(88 days)
Bonelogic software is intended to be used by specialized medical practitioners to assist in the characterization of human anatomy with 3D visualization and specific measurements. The medical image modalities intended to be used in the software are computed tomography (CT) images, cone beam computed tomography (CBCT) images and weight-bearing cone beam CT (WBCT) images. The intended patient population is adults over 16 years of age.
Bonelogic software contains the measurement template with a set of distance and angular measurements can be used for diagnostic purposes. The three dimensional (3D) models are displayed and can be manipulated in the software. Together, the information from the measurements and the 3D visualization can be used for treatment planning in the field of orthopedics (foot and ankle, and hand wrist). The 3D models can be outputted from the software for traditional or additive manufacturing. The physical models generated based on the 3D digital models are not intended for diagnostic use.
Bonelogic product is a software tool to be used by specialized medical practitioners. The software tool is aimed to help the user in the characterization of human anatomy, and identifying possible trauma or deformities, the diagnose and the treatment planning should always be based on the professional skills of the specialist doctor. The medical image modalities intended to be used in the software are computed tomography (CT) images, cone beam computed tomography (CBCT) images and weight-bearing cone beam CT (WBCT) images. Bonelogic software has got a modular architecture. The software includes following functionality:
- Importing medical images in DICOM format
- Viewing of DICOM data
- Selecting a region of interest using generic segmentation tools
- Segmenting specific anatomy using dedicated semi-automatic tools or fully automatic algorithms
- Verifying and editing a region of interest
- Calculating a digital 3D model and editing the model
- Measuring on 3D models
- Exporting images, measurements, and 3D models to third-party packages
- Planning treatments on the 3D models
- Interfacing with packages for Finite Element Analysis
The Bonelogic software aims to assist specialized medical practitioners in characterizing human anatomy using 3D visualization and specific measurements. The device processes CT, CBCT, and WBCT images to create 3D models and perform measurements for diagnostic and treatment planning purposes, particularly in orthopedics (foot and ankle, hand and wrist).
Here's a breakdown of the acceptance criteria and the study proving the device meets them:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Implicit) | Reported Device Performance |
|---|---|
| Geometric Accuracy of 3D Models: The subject device's 3D models should be geometrically accurate when compared to models from a predicate device. | The geometric accuracy of 3D virtual models created in the subject device (Bonelogic) was assessed against similar virtual models created with the predicate device (Mimics Medical, K183105). The study concluded that performance testing demonstrated device performance and substantial equivalence to the predicate device. Although no specific metrics (e.g., mean absolute difference, Dice similarity coefficient) are provided, the general statement supports the criterion. |
| Measurement Accuracy: Manual measurements of radiological parameters performed by clinicians should be comparable to measurements generated by the subject device. | The measurement accuracy in the subject device (Bonelogic) was assessed by comparing manual measurements of radiographical parameters against the same measurements created in the subject device. Manual measurements were performed by clinicians. The study concluded that performance testing demonstrated device performance and substantial equivalence to the predicate device. Again, specific quantitative error metrics are not reported in this summary. |
| Performance Testing against Defined Requirements: The device should meet its defined requirements. | Verification against defined requirements via performance testing was conducted. This included testing on measurement repeatability. The study concluded that all performance testing conducted demonstrated device performance and substantial equivalence. No specific details about the defined requirements or quantitative results of the repeatability testing are included. |
| Clinical Validation (Usability/Accuracy): The accuracy of the 3D virtual models generated by the device should be validated against original DICOM imaging data, and the usability in a clinical setting should be validated. | Validation for the subject device against user needs via clinical validation for the usability of 3D models in a clinical setting was performed. Clinical validation for the accuracy of 3D virtual models against original DICOM imaging data was also performed. The general conclusion is that performance testing demonstrated device performance and substantial equivalence to the predicate device. However, specific details about the clinical validation results (e.g., user satisfaction, quantitative accuracy metrics) are not provided. |
| Overall Substantial Equivalence: The device should be as safe and effective and perform as well as the predicate device. | The summary explicitly states: "A comparison of intended use and technological characteristics combined with performance data demonstrates that Bonelogic software is substantially equivalent to the predicate device Mimics Medical (K183105). Minor differences in intended use and technological characteristics exist, but performance data demonstrates that Bonelogic software is as safe and effective and performs as well as the predicate device." |
2. Sample Size Used for the Test Set and Data Provenance
- Geometric Accuracy Study (3D Models): The test set included "DICOM images containing foot and ankle anatomy." The specific sample size (number of cases or patients) is not provided.
- Measurement Accuracy Study: The test set included "DICOM images containing hand and wrist anatomy." The specific sample size is not provided.
- Provenance: The document does not specify the country of origin of the data or whether the data was retrospective or prospective.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
- For the measurement accuracy study, "manual measurements were performed by clinicians." The number of clinicians is not specified. Their specific qualifications (e.g., "radiologist with 10 years of experience") are also not detailed beyond "clinicians."
- For the accuracy of 3D virtual models, the validation was against "original DICOM imaging data." It is implied that human experts would interpret this raw data, but the number and qualifications of these experts are not explicitly stated.
4. Adjudication Method for the Test Set
The document does not describe any specific adjudication method (e.g., 2+1, 3+1) used to establish ground truth for the test set. For the measurement accuracy, it mentions "manual measurements were performed by clinicians," implying these measurements served as a ground truth comparison, but how consensus was reached if multiple clinicians were involved is not detailed.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
The document does not mention an MRMC comparative effectiveness study that directly quantifies human readers' improvement with AI vs. without AI assistance. The performance studies primarily focus on comparing the device's output to manual measurements or predicate device outputs, not on the interaction of the device with human readers and its impact on their diagnostic performance.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study
Yes, standalone performance was assessed.
- The "geometric accuracy of 3D virtual models created in the subject device Bonelogic software" was compared against the predicate device's models. This tests the algorithm's output directly.
- The "measurement accuracy in the subject device Bonelogic software" was assessed by comparing the device's generated measurements against manual measurements. This also evaluates the algorithm's standalone capability.
- Verification via "performance testing on measurements repeatability" assesses the algorithm's consistency.
7. Type of Ground Truth Used
- Expert Consensus: Implied for the "manual measurements performed by clinicians" in the measurement accuracy study.
- Comparison to Predicate Device Output: For the geometric accuracy of 3D models, the ground truth was essentially the output of the predicate device (Mimics Medical).
- Original DICOM Imaging Data: For the accuracy of 3D virtual models, they were validated against "original DICOM imaging data," which serves as the foundational ground truth.
8. Sample Size for the Training Set
The document does not provide any information about the sample size used for the training set for the Bonelogic software's algorithms (e.g., for segmentation, measurement automation).
9. How Ground Truth for the Training Set Was Established
The document does not provide details on how the ground truth for any potential training set was established. Given the nature of the device (image processing for measurements and 3D models), it is likely that expert-annotated CT/CBCT/WBCT images would be used for training, but this is not mentioned.
§ 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).