Search Results

The MirrorMe3D Modeling System is intended for use as an image processing system for the transfer of 3D medical images. The MirrorMe3D Modeling System is also intended as a visualization system for measuring and treatment planning for aesthetic facial soft tissue. The input data is processed by the System using off-the-shelf modeling software and the result is an output data file that may then be provided as a digital model or used as input for the additive manufacturing of a physical anatomic model, which is not for diagnostic use. The MirrorMe3D Modeling System should only be used in conjunction with expert clinical judgment and is not intended for diagnostic use. MirrorMe3D trained personnel will use off-the-shelf software to assist users in creating the 3D virtual (or digital) model that depicts the surgeon's intended outcome. The anatomic models are not for diagnostic use.

Device Description

The MirrorMe3D Modeling System is image processing software that enables the input and visualization of 3D medical imaging with output files that can be virtual or physical 3D anatomic models. The Modeling System software is used for visualization of preoperative treatment planning options, with measurement functionality, for surgery of the aesthetic facial soft tissue.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study information you requested, based on the provided document:

Acceptance Criteria and Device Performance

The document describes that for each device produced by MirrorMe3D, there are checks to ensure the system conforms to specifications and is fit for its intended use. There isn't a single, aggregated table of numerical acceptance criteria for the entire device as one might find for a standalone AI algorithm. Instead, the "acceptance criteria" are tied to individual product validation and process auditing.

However, based on the text, we can infer some key performance aspects being tested.

1. Table of Acceptance Criteria (Inferred) and Reported Device Performance

Acceptance Criteria Category	Specific Acceptance Criteria (Inferred)	Reported Device Performance/Verification Method
Input Data Integrity	Input imaging data is free from corruption and suitable for processing.	"MirrorMe3D checks the integrity of the input imaging data."
Treatment Option Visualization Accuracy	The visualization of patient-specific treatment options accurately reflects the intended plan and is acceptable to the doctor.	"validates the visualization of the patient specific treatment options through doctor and staff review, conducts testing and a verification of the model design files." Additionally, "The approval of the design of the anatomic model depicting the intended treatment outcome by the Doctor is required and indicates design acceptance."
Model Design File Verification	The digital model design files adhere to specifications and accurately represent the intended outcome.	"conducts testing and a verification of the model design files."
Physical Product Quality	Physical 3D printed models are free from visual defects and meet established quality protocols.	"visually inspections all physical products using a quality protocol."
Geometric Accuracy of Additive Manufacturing	Additively manufactured outputs maintain geometric accuracy within an established tolerance range.	"The model production process is tested on a monthly basis to confirm the additively manufactured outputs meet conformance standards and maintain geometric accuracy within an established tolerance range."
Off-The-Shelf (OTS) Software Performance	The off-the-shelf software programs maintain acceptable tolerances and reasonable measurement parameters for modeling.	"Software testing is periodically conducted to determine if the modeling maintains acceptable tolerances and is within reasonable measurement parameters and documentation was provided as recommended by the FDA Guidance for 'Off-The-Shelf Software Use in Medical Devices'." Risk analysis determined "the severity of the harm that could result from a software failure is a minor level of hazard to patients."

Important Note: The document describes a workflow and quality control process for each individual product rather than a large-scale statistical study of the device as a whole against predefined performance metrics like sensitivity/specificity for a diagnostic device. The MirrorMe3D Modeling System is framed as an "image processing system" and "visualization system" that relies heavily on human expertise and approval.

Study Information

The document describes process validation and quality control measures for the manufacturing and utilization of the MirrorMe3D Modeling System for each specific case, rather than a single, large-scale clinical performance study with a test set in the traditional sense of an AI/diagnostic device submission.

Thus, many of your requested items for a "study" (like test set sample size, expert qualifications for ground truth establishment, MRMC studies) are not directly applicable or explicitly detailed in the provided text, as the "study" is more of an ongoing quality assurance and product-specific validation process.

Let's address the points as best as possible given the provided text:

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

Sample Size: Not applicable in the context of a single retrospective/prospective test set. The document states their process: "For each device, MirrorMe3D checks..." indicating that every individual case processed by MirrorMe3D undergoes a series of checks and reviews.
Data Provenance: Not specified. The input is "3D medical images," but the source (country, hospital) is not mentioned. Given it's a service, the data would come from individual clinicians/patients. The document also doesn't specify if the input images for the internal process validation were retrospective or prospective.

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

Number of Experts: For each case, at least one surgeon (Doctor) is involved in establishing the "ground truth" (i.e., the intended treatment outcome). MirrorMe3D trained personnel also assist.
Qualifications of Experts: The primary "expert" establishing the desired outcome (which acts as the ground truth for the model's design) is the "Doctor" (Surgeon). Specific qualifications like years of experience or board certification are not detailed in this document but are implicitly understood for a surgeon using such a system for treatment planning. MirrorMe3D also employs "trained personnel" to assist.

4. Adjudication method for the test set:

Adjudication Method: "The approval of the design of the anatomic model depicting the intended treatment outcome by the Doctor is required and indicates design acceptance." This suggests a single-expert approval model for each case. There is no mention of a multi-reader consensus or 2+1/3+1 adjudication for a test set.

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 done according to the provided text. The device is described as a "visualization system for measuring and treatment planning" and "image processing software" where "MirrorMe3D trained personnel will use off-the-shelf software to assist users in creating the 3D virtual (or digital) model that depicts the surgeon's intended outcome." This process does not describe an AI-assisted diagnostic workflow with human readers.

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

No, a standalone algorithm-only performance study was NOT done. The system explicitly requires "MirrorMe3D trained personnel" and "expert clinical judgment" (from the Doctor/Surgeon) to create the models and approve the design. It's a human-in-the-loop system. The document states, "MirrorMe3D trained personnel will use off-the-shelf software to assist users in creating the 3D virtual (or digital) model that depicts the surgeon's intended outcome."

7. The type of ground truth used:

The primary "ground truth" for the treatment plan and subsequent model design is the "surgeon's intended outcome" or "expert clinical judgment" as approved by the Doctor/Surgeon. For the physical models, the ground truth is adherence to the approved digital design and established geometric accuracy tolerances.

8. The sample size for the training set:

Not applicable as this document describes a quality control and process validation framework for a system that uses "off-the-shelf software" and relies on human intervention, rather than a deep learning AI model that requires a dedicated training set. The "off-the-shelf software" would have been developed and "trained" by its respective manufacturers, but that is not part of this submission for the MirrorMe3D system.

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

Not applicable. As explained in point 8, the MirrorMe3D system itself doesn't refer to a "training set" in the context of an AI algorithm that learns from data.

Ask a Question

Ask a specific question about this device

Page 1 of 1