(404 days)
The Digimamo D ® Full-Field Digital Mammography System is a device intended to produce planar digital x-ay images of the entire breast. The Digimamo D is indicated for generating mammographic images that can be used for screening and diagnosis of breast cancer. The Digimamo D is intended to be used in the same clinical applications as traditional film/screen systems.
This is a mammography system available in only a digital version. The main characteristics are: Operation Panel/Console, X-ray tube information, Filter Selection, Window Material and thickness, Focal Spot Sizes, Operating principle, Radiological Characteristics, Column and Gantry, Flat Panel Detector, Workstation.
The provided text describes the Digimamo D Full-Field Digital Mammography System and its comparison to a predicate device (Adani MammoScan K172027) and a reference device (DRTECH RSM2430C K170930). The information primarily focuses on the device's technical specifications and a summary of non-clinical and clinical testing performed to establish substantial equivalence.
Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:
Acceptance Criteria and Reported Device Performance
The acceptance criteria are not explicitly listed in a table format with specific quantitative thresholds. Instead, the document states that the device was tested to establish "substantial equivalence" to a predicate device, meaning it is "as safe and effective." The performance is deemed acceptable if the images are of "sufficiently acceptable quality for mammographic usage."
However, we can infer some "acceptance" from the successful outcome of the comparison and the conclusion of substantial equivalence. The key is that the new device's performance aligns with the predicate device and that the images were found to be diagnostic quality by experts.
| Acceptance Criteria Category | Description of Performance/Evidence Provided in Document |
|---|---|
| Device Safety | Passed various IEC 60601 series standards (60601-1, 60601-1-2, 60601-1-3, 60601-2-28, 60601-2-45) by 3rd party Nationally Recognized Testing Laboratories. Complies with 21 CFR 1020.30 and 1020.31. Firmware validated per FDA Guidance (May 11, 2005). Cybersecurity recommendations observed. |
| Device Effectiveness (Image Quality) | Clinical image evaluation by expert radiologists determined that the clinical images were of "sufficiently acceptable quality for mammographic usage." The device employs the same imaging panel/software cleared by DRTECH in RSM2430C (K170930), and it "worked properly and produced diagnostic quality mammographic images." |
| Substantial Equivalence to Predicate Device | The document concludes that "the new Digimamo D System is as safe and effective as the predicate device, have few technological differences, and has the same indications for use, thus rendering it substantially equivalent to the predicate device." This implicitly means the performance is comparable to the predicate for all relevant aspects (e.g., image quality, dose, functionality). |
| Indicated Use | The device's Indications for Use are stated to be "SAME" as the predicate device: "intended to produce planar digital x-ray images of the entire breast. The Digimamo D is indicated for generating mammographic images that can be used for screening and diagnosis of breast cancer. The Digimamo D is intended to be used in the same clinical applications as traditional film/screen systems." Performance meets this stated intended use. |
| Hardware & Software Requirements | The manufacturer states they "meet the hardware and software requirements expressed in that submission" (referring to the DRTECH submission K170930). This implies their system's components and software function as expected and are comparable to a cleared device. |
Study Details:
-
Sample size used for the test set and the data provenance:
- Test Set Sample Size: "Sixty images" were evaluated. The document further clarifies "Totally, there are 12 sets of images presented in this review." Each set consists of four standard views (right CC, right MLO, left CC, left MLO), so 12 sets * 4 views/set = 48 images initially. However, it also states "If requested by the radiologist, additional image with magnification was taken." This suggests more than 48 images could have been evaluated, up to 60 total. It's not explicitly clear if all "sixty images" were unique patient cases or if "12 sets of images" corresponds to 12 patient cases, each with multiple views. The phrasing "Sixty images were evaluated" followed by "Totally, there are 12 sets of images" could be interpreted as 60 individual images selected from these 12 sets, or 60 total images including any additional magnification views for the 12 cases.
- Data Provenance: Images were taken during the "standard screening procedure in Fundação Municipal de Saúde de Macaé, Rio de Janeiro, Brazil." This indicates the data is retrospective as it was already acquired.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Number of Experts: "Three Board Certified US based radiologists."
- Qualifications of Experts: "US-based MQSA qualified radiologists." MQSA (Mammography Quality Standards Act) qualification ensures competence in mammography interpretation.
-
Adjudication method for the test set:
- The document states, "Images were reviewed by US-based MQSA qualified radiologists. The results of clinical image evaluation determined that the clinical images reviewed by the expert radiologists were of sufficiently acceptable quality for mammographic usage."
- There is no explicit description of an adjudication method (e.g., 2+1, 3+1 consensus). It sounds like a qualitative assessment by individual radiologists, and the collective outcome was that the images were acceptable. It doesn't describe a process for resolving disagreements or reaching a formal consensus for a specific ground truth label.
-
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, an MRMC comparative effectiveness study was not done. The study described is a clinical image evaluation by human readers of images produced by the device to confirm diagnostic quality, not a study comparing human performance with vs. without AI assistance. The device itself is a full-field digital mammography system, not an AI-based reading aid.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Not explicitly described as a standalone algorithm study. The device is an imaging system (hardware + software for image acquisition and display). The "clinical testing" refers to human readers evaluating the image output of the system, not an algorithm's classification performance. The "digital panel software employed was already reviewed by FDA in the reference submission list," which implies its performance characteristics (e.g., image quality metrics) would have been assessed as part of its original clearance. This is an assessment of the entire system's output through human review.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- The "ground truth" for the clinical image evaluation was the qualitative assessment by the three Board Certified US-based MQSA qualified radiologists that the images were of "sufficiently acceptable quality for mammographic usage." This is a form of expert opinion/consensus on image quality for diagnostic purposes. It is not described as being based on pathology or clinical outcomes.
-
The sample size for the training set:
- The document does not mention a training set sample size as this device is a mammography imaging system (hardware and associated acquisition/processing software), not an AI algorithm requiring a training set in the conventional sense (e.g., for classification or detection). The software involved (digital panel software) was from a previously cleared device (DRTECH RSM2430C K170930), implying its development and validation were part of that prior submission.
-
How the ground truth for the training set was established:
- As no "training set" for an AI algorithm is described for this device's new performance evaluation, the method for establishing ground truth for a training set is not applicable/not provided in this document.
§ 892.1715 Full-field digital mammography system.
(a)
Identification. A full-field digital mammography system is a device intended to produce planar digital x-ray images of the entire breast. This generic type of device may include digital mammography acquisition software, full-field digital image receptor, acquisition workstation, automatic exposure control, image processing and reconstruction programs, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II (special controls). The special control for the device is FDA's guidance document entitled “Class II Special Controls Guidance Document: Full-Field Digital Mammography System.”See § 892.1(e) for the availability of this guidance document.