(132 days)
iBreastExam Gen II device is intended to produce a surface pressure map of the breast as an aid in documenting palpable breast lesions identified during a clinical breast exam. iBreastExam Gen II device is intended for use by a qualified healthcare professional trained in its use and is not for home use.
iBreastExam Gen II is a visual mapping system, intended to produce a surface pressure map of the breast as an aid in documenting breast lesions identified during a clinical breast exam. The use of iBreastExam Gen II does not involve the use of any ionizing radiation. The iBreastExam Gen II system consists of a group of hardware and software components. The hardware component(s) consist of: iBreastExam Gen II Scanhead, Computer System, USB Wall Charger and Charging Cables. The software component(s) consist of: iBEConnect Software Program.
The iBreastExam Gen II device is intended to produce a surface pressure map of the breast to aid in documenting palpable breast lesions identified during a clinical breast exam. It is not for diagnosis, but purely a documentation tool. Clinical management decisions should be based on clinical and diagnostic examinations (e.g., ultrasound or mammography).
Here's a breakdown of the acceptance criteria and study details:
1. Table of Acceptance Criteria and Reported Device Performance
The provided text describes performance testing related to accuracy and reproducibility, but it does not explicitly state specific pass/fail acceptance criteria (e.g., "accuracy must be greater than X%") for the device's performance characteristics in a numerical form. Instead, the conclusions for each test conducted state that the device "produced accurate and reproducible results." The closest numerical performance characteristic mentioned is "Detectable lesion size: 5mm and above" for both the subject and predicate devices, which can be inferred as a performance expectation.
| Acceptance Criteria (Inferred/Stated) | Reported Device Performance |
|---|---|
| Inter-observer accuracy and reproducibility (of location, size, stiffness, and shape) | iBreastExam Gen II produced accurate and reproducible results. |
| Intra-observer accuracy and reproducibility (of location, size, stiffness, and shape) | iBreastExam Gen II produced accurate and reproducible results. |
| Inter-system accuracy and reproducibility (of location, size, stiffness, and shape) | iBreastExam Gen II produced accurate and reproducible results. |
| Detectable lesion size | 5mm and above (same as predicate) |
| Output of the device (compared to predicate) | Output of iBreastExam Gen II system is identical to that of predicate iBreastExam (K142926). |
| Electrical and mechanical safety, and electromagnetic compatibility | All test results were satisfactory and did not raise any additional safety or effectiveness concerns. |
2. Sample Size Used for the Test Set and Data Provenance
For the inter-observer, intra-observer, and inter-system bench tests, the sample size used was 20 subjects.
The tests were performed on silicone phantoms constructed by layering slabs of silicon with lesions placed in between. This indicates that the testing was conducted prospectively in a controlled lab environment using simulated tissue, rather than on human patients or retrospective clinical data. Therefore, there is no country of origin for patient data in this context.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts
The ground truth for the test set (silicone phantoms) was established by the physical construction of the phantoms themselves, which included lesions of defined location, size, stiffness, and shape. The "experts" in this context were effectively the creators of the phantoms, who embedded the known characteristics.
The testing methodology involved "recruits" (20 subjects) who were "instructed on how to use iBreastExam Gen II and was allowed to perform several practice exams prior to the study." These recruits were the operators performing the measurements, not experts establishing ground truth or interpreting results, as the device's purpose is documentation of palpable lesions. The document does not specify qualifications for these recruits beyond being trained to use the device.
4. Adjudication Method for the Test Set
There was no adjudication method described in the context of establishing ground truth for the test set, as the ground truth was inherently known from the design of the silicone phantoms (known lesion location, size, stiffness, and shape).
For the inter-observer and intra-observer tests, the reproducibility and accuracy were measured against this known phantom ground truth, and for inter-system, it was comparing measurements from different devices against the same phantom.
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
There was no MRMC comparative effectiveness study mentioned in the provided text. The device, iBreastExam Gen II, is described as a "breast lesion documentation system" that produces a "surface pressure map." Its function is to aid in documenting already palpable lesions and is not intended for diagnosis. The study described focuses on the device's accuracy and reproducibility in measuring known characteristics on phantoms when operated by different individuals or with different devices, not on how it assists human readers in diagnosis or detection. The device itself is not an AI-assisted diagnostic tool.
6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done
The provided text describes direct interaction between a human operator ("recruit") and the device. The device collects data based on human operation (placing the scanhead on the phantom) and then presents a tactile map via software. The "iBEConnect Software Program" provides positioning guidance and processes data. However, the performance tests described (inter-observer, intra-observer, inter-system) inherently include a human operator in the loop. The device is explicitly stated to be for use by a "qualified healthcare professional trained in its use." Therefore, a standalone (algorithm only without human-in-the-loop) performance study was not explicitly described or, due to the nature of the device, likely conducted as its intended use requires interaction.
7. The Type of Ground Truth Used
The ground truth used for the performance testing was known physical characteristics of silicone phantoms (defined lesion location, size, stiffness, and shape). This is a form of engineered or simulated ground truth, rather than expert consensus, pathology, or outcomes data from human patients.
8. The Sample Size for the Training Set
The document does not describe a "training set" in the context of machine learning or AI. The tests performed are verification and validation of the device's measurement capabilities. If any internal calibration or parameter setting during device development involved data, it is not mentioned. The user training described for the 20 recruits refers to training them on how to use the device, not training the device's internal algorithms.
9. How the Ground Truth for the Training Set Was Established
Since no specific "training set" for the device's algorithms or software was described (beyond general software validation), the method for establishing its ground truth is not applicable or detailed in the provided information. The device functions by physical principles (capacitive pressure sensors) and software processing, not through a learning algorithm that requires a separate training dataset with established ground truth.
§ 884.2990 Breast lesion documentation system.
(a)
Identification. A breast lesion documentation system is a device for use in producing a surface map of the breast as an aid to document palpable breast lesions identified during a clinical breast examination.(b)
Classification. Class II (special controls). The device, when it is a breast examination recording sheet, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 884.9. The special control is FDA's guidance entitled “Class II Special Controls Guidance Document: Breast Lesion Documentation System.” See § 884.1(e) for the availability of this guidance document.