(54 days)
Saige-Dx analyzes digital breast tomosynthesis (DBT) mammograms to identify the presence or absence of soft tissue lesions and calcifications that may be indicative of cancer. For a given DBT mammogram, Saige-Dx analyzes the DBT image stacks and the accompanying 2D images, including full field digital mammography and/or synthetic images. The system assigns a Suspicion Level, indicating the strength of suspicion that cancer may be present, for each detected finding and for the entire case. The outputs of Saige-Dx are intended to be used as a concurrent reading aid for interpreting physicians on screening mammograms with compatible DBT hardware.
Saige-Dx is a software device that processes screening mammograms using artificial intelligence to aid interpreting radiologists. By automatically detecting the presence or absence of soft tissue lesions and calcifications in mammography images, Saige-Dx can help improve reader performance, while also reducing reading time. The software takes as input a set of x-ray mammogram DICOM files from a single digital breast tomosynthesis (DBT) study and generates finding-level outputs for each image analyzed, as well as an aggregate case-level assessment. Saige-Dx processes both the DBT image stacks and the associated 2D images (full-field digital mammography (FFDM) and/or synthetic 2D images) in a DBT study. For each image, Saige-Dx outputs bounding boxes circumscribing any detected findings and assigns a Finding Suspicion Level to each finding, indicating the degree of suspicion that the finding is malignant. Saige-Dx uses the results of the finding-level analysis to generate a Case Suspicion Level, indicating the degree of suspicion for malignancy across the case. Saige-Dx encapsulates the finding and case-level results into a DICOM Structured Report (SR) object containing markings that can be overlaid on the original mammogram images using a viewing workstation and a DICOM Secondary Capture (SC) object containing a summary report of the Saige-Dx results.
Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) clearance letter for Saige-Dx:
1. Table of Acceptance Criteria and Reported Device Performance
The provided document indicates that the primary endpoint of the standalone performance testing was to demonstrate non-inferiority of the subject device (new Saige-Dx version) to the predicate device (previous Saige-Dx version). Specific quantitative acceptance criteria (e.g., AUC, sensitivity, specificity thresholds) are not explicitly stated in the provided text. However, the document states:
"The test met the pre-specified performance criteria, and the results support the safety and effectiveness of Saige-Dx updated AI model on Hologic and GE exams."
| Acceptance Criteria (Not explicitly quantified in source) | Reported Device Performance |
|---|---|
| Non-inferiority of subject device performance to predicate device performance. | "The test met the pre-specified performance criteria, and the results support the safety and effectiveness of Saige-Dx updated AI model on Hologic and GE exams." |
| Performance across breast densities, ages, race/ethnicities, and lesion types and sizes. | Subgroup analyses "demonstrated similar standalone performance trends across breast densities, ages, race/ethnicities, and lesion types and sizes." |
| Software design and implementation meeting requirements. | Verification testing including unit, integration, system, and regression testing confirmed "the software, as designed and implemented, satisfied the software requirements and has no unintentional differences from the predicate device." |
2. Sample Size for the Test Set and Data Provenance
- Sample Size for Test Set: 2,002 DBT screening mammograms from unique women.
- 259 cancer cases
- 1,743 non-cancer cases
- Data Provenance:
- Country of Origin: United States (cases collected from 12 diverse clinical sites).
- Retrospective or Prospective: Retrospective.
- Acquisition Equipment: Hologic (standard definition and high definition) and GE images.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
The document mentions: "The case collection and ground truth lesion localization processes of the newly collected cases were the same processes used for the previously collected test dataset (details provided in K220105)."
- While the specific number and qualifications of experts for the ground truth of the current test set are not explicitly detailed in this document, it refers back to K220105 for those details. It implies that a standardized process involving experts was used.
4. Adjudication Method for the Test Set
The document does not explicitly describe the adjudication method (e.g., 2+1, 3+1) used for establishing ground truth for the test set. It states: "The case collection and ground truth lesion localization processes of the newly collected cases were the same processes used for the previously collected test dataset (details provided in K220105)." This suggests a pre-defined and presumably robust method for ground truth establishment.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- Was it done? Yes.
- Effect Size: The document states: "a multi-reader multi-case (MRMC) study was previously conducted for the predicate device and remains applicable to the subject device." It does not provide details on the effect size (how much human readers improve with AI vs. without AI assistance) within this document. Readers would need to refer to the K220105 submission for that information if it was presented there.
6. Standalone (Algorithm Only) Performance Study
- Was it done? Yes.
- Description: "Validation of the software was conducted using a retrospective and blinded multicenter standalone performance testing under an IRB approved protocol..."
- Primary Endpoint: "to demonstrate that the performance of the subject device was non-inferior to the performance of the predicate device."
7. Type of Ground Truth Used
- The ground truth involved the presence or absence of cancer, with cases categorized as 259 cancer and 1,743 non-cancer. The mention of "ground truth lesion localization processes" implies a detailed assessment of findings, likely involving expert consensus and/or pathology/biopsy results to confirm malignancy. Given it's a diagnostic aid for cancer, pathology is the gold standard for confirmation.
8. Sample Size for the Training Set
- Training Dataset: 161,323 patients and 300,439 studies.
9. How the Ground Truth for the Training Set Was Established
- The document states: "The Saige-Dx algorithm was trained on a robust and diverse dataset of mammography exams acquired from multiple vendors including GE and Hologic equipment."
- While it doesn't explicitly detail the method of ground truth establishment for the training set (e.g., expert consensus, pathology reports), similar to the test set, for a cancer detection AI, it is highly probable that the ground truth for the training data was derived from rigorous clinical assessments, including follow-up, biopsy results, and/or expert interpretations, to accurately label cancer and non-cancer cases for the algorithm to learn from. The implied "robust and diverse" nature of the training data suggests a comprehensive approach to ground truth.
§ 892.2090 Radiological computer-assisted detection and diagnosis software.
(a)
Identification. A radiological computer-assisted detection and diagnostic software is an image processing device intended to aid in the detection, localization, and characterization of fracture, lesions, or other disease-specific findings on acquired medical images (e.g., radiography, magnetic resonance, computed tomography). The device detects, identifies, and characterizes findings based on features or information extracted from images, and provides information about the presence, location, and characteristics of the findings to the user. The analysis is intended to inform the primary diagnostic and patient management decisions that are made by the clinical user. The device is not intended as a replacement for a complete clinician's review or their clinical judgment that takes into account other relevant information from the image or patient history.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include:
(i) A detailed description of the image analysis algorithm, including a description of the algorithm inputs and outputs, each major component or block, how the algorithm and output affects or relates to clinical practice or patient care, and any algorithm limitations.
(ii) A detailed description of pre-specified performance testing protocols and dataset(s) used to assess whether the device will provide improved assisted-read detection and diagnostic performance as intended in the indicated user population(s), and to characterize the standalone device performance for labeling. Performance testing includes standalone test(s), side-by-side comparison(s), and/or a reader study, as applicable.
(iii) Results from standalone performance testing used to characterize the independent performance of the device separate from aided user performance. The performance assessment must be based on appropriate diagnostic accuracy measures (
e.g., receiver operator characteristic plot, sensitivity, specificity, positive and negative predictive values, and diagnostic likelihood ratio). Devices with localization output must include localization accuracy testing as a component of standalone testing. The test dataset must be representative of the typical patient population with enrichment made only to ensure that the test dataset contains a sufficient number of cases from important cohorts (e.g., subsets defined by clinically relevant confounders, effect modifiers, concomitant disease, and subsets defined by image acquisition characteristics) such that the performance estimates and confidence intervals of the device for these individual subsets can be characterized for the intended use population and imaging equipment.(iv) Results from performance testing that demonstrate that the device provides improved assisted-read detection and/or diagnostic performance as intended in the indicated user population(s) when used in accordance with the instructions for use. The reader population must be comprised of the intended user population in terms of clinical training, certification, and years of experience. The performance assessment must be based on appropriate diagnostic accuracy measures (
e.g., receiver operator characteristic plot, sensitivity, specificity, positive and negative predictive values, and diagnostic likelihood ratio). Test datasets must meet the requirements described in paragraph (b)(1)(iii) of this section.(v) Appropriate software documentation, including device hazard analysis, software requirements specification document, software design specification document, traceability analysis, system level test protocol, pass/fail criteria, testing results, and cybersecurity measures.
(2) Labeling must include the following:
(i) A detailed description of the patient population for which the device is indicated for use.
(ii) A detailed description of the device instructions for use, including the intended reading protocol and how the user should interpret the device output.
(iii) A detailed description of the intended user, and any user training materials or programs that address appropriate reading protocols for the device, to ensure that the end user is fully aware of how to interpret and apply the device output.
(iv) A detailed description of the device inputs and outputs.
(v) A detailed description of compatible imaging hardware and imaging protocols.
(vi) Warnings, precautions, and limitations must include situations in which the device may fail or may not operate at its expected performance level (
e.g., poor image quality or for certain subpopulations), as applicable.(vii) A detailed summary of the performance testing, including test methods, dataset characteristics, results, and a summary of sub-analyses on case distributions stratified by relevant confounders, such as anatomical characteristics, patient demographics and medical history, user experience, and imaging equipment.