(136 days)
Sim&Size enables visualization of cerebral blood vessels for preoperational planning and sizing for neurovascular interventions and surgery. Sim&Size also allows for the ability to computationally model the placement of neurointerventional devices.
General functionalities are provided such as:
- Segmentation of neurovascular structures
- Automatic centerline detection
- Visualization of X-Ray based images for 2D review and 3D reconstruction
- Placing and sizing tools
- Reporting tools
Information provided by the software is not intended in any way to eliminate, replace or substitute for, in whole or in part, the healthcare provider's judgment and analysis of the patient's condition.
The Sim&Size software is a medical device intended to provide a 3D view of the final placement of implants. It uses an image of the patient produced by 3D rotational angiography. It offers clinicians the possibility of computationally model neurovascular implantable medical devices (IMD) in the artery or in the aneurysm to be treated through endovascular surgery, IMD such as the flow-diverters (FD) and the intrasaccular devices (ISD).
Sim&Size is a software designed with two modules. FDsize is the module that allows to preoperationally plan the choice of size of flow-diverter devices. IDsize is the module that allows to preoperationally plan the choice of size of intrasaccular devices.
Associated to these two modules, a common module is intended to import DICOM and to provide a 3D reconstruction of the vascular tree in the surgical area.
Sim&Size has been simplified as much as possible to guide the user in an intuitive way in order to reduce the total number of actions required and thus to optimize the time taken to obtain the desired results. There are currently seven steps required to choose the optimal size of an IMD to be placed:
- 1- Importing the images: the 3D rotational angiography DICOM files are imported.
- 2- Selecting of the region of interest (ROI): the user positions and focuses a sphere in the placement zone.
- 3- Threshold validation: the user checks the accuracy of the automatically extracted arterial wall. The threshold can be adjusted if needed.
- 4- Choosing the entry point: the user clicks on the entry point to the arterial network in order to retrieve the vessel centerlines.
- 5- Correct automatically or manually the centerline if needed: the user corrects the centerline going through a vessel fusion with the automatic tool or manually.
- 6- Initializing the implant: the user selects an IMD reference and the ideal placement zone.
- 7- Sizing the implant: IMD apposition is shown by a color chart in the 3D view. The user can change the IMD reference and placement zone to complete the planning for the intervention.
Patient images can be imported into Sim&Size in two ways: the user has the ability to transfer images using an external storage device (e.g. USB stick) or to retrieve images directly from Scanners Workstation if the option is enabled (only the retrieve function is possible).
The Sim&Size software is compatible with the operating systems MS Windows and Mac OS, when it is first installed a check is done to verify if the user's computer meets the minimum requirements for the use of the software. When a new version of the software is available, the update can be done by the user through a link send by Sim&Cure, the user then follow the instructions indicated in the user manual or by the use of the updater tool SacUpdates that notifies the user, then assists the download and the installation of the last version.
The computational modeling of three devices are supported by the software: Medtronic Pipeline Flex Embolization Device (PED - P100018/S015) flow diverter: Stryker Surpass Evolve Flow Diverter System (Evolve - P170024 S003); and Microvention Woven EndoBridge Aneurysm Embolism System (WEB - P170032) intrasaccular devices. The Medtronic Pipeline Flex Embolization Device, Stryker Surpass Evolve Flow Diverter and Microvention Woven EndoBridge Aneurysm Embolism System devices referenced here are FDA-approved neurointerventional devices.
The device is Sim&Size.
1. Table of Acceptance Criteria & Reported Device Performance
The document does not explicitly provide a table of acceptance criteria with numerical thresholds. Instead, it describes various performance tests and states that they were "passed." The general acceptance criterion implied is that the software performs its intended functions accurately and safely, and that the new features or changes do not negatively impact the device's performance or safety compared to the predicate device.
| Performance Metric/Test | Acceptance Criteria (Implicit) | Reported Device Performance |
|---|---|---|
| Software Verification & Validation | Device functions as intended without failure. No regression from previous version. | "All the continuous, supervised and acceptance tests are pass with the version 1.0.6." |
| DICOM Image Importation | Successful import of DICOM images. | Tests passed. |
| Patient Manager | Functionality as designed. | Tests passed. |
| Image Display & Processing | Correct display and processing of images. | Tests passed. |
| Visualization of Anatomic Reconstruction | Accurate visualization. | Tests passed. |
| Report Creation & Visualization | Correct report generation and display. | Tests passed. |
| Fusion Correction (Automatic & Manual) | Proper functionality of fusion correction. | "The new feature of manual fusion correction has been tested and the tests are passed." |
| Cybersecurity | Compliance with cybersecurity standards. | "The cybersecurity testing has been improved and all the tests are passed." |
| Non-Regression Testing | No unintended alterations due to new features. | "There is no regression between the predicate device and the version 1.0.6." |
| FDsize Module: Flow Diverter Length & Apposition Computation | Conform to mathematical definition. | Verification testing checks conformity. |
| FDsize Module: Prediction of IMD implantation | Accurate prediction compared to experimental and in-vivo data. | Validation tests ensure simulation model validity for new IMDs, proper calibration, and predictability. |
| IDsize Module: New Mechanical Solver Verification | Verification of new in-house solver. | "A variety of verification test cases were performed... proper verification of the computational model." |
| IDsize Module: Accuracy of Computational Model | Accurate prediction compared to experimental and in-vivo data. | Validation tests show accuracy, robustness, and similar overall accuracy to the old model. |
| Impact of New IMDs (FDsize) | No impact on safety and performance. | "The inclusion of the new Implantable Medical Devices databases in the Sim&Size software have no impact on of the safety and performance of the device." |
| Impact of New Mechanical Solver (IDsize) | No impact on safety and performance. | "Integration of the new mechanical solver and the model change in the Sim&Size software have no impact on of the safety and performance of the device." |
2. Sample Size for the Test Set and Data Provenance
The document does not explicitly state a numerical sample size for the test set. However, it mentions:
- FDsize module validation:
- "Experimental benchtests to perform optical acquisitions of new IMD devices samples in both unconstrained and constrained configurations." (Likely in-vitro)
- "Realistic in vitro (silicone model) datasets in which the predictability of the simulation model is assessed comparing in-vitro and virtual Flow Diverters devices implanted in silicone phantom of patients presenting with intracranial aneurysms." (In-vitro)
- "In vivo studies for which the results are based on comparisons between FD implanted in patients presenting with intracranial aneurysms and virtual FD deployment." (Clinical, retrospective/prospective not specified, but likely retrospective based on existing patient data.)
- IDsize module validation:
- "Experimental benchtests which assess the accuracy of the IDsize computational model in a well-controlled experimental configuration." (Likely in-vitro)
- "Realistic in vitro (silicone model) datasets in which a comparison is done between implanted in silicon phantoms of idealized aneurysms anatomies and virtual WEB computationally modeling with Sim&Size." (In-vitro)
- "In vivo studies for which the results are based on comparisons between WEB implanted in patients presenting with intracranial aneurysms and virtual WEB deployment." (Clinical, retrospective/prospective not specified, but likely retrospective.)
The country of origin for the data is not specified.
3. Number of Experts and Qualifications for Ground Truth
The document does not specify the number or qualifications of experts used to establish ground truth for the test set. For "in vivo studies," it implies comparison with actual patient outcomes, which would inherently involve clinical assessment by qualified medical professionals, but this is not detailed for the ground truth establishment process itself.
4. Adjudication Method
The document does not describe any specific adjudication method (e.g., 2+1, 3+1) for establishing ground truth within the context of the performance testing.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
The document does not mention an MRMC comparative effectiveness study that assesses the effect size of AI assistance on human reader improvement. The device description explicitly states: "Information provided by the software is not intended in any way to eliminate, replace or substitute for, in whole or in part, the healthcare provider's judgment and analysis of the patient's condition." This indicates it's a planning tool, not a diagnostic AI intended to assist interpretation, so an MRMC study in that sense might not be applicable or expected.
6. Standalone Performance
Yes, a standalone performance assessment was done. The "Performance Testing - Bench" section describes verification and validation tests for the FDsize and IDsize modules, which evaluate the computational modeling capabilities of the software itself against mathematical definitions, experimental setups (in-vitro), and in-vivo comparisons. This indicates the algorithm's performance without direct human-in-the-loop diagnostic interpretation. The software is a planning tool, meant to be used by a human, but the validity of its computations were tested independently.
7. Type of Ground Truth Used
The ground truth types used appear to be:
- Mathematical definitions: For verification of computed flow diverter length and apposition.
- Experimental measurements: Obtained from "experimental benchtests" and "realistic in vitro (silicone model) datasets." This involves physical measurements from devices implanted in phantoms.
- In-vivo patient outcomes/observations: For "in vivo studies," where virtual deployments are compared to actual deployments in patients. This implies clinical observations or post-procedural imaging.
8. Sample Size for the Training Set
The document does not specify a sample size for the training set. It primarily describes verification and validation of the computational model and the solver, rather than a machine learning model that would require a distinct training set. The changes in this submission relate to adding new IMD databases and a new mechanical solver, which typically involve calibrating parameters for these specific devices or equations, rather than training a deep learning model from scratch on a large image dataset.
9. How Ground Truth for the Training Set Was Established
As a specific "training set" for a machine learning model is not explicitly mentioned and the focus is on computational modeling and solver verification/validation, the concept of ground truth establishment for a training set in the typical AI sense is not detailed. The "ground truth" referenced for validation is against mathematical definitions, experimental data (in-vitro), and in-vivo comparisons, as described in point 7.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).