(44 days)
The FRAX™ 10-Year Fracture Risk software option is an accessory to currently marketed GE Lunar bone densitometer devices, which are intended to estimate the bone mineral density and body composition (lean and fat tissue mass) of patients when medically indicated by their physicians. This software option is intended to provide an assessment of 10-year fracture risk. The option provides an estimate of 10-year probability of hip fracture and 10-year probability of a major osteoporotic fracture (clinical spine, forearm, hip or shoulder fracture). This estimate is based on the patient's age, sex, country, ethnicity, height, weight, femur neck BMD T-score, and the presence of several risk factors and is computed using the FRAX™ Fracture Risk Assessment Tool endorsed by the World Health Organization (WHO). The tool has been validated for men and post-menopausal women between 40 and 90 years old. The output is provided in a separate screen display and report that can be viewed or printed or exported to an optional physician report generator tool. The results can be used by a physician in conjunction with other clinical risk factors as an aid in the diagnosis of osteoporosis and medical conditions leading to reduced bone density, and ultimately in the assessment of fracture risk.
This software option is intended to provide an assessment of 10-year fracture risk. The option provides an estimate of 10-year probability of hip fracture and 10-year probability of a major osteoporotic fracture (clinical spine, forearm, hip or shoulder fracture). This estimate is based on the patient's age, sex, country, ethnicity, height, weight, femur neck BMD T-score, and the presence or absence of several risk factors and is computed using the FRAX™ Fracture Risk Assessment Tool endorsed by the World Health Organization (WHO). The tool has been validated for men and post-menopausal women between 40 and 90 years old. The output is provided in a separate screen display and report that can be viewed or printed or exported to an optional physician report generator tool.
This response is based on the provided text. The document does not explicitly state acceptance criteria in a quantitative manner, nor does it provide a direct "device performance" in terms of clinical outcomes that are independently measured against such criteria. Instead, the submission relies on demonstrating substantial equivalence to predicate devices and conformance to published data and the WHO FRAX algorithm.
Here's a breakdown of the requested information based on the provided text:
Acceptance Criteria and Study to Prove Device Meets Acceptance Criteria
1. A table of acceptance criteria and the reported device performance:
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Consistency with WHO FRAX Algorithm Output | Outputs from the GE Lunar FRAX™ 10-year Fracture Risk Software were consistent with the independent outputs published that utilized the WHO 10-year fracture probability algorithm for specific age, sex, country, ethnicity, height, weight, femur neck BMD T-score, and the presence of several clinical risk factors. |
| Substantial Equivalence to Predicate Devices | The GE Lunar FRAX™ 10-year Fracture Risk Software Option is of a comparable type and substantially equivalent to the Fracture Risk for DPX bone densitometers and Fracture Risk from BMD using Hologic QDR x-ray Bone Densitometers. It has the same technological characteristics, is comparable in key safety and effectiveness features, it utilizes similar design, construction, and materials, and has similar intended uses as the predicate devices. |
| Conformance to Design Specifications and Industry Standards | The device has been evaluated for conformance to its design specifications and applicable industry standards for software development. The design and development process conforms with 21 CFR 820 and ISO 13485 quality systems. |
| Clinical Validation (FRAX Model) | The FRAX™ tool has been validated for men and post-menopausal women between 40 and 90 years old through extensive published literature describing its development and clinical validation. |
2. Sample size used for the test set and the data provenance:
- Test Set (for software output consistency): The document states "Design verification tests of the software option were performed using a series of test cases from published data." No specific sample size for these test cases is provided, nor are details on the country of origin or whether the data was retrospective or prospective. It implies the use of existing, published examples of FRAX calculations.
- Validation Cohorts (for FRAX model itself):
- Sample Size: 230,486 subjects.
- Follow-up: 1,208,528 patient years of follow-up.
- Data Provenance: "Eleven independent, population-based cohorts that did not participate in the model synthesis." The earlier description of the model development mentioned "centers in North America, Europe, Asia, and Australia," so it's reasonable to infer a similar, broad geographic distribution for the validation cohorts. The data is described as "follow up data from nine prospective, population-based cohorts" for model development, and implies the validation data is also prospective given the "patient years of follow-up."
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
This information is not provided for the software’s design verification test cases. The "ground truth" for these tests was the "independent outputs published that utilized the WHO 10-year fracture probability algorithm." This implies that the algorithm itself, as published by the WHO, served as the reference standard, rather than a panel of experts specifically forming a ground truth for this device's test set.
For the FRAX model's overall development and validation, experts presumably developed and endorsed the WHO FRAX algorithm, but the document does not specify their number or qualifications as part of this device's testing.
4. Adjudication method for the test set:
This information is not provided. The software performance was verified by checking consistency with "independent outputs published," which suggests a direct comparison method rather than an adjudication process involving human experts for the software itself.
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:
No MRMC comparative effectiveness study was performed or cited for this specific device. The device is an algorithm that provides a risk score, not an AI to assist human readers in image interpretation. No human-in-the-loop performance improvement data is mentioned.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
Yes, in essence. The primary "test" performed for this device was a non-clinical evaluation comparing its algorithm's outputs to "independent outputs published that utilized the WHO 10-year fracture probability algorithm." This is a standalone performance assessment of the algorithm's consistency with the established WHO FRAX standard.
7. The type of ground truth used:
- For the device's software verification: The ground truth was the published outputs of the WHO 10-year fracture probability algorithm.
- For the underlying FRAX model's validation: The ground truth was observed fracture events (hip fractures and other osteoporotic fractures) recorded during "person years of follow-up" in large, population-based cohorts.
8. The sample size for the training set:
- For the FRAX model development: The model was "developed from baseline and follow up data from nine prospective, population-based cohorts... consisting of 46,340 people and 189,852 person years of follow-up." This represents the core "training" data for the FRAX algorithm itself.
- For the GE Lunar device's software: No specific "training set" for the software is mentioned, as it implements an existing, published algorithm (FRAX).
9. How the ground truth for the training set was established:
- For the FRAX model development:
- Clinical Risk Factors: "Appropriate clinical risk factors for fracture were determined from eleven large cohort studies."
- Outcome Data: The ground truth was the occurrence of actual fracture events (hip fractures and other osteoporotic fractures) observed during the "189,852 person years of follow-up." These events were "reported," implying clinical diagnosis and documentation.
- The data was "generally randomly selected and analyzed according to accepted epidemiological practices" and "published in peer-reviewed journals," suggesting rigorous methodology for establishing these outcomes.
§ 892.1170 Bone densitometer.
(a)
Identification. A bone densitometer is a device intended for medical purposes to measure bone density and mineral content by x-ray or gamma ray transmission measurements through the bone and adjacent tissues. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.