials, which are not part of this initial 510(k) summary for a "Picture Archiving and Communication System (Medical Imaging Software)" with image processing features.

Ask a specific question about this device

VelocityAIS is a stand-alone software product that provides the physician a means for comparison of medical imaging data from multiple DICOM conformant imaging modality sources. It allows the display, annotating, volume rendering, registration and fusing of medical images as an aid during use by diagnostic radiology, oncology, radiation therapy planning and other medical specialties. VelocityAIS is not intended for mammography diagnosis.

VelocityAIS is a stand-alone software product that provides medical image processing designed to facilitate the oncology or other clinical specialty work flow by allowing the comparison of medical imaging data from different modalities, points in time, and/or scanning protocols. The product provides users with the means to display, co-register and fuse medical images from multiple modalities including PET, SPECT, CT, and MR, draw Regions of Interest (ROI), calculate, and report relative differences in pixel intensities, Standardized Uptake Value (SUV) or other values within those regions, and import and export results to and from commercially available radiation treatment planning systems and PACS devices. VelocityAIS is used as a stand-alone application on recommended Off-The-Shelf (OTS) computers supplied by the company or by the end-user.

Here's a breakdown of the acceptance criteria and the study information for K070248 (Velocity AIS), based on the provided text:

Important Note: The provided document is a 510(k) summary for a pre-market notification, not a detailed study report. As such, it primarily focuses on establishing substantial equivalence to predicate devices and demonstrating that the software meets its functional specifications. It does not contain extensive details about clinical performance studies with specific statistical acceptance criteria for diagnostic accuracy, reader studies, or detailed ground truth methodologies that would be typical for a device making a diagnostic claim based on AI.

Acceptance Criteria and Reported Device Performance

The document does not explicitly present specific numerical "acceptance criteria" for diagnostic performance (e.g., sensitivity, specificity, AUC) or a comparative table of "reported device performance" against such criteria.

Instead, the core of the acceptance criteria for this 510(k) is based on:

• Functional Specifications and Performance Requirements: The device must meet its own defined functional specifications and performance requirements.
• Substantial Equivalence: The device must be demonstrated as "substantially equivalent" to legally marketed predicate devices, meaning it is similar in characteristics, materials, features, technological features, intended use, and indications for use, and does not pose any new issues of safety and effectiveness.

The reported device "performance" is implicitly stated as:

Study Details (Based on available information)

Given the nature of the 510(k) summary, specific details about "studies" in the context of diagnostic accuracy or deep learning models are limited. The provided text describes primarily the verification and validation (V&V) of the software's functional specifications and its comparison to predicate devices, rather than a clinical performance study as one might expect for a new diagnostic AI.

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

  • Sample Size: Not specified for any "test set" in the context of diagnostic performance. The document only mentions "medical imaging data from multiple DICOM conformant imaging modality sources."
  • Data Provenance: Not specified. No mention of country of origin or whether the data was retrospective or prospective.

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

  • Not specified. There is no mention of experts or ground truth establishment for a diagnostic test set. The validation seems to be primarily functional.

• 4. Adjudication method for the test set:

  • Not specified. This is not applicable given the lack of detail on a diagnostic test set and ground truth establishment.

• 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, an MRMC comparative effectiveness study is not mentioned or described. The device is characterized as aiding "during use by diagnostic radiology, oncology, radiation therapy planning and other medical specialties," but no study on human reader improvement with or without AI assistance is detailed.

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

  • The document states, "VelocityAIS is a stand-alone software product." This implies independent functionality. However, it's not a "standalone performance study" in the sense of an algorithm making a diagnostic claim without human oversight. Its stated use is "as an aid" for physicians, suggesting a human-in-the-loop model for clinical decision-making. No metrics for standalone diagnostic accuracy are provided.

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

  • Not specified in the context of clinical or diagnostic performance. The ground truth refers to whether the software meets its internal "functional specifications and performance requirements," rather than a clinical truth.

• 8. The sample size for the training set:

  • Not applicable/not specified. This device predates the widespread use of deep learning in medical imaging diagnostics for which a "training set" of patient data would be a primary concern. Its function involves image display, co-registration, fusion, ROI drawing, and calculation, which are rule-based software functionalities, not statistical learning from a large training dataset.

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

  • Not applicable/not specified, for the same reasons as #8. Ground truth for training would not be relevant for this type of software as described.

Ask a specific question about this device