(422 days)
The sterEOS Workstation is intended for use in the fields of musculoskeletal radiology and orthopedics in both pediatric and adult populations as a general PACS device for acceptance, transfer, display, storage, and digital processing of 2D X-ray images of the musculoskeletal system including interactive 2D measurement tools.
When using 2D X-ray images obtained with the Biospace EOS System (K071546), the sterEOS Workstation provides interactive 3D measurement tools to aid in the analysis of scoliosis and related disorders and deformities of the spine in adult patients as well as pediatric patients 7 years and older. The 3D measurement tools include interactive analysis based on a model of bone structures derived from an a priori image data set from 175 patients (91 normal patients, 47 patients with moderate idiopathic scoliosis and 37 patients with severe idiopathic scoliosis), and dry isolated vertebrae data. The model of bone structures is not intended for use in patients with a Cobb's angle > 50 degrees and is not intended for use to assess individual vertebral abnormalities.
The sterEOS Workstation is a general picture archiving and communications storage system for acceptance, transfer, display, storage, and digital processing of 2D X-ray images of the musculoskeletal system, including interactive 2D measurement tools.
When used with 2D X-ray images obtained with the Biospace EOS System (K071546), the sterEOS Workstation provides interactive 3D measurement tools to aid in the analysis of scoliosis and related disorders and deformities of the spine.
The provided text indicates that the sterEOS Workstation is a general PACS device with 2D and 3D measurement tools, particularly for scoliosis and spinal deformities. The performance data section broadly mentions "Accuracy studies of the 3D measurements were performed by using either anthropomorphic phantoms or patient images acquired with EOS system in hospitals. Results of these studies validate the 3D measurement tools of the software and demonstrate the equivalent performance of the device in both adult and pediatric populations."
However, the input does not contain specific acceptance criteria (e.g., minimum accuracy percentages, confidence intervals, comparison to a gold standard with numerical thresholds) or the detailed results of these studies. It also lacks information on sample sizes for test sets, data provenance, expert qualifications, adjudication methods, details of MRMC studies, standalone performance specifics, training set size, or how its ground truth was established.
Therefore, many parts of your request cannot be directly answered from the provided text.
Here's a summary of what can be extracted or inferred based on the given document, and where information is missing:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Not specified in document | "Accuracy studies of the 3D measurements... validate the 3D measurement tools of the software and demonstrate the equivalent performance of the device in both adult and pediatric populations." |
2. Sample size used for the test set and the data provenance
- Sample Size for Test Set: Not specified. The document mentions an "a priori image data set from 175 patients" used for a model of bone structures, but this is explicitly not the test set for performance evaluation of the 3D measurement tools themselves. This 175-patient dataset likely contributed to the development of the model, not necessarily the independent verification of its accuracy against new data.
- Data Provenance: "patient images acquired with EOS system in hospitals." The country of origin is not specified.
- Retrospective or Prospective: Not specified.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Not specified. The document does not mention the use of experts for ground truth establishment in the performance studies.
4. Adjudication method for the test set
- Not specified.
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. The document focuses on the software's 3D measurement accuracy rather than its impact on human reader performance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- The performance studies seem to focus on the accuracy of the "3D measurement tools of the software," implying a standalone evaluation of the algorithm's output for these measurements. However, the exact methodology (e.g., whether a human user manually applied the tools, or if it was fully automated) is not detailed. The device is described as having "interactive 3D measurement tools," which might suggest human interaction is part of its intended use and thus part of its performance.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- For studies using anthropomorphic phantoms, the ground truth would be based on the known, precise dimensions of the phantom.
- For studies using patient images, the type of ground truth is not explicitly stated. Given it's measuring 3D anatomical structures from 2D X-rays, the ground truth would likely involve highly precise 3D imaging (e.g., CT, MRI) or a manual gold standard obtained by expert measurement using high-resolution 3D data if available, or direct measurements on the phantoms.
8. The sample size for the training set
- The document mentions "a model of bone structures derived from an a priori image data set from 175 patients (91 normal patients, 47 patients with moderate idiopathic scoliosis and 37 patients with severe idiopathic scoliosis), and dry isolated vertebrae data." This 175-patient dataset, along with dry isolated vertebrae data, served as the basis for developing the bone structure model used by the 3D measurement tools. This would be considered the data used for training/development of the model.
9. How the ground truth for the training set was established
- For the "a priori image data set from 175 patients," the method for establishing ground truth for training the bone structure model is not detailed.
- For "dry isolated vertebrae data," the ground truth would likely be established through precise physical measurements of these specimens.
§ 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).