Search Results

The MicroDose SI (model L50) is indicated for generating mammographic images that can be used for screening and diagnosis of breast cancer. The MicroDose SI (model L50) is intended to be used in the same clinical applications as traditional film/screen systems.

This application is to add functionality to the previously cleared MicroDose SI, Model L50, (K123995) that allows for breast density measurements. The L50 submission described that the detector electronics have been modified to do spectral imaging through the functionality to discriminate between high and low energy photons based on an energy threshold. The first application of this technology is Spectral Breast Density Measurement where breast glandularity and thickness is measured simultaneously by the spectral difference in X-ray attenuation between adipose and fibroglandular tissue. The MicroDose'Sl can thereby measure volumetric glandularity and breast thickness for each pixel in the image. Total breast volume is calculated by integrating the breast thickness over the breast region. taking pixel size and beam geometry into account. Fibroglandular thickness is calculated as the product of breast thickness and volumetric glandularity for each pixel. Total fibroglandular volume is then calculated by integrating the fibroglandular thickness over the breast region. The total breast volumetric glandularity is then given as the ratio of fibroglandular to breast volume. The correlation between the MicroDose Density Score and the radiologist's BI-RADS score has The following Information is sent as part of the content of a DICOM Structured Report, SR, to selected destinations (for example a PACS). Breast analysis information is also included as private attributes in the DICOM header of the acquired images.

The Philips MicroDose SI (model L50) is an add-on functionality to a previously cleared full-field digital mammography (FFDM) system that allows for breast density measurements. The device measures volumetric glandularity, glandular volume, breast volume, and provides a breast density score correlated to BI-RADS categories.

Here's an analysis of its acceptance criteria and the study performance:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the MicroDose SI Breast Density Measurement application were established by evaluating its parameters against industry standard breast tissue equivalent phantoms and a substantial clinical screening population dataset. The exact quantitative acceptance criteria (e.g., specific thresholds for correlation coefficients or agreement percentages) are not explicitly stated in the provided text. However, the study aimed to confirm several aspects:

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size: "a substantial clinical screening population dataset"
• Data Provenance: The document does not explicitly state the country of origin. It mentions "a substantial clinical screening population dataset," implying real-world clinical data. It is a retrospective study based on an existing dataset.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

• Number of Experts: Not explicitly stated, but "MQSA qualified radiologists" were involved.
• Qualifications of Experts: "MQSA qualified radiologists." MQSA (Mammography Quality Standards Act) qualification ensures radiologists are trained and certified to interpret mammograms in the United States.

4. Adjudication Method for the Test Set

The document does not explicitly state an adjudication method (e.g., 2+1, 3+1). It states "the BI-RADS breast composition score given by MQSA qualified radiologists," implying that the radiologists' scores were used as a reference for comparison, but doesn't detail how discrepancies among multiple radiologists would have been resolved if applicable.

5. Multi-Reader-Multi-Case (MRMC) Comparative Effectiveness Study

• Was an MRMC study done? No, a traditional MRMC comparative effectiveness study, which typically evaluates human readers with and without AI assistance, was not explicitly described. The study focused on the correlation between the device's density scores and radiologists' BI-RADS scores, as well as internal consistency and expected biological correlations, rather than the improvement of human reader performance with AI assistance.
• Effect size of human reader improvement: Not applicable, as an MRMC comparitive effectiveness study with human-in-the-loop performance measurement was not described.

6. Standalone Performance (Algorithm Only without Human-in-the-Loop Performance)

Yes, a standalone performance assessment was conducted. The MicroDose SI L50 Breast Density Measurement application's ability to measure and output volumetric glandularity, glandular volume, breast volume, and a breast density score is an inherent standalone function. The study evaluated these independent measurements against:

• Industry standard breast tissue equivalent phantoms.
• Correlation with radiologists' BI-RADS scores.
• Internal consistency (across views, laterality).
• Expected biological correlations (with age, breast volume).

7. Type of Ground Truth Used

The ground truth for the clinical correlation aspects of the test set was:

• Expert Consensus: Specifically, "the BI-RADS breast composition score given by MQSA qualified radiologists." This implies expert interpretation of the mammograms.
• Physiological/Biological Correlates: Age and breast volume were used to confirm expected biological relationships with volumetric glandularity.
• Physical Phantoms: Used to validate density measurements against known standards.

8. Sample Size for the Training Set

The document does not specify the sample size for the training set. It mentions "a substantial clinical screening population dataset" for the evaluation (test set), but does not distinguish this from a training set.

9. How the Ground Truth for the Training Set Was Established

The document does not provide details on how the ground truth for any potential training set was established. The focus of the provided text is on the validation activities of the developed application.

Ask a specific question about this device

The MicroDose SI (model L50) is indicated for generating mammographic images that can be used for screening and diagnosis of breast cancer. The MicroDose SI (model L50) is intended to be used in the same clinical applications as traditional film/screen systems.

The MicroDose SI (model L50) is a full-field digital mammography system comprised of an image acquisition system, a gantry and an acquisition workstation computer equipped with a keyboard, a keypad, a mouse, and a monitor. The image acquisition system includes a digital detector of photon counting technology, x-ray tube (with tungsten target and aluminum filtration), high voltage generator, compression paddle(s), and multi-slit collimator. The acquisition workstation is the user interface for preparing and initiating image acquisition, image processing, and image transfer to the desired destination (e.g. PACS) for diagnosis and archiving.

The MicroDose SI (model L50) detector is based on photon counting technology and consists of a large number of crystalline silicon strip detectors. The technology enables high detection efficiency of photons and efficient rejection of electronic noise. The MicroDose SI (model L50) uses a multi-slit scanning technique that prevents image degradation caused by scattered radiation by removing photons scattered in the breast and not directed towards the detector. These factors combine into a dose efficient system.

The MicroDose SI (model L50) provides three exposure modes; manual, automatic (parameters predefined based on compressed breast thickness), and SmartAEC continuously adjusts the exposure based on measured transmission from the leading detector edge.

The changes compared to the L30 model involves a minor modification to the collimator, such that the range of breast thickness that can be imaged on the device is slightly expanded.

The provided 510(k) summary for the PHILIPS MicroDose SI (model L50) does not contain a detailed study proving device performance against specific acceptance criteria. This document focuses on demonstrating substantial equivalence to a predicate device (Philips MicroDose L30) rather than presenting a performance study with acceptance criteria.

However, based on the information provided, we can infer some aspects and highlight what is missing:

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

This information is not provided in the given 510(k) summary. The document states "The equivalence in performance is supported through the performed verification and validation activities" and "physical laboratory test results," but it does not detail the nature, sample size, or provenance of these tests.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications

This information is not provided in the given 510(k) summary. Given the focus on technical equivalence through minor modifications, it's possible that clinical ground truth establishment with expert readers was not deemed necessary for this specific submission, relying instead on the predicate's prior approvals.

4. Adjudication Method for the Test Set

This information is not provided in the given 510(k) summary.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

A MRMC comparative effectiveness study is not mentioned in the provided 510(k) summary. The submission focuses on demonstrating substantial equivalence based on technical characteristics and minor modifications, not on a direct comparison of human reader performance with or without AI assistance.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

A standalone performance study of the algorithm is not discussed in the provided 510(k) summary. The device is a "Full-Field Digital Mammography X-ray System," implying it generates images for human interpretation, not an AI-only diagnostic tool.

7. Type of Ground Truth Used

The specific type of ground truth used for any "verification and validation activities" related to the L50's performance is not explicitly stated in this document. Given that it's an imaging device for breast cancer screening and diagnosis, typical ground truth would involve pathology reports or long-term follow-up for cancer detection, often assessed by expert consensus on image interpretation. However, this is not detailed here.

8. Sample Size for the Training Set

The provided 510(k) summary describes a physical imaging device, not an AI algorithm that requires a training set. Therefore, this information is not applicable in the context of this specific submission.

9. How the Ground Truth for the Training Set Was Established

Since the device itself is not an AI algorithm with a training set, this information is not applicable.

Summary of Missing Information:

This 510(k) summary is for a hardware device (mammography system) that has undergone minor modifications compared to a cleared predicate. As such, it primarily focuses on demonstrating technical equivalence and that the changes do not introduce new safety or effectiveness concerns.

It does not provide details about:

• Specific quantitative performance acceptance criteria (e.g., image quality metrics, diagnostic accuracy).
• Clinical study designs (sample sizes, provenance, expert qualifications, adjudication methods).
• AI algorithm training or testing data (as it's not an AI diagnostic device).
• MRMC studies.

The "study" mentioned vaguely as "verification and validation activities" and "physical laboratory test results" is implied to support the claim of equivalence, but no specifics are provided in this regulatory document.

Ask a specific question about this device

The Philips MicroDose L30 is indicated for generating mammographic images that can be used for screening and diagnosis of breast cancer. The Philips MicroDose L30 is intended to be used in the same clinical applications as traditional film/screen systems.

The Philips MicroDose L30 is a full-field digital mammography system comprised of an image acquisition system, a gantry and an acquisition workstation computer equipped with a keyboard, a keypad, a mouse, and a monitor. The image acquisition system includes a digital detector of photon counting technology, x-ray tube (with tungsten target and aluminum filtration), high voltage generator, compression paddle(s), and multi-slit collimator. The acquisition is the user interface for preparing and initiating image acquisition, image processing, and image transfer to the desired destination (e.g. PACS) for diagnosis and archiving. The Philips MicroDose L30 detector is based on photon counting technology and consists of a large number of crystalline silicon strip detectors. The technology enables high detection efficiency of photons and efficient rejection of electronic noise. The Philips MicroDose 130 uses a multi-slit scanning technique that prevents image degradation caused by scattered radiation by removing photons scattered in the breast and not directed towards the detector. These factors combine into a dose efficient system. The Philips MicroDose 130 provides three exposure modes; manual, automatic (parameters predefined based on compressed breast thickness), and SmartAEC continuously adjusts the exposure based on measured transmission from the leading detector edge.

The provided document (K120255) describes a 510(k) premarket notification for the Philips MicroDose L30 Full-Field Digital Mammography X-ray System. This submission focuses on obtaining clearance for a quantitative reduced dose claim for the MicroDose system based on supporting clinical data, as the device itself (including its indications for use and core technology) is substantially equivalent to a previously cleared predicate device (Sectra MicroDose Mammography L30, K110025).

Therefore, the document does not contain details about acceptance criteria or a study proving the device meets those criteria in the context of typical diagnostic performance (e.g., sensitivity, specificity, AUC for cancer detection). Instead, it refers to the original 510(k) (K110025) for performance data and highlights that the current submission is about the reduced dose claim.

However, based on the information provided, we can infer some aspects and acknowledge what is missing.

Here's an analysis based on the provided text, addressing the points where information is available or can be inferred, and explicitly stating when information is not provided:

1. A table of acceptance criteria and the reported device performance

   • Not explicitly provided in the context of diagnostic accuracy for cancer detection for this specific submission (K120255).
   • The document states that the original 510(k) (K110025) for the Sectra MicroDose Mammography L30 did contain performance data from non-clinical testing. This included:
     • Sensitometric response
     • Spatial resolution
     • Noise analysis
     • Signal-to-noise-ratio transfer - DQE
     • Dynamic range
     • Automatic exposure control performance
     • Phantom testing
     • Patient radiation dose
   • The predicate device demonstrated that it "performed as well as or better than the predicate devices in all relevant areas" for these technical parameters.
   • For the reduced dose claim in K120255, the acceptance criteria would implicitly be related to demonstrating a reduction in patient radiation dose while maintaining image quality sufficient for the stated indications. The document mentions "consistent levels of dose reduction with the use of the MicroDose system compared to other FFDM and film screen systems" based on published studies. Specific numerical acceptance criteria for this dose reduction are not provided.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

   • For the original device performance (K110025) referred to, the document mentions "clinical evidence including published studies on various national mammography screening programs." This suggests prospective or retrospective data from real-world screening programs, but specific sample sizes, countries of origin, or detailed study designs are not provided in this document.
   • For the reduced dose claim in K120255, no specific test set or study details like sample size or provenance are provided, beyond the general reference to "supporting clinical data" and "published studies."

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

   • Not provided. This information would typically be detailed in a clinical study report, which is not part of this summary.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

   • Not provided.

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 provided. This device is a digital mammography X-ray system, not an AI software for interpretation assistance. Therefore, an MRMC study comparing human readers with and without AI assistance is generally not applicable to this type of device itself. The clinical evidence referenced concerned the performance of the system for dose reduction and image quality (in K110025).

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

   • Not applicable/Not provided. This is a hardware device (X-ray system), not an algorithm for image interpretation. Its performance is evaluated on its ability to produce images, which are then interpreted by human readers.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

   • Not provided for the referenced "published studies" or previous clearance. For mammography screening, ground truth often involves a combination of imaging follow-up, biopsy/pathology results, and expert consensus (e.g., radiologists' final assessments over time).

8. The sample size for the training set

   • Not applicable/Not provided. This refers to a hardware system, not a machine learning algorithm that requires a training set. The system itself is "trained" through engineering and calibration processes, not data sets in the AI sense.

9. How the ground truth for the training set was established

   • Not applicable/Not provided. (See point 8).

Summary of what's described in the document:

The Philips MicroDose L30 obtained clearance through a 510(k) submission, asserting substantial equivalence to the Sectra MicroDose Mammography L30 (K110025). The current submission (K120255) specifically sought clearance for a quantitative reduced dose claim.

• Acceptance Criteria & Performance: For the initial device (K110025), non-clinical test data (sensitometric response, spatial resolution, noise, DQE, dynamic range, AEC performance, phantom testing, patient radiation dose) showed performance "as well as or better than the predicate devices." For the reduced dose claim, the device demonstrated "consistent levels of dose reduction" in comparison to other FFDM and film-screen systems, as supported by "published studies on various national mammography screening programs." The specific numerical acceptance criteria for this dose reduction are not detailed in this summary.
• Study Type: The evidence for the reduced dose claim comes from "supporting clinical data" and "published studies," indicating these were likely clinical evaluations focused on demonstrating dose reduction while maintaining diagnostic image quality. Details on these studies (e.g., sample size, provenance, ground truth, expert involvement) are not provided in this 510(k) summary.
• The device is a Full-Field Digital Mammography X-ray System, not an AI-powered diagnostic tool, so many questions regarding AI-specific evaluations (MRMC, standalone algorithm, training sets) are not applicable.

Ask a specific question about this device