Search Results

UniSyn is a software application for image registration and fusion display of scanned image data from CT, PET, SPECT, MR and other medical scanners. It is to be used by qualified radiology and nuclear medicine professionals. UniSyn creates multi-planar reformat and maximum intensity projection displays of the data and provides measurements such as area, volume and Standard Uptake Values for user defined regions on the image.

For use with internally administered radioactive products. UniSyn can estimate radiation dose from internalized radioactivity in the human body as a result of a diagnostic or therapeutic medical procedure involving radioactive materials. UniSyn should not be used to deviate from approved product dosing and administration instructions. Refer to the product's prescribing information for instructions.

UniSyn Molecular Imaging (MI) is a Software as a Medical Device (SaMD) that supports the visualization, manipulation and analysis of medical image data acquired or used in radiology and nuclear medicine centers. UniSyn Ml is only intended to be used by qualified radiology and nuclear medicine professionals. Univer-interface components: a patient study browser and the UniSyn MI viewer. The software is available in both thick and can be integrated to launch from PACS software.

Using UniSyn MI users can coregister anatomical and visualize them in fused and/or standalone display, e.g. single or multi-modal combinations of PET, SPECT, CT, and MR images. Users can also visualize and process planar nuclear medicine (NM) images acquired as single of multi-frame images. The layout of the UniSyn MI viewer is highly customisable, a typical layout for a PET/CT study would include of the PET and CT series in multiplanar reformatted (MPR) views as well as a 3D maximum-intensity (MIP) projection rendering of the PET series.

UniSyn MI provides tools to zoom, pan, stack, and window-level the displayed series. Our triangulation tool can be used to localize a single anatomical point of interest among all MPR and MP views of the rest (RO) tools are available to delineate 2D and 3D regions and then compute image statistics within those regions, e.g. ROI area/volumes, minimum, mean and standard deviation of image pixel values. Various image segmentation tools are included with UniSyn MI to facilitate ROI delineation based on image pixel data.

UniSyn MI includes a tool for absorbed dose estimation associated with internally deposited with diagnostic and therapeutic medical procedures. Absorbed dose estimates are based on single- or multi-time point activity measurements of molecular images and absorbed dose coefficients (S-Values) that are based on computational human models.

Once a user has completed their review or analysis of a given study, UniSyn MI provides tools to generate reports and export exemplary image data to share with referring physicians to substantiate their findings.

The medical device described in the document is "UniSyn Molecular Imaging (6-3-1)". It is a software application for image registration, fusion display, and analysis of medical image data (CT, PET, SPECT, MR) used by radiology and nuclear medicine professionals. It also estimates radiation dose from internalized radioactivity.

Here's an analysis of the provided information regarding acceptance criteria and the study that proves the device meets them:

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

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

• Normal Organ Dosimetry: "Published data" and "published literature" were used for comparison. The specific number of patients or organs within these published data sets is not specified. Data provenance is implied to be from existing medical literature.
• Tumor Dosimetry: Comparisons were made using "tumor sizes ranging from 3.9 to 600 cc," implying a range of scenarios were tested against the OLINDA/EXM v1.0 sphere model. The specific number of tumor cases or datasets used is not specified. Data provenance is implied to be through comparison with an existing, cleared device (OLINDA/EXM v1.0).
• Radionuclides Evaluated: Fluorine-18, Gallium-177, Technetium-99m, and Yttrium-90 were evaluated. The number of samples for each radionuclide is not specified.
• The document does not specify countries of origin, nor whether the data was retrospective or prospective. It mainly focuses on comparison with existing validated models and literature.

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

The document does not specify the number of experts or their qualifications used to establish ground truth for the test set. Ground truth was established by:

• Comparison to "values from published data" for normal organ dosimetry.
• Comparison to the "sphere model of OLINDA/EXM v1.0 (K033960)" for tumor dosimetry.

4. Adjudication method for the test set

The document does not specify an adjudication method for the test set. The testing involved direct comparison to reference values from published literature or a validated software model, rather than expert adjudication of device 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

A multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly mentioned for the UniSyn Molecular Imaging device in this document. The study focused on the validation of the dosimetry model's accuracy against established references, not human-in-the-loop performance or improvement with AI assistance.

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

Yes, a standalone performance evaluation was done for the dosimetry model. The testing involved comparing the device's dose estimates directly against reference values from published data and another FDA-cleared device (OLINDA/EXM v1.0). This evaluates the algorithm's performance in isolation from human interpretation.

For functionalities shared with the predicate (image registration, visualization, measurements, etc.), "existing verification and validation testing protocols" were used, which would also likely be standalone performance evaluations of the software's capabilities.

7. The type of ground truth used

The ground truth used was:

• Published Literature/Data: For normal organ dosimetry, comparison was made to established values in published literature.
• Validated Software Model: For tumor dosimetry, comparison was made to the sphere model of an FDA-cleared device, OLINDA/EXM v1.0 (K033960), which itself serves as a recognized ground truth for dose estimation in that context.

8. The sample size for the training set

The document does not specify the sample size for any training set. Given that the testing methods involve comparisons to established models and published data, it is likely that the dosimetry model relies on pre-existing scientific understanding and computational models rather than a machine learning approach that requires a distinct "training set."

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

As no specific training set is mentioned (implying a non-machine learning approach for the core dosimetry calculations), the establishment of ground truth for a training set is not applicable in this document. The underlying principles and S-values used in the dosimetry model would be based on established scientific principles and data.

Ask a specific question about this device