SIEMENS MAMMOMAT NOVATIONDR FULL FIELD DIGITAL MAMMOGRAPHY SYSTEM
Device Facts
| Record ID | P030010 |
|---|---|
| Device Name | SIEMENS MAMMOMAT NOVATIONDR FULL FIELD DIGITAL MAMMOGRAPHY SYSTEM |
| Applicant | Siemens Medical Solution |
| Product Code | MUE · Radiology |
| Decision Date | Aug 20, 2004 |
| Decision | APRL |
| Regulation | 21 CFR 892.1715 |
| Device Class | Class 2 |
Indications for Use
The Siemens Mammomat Novation™ Full Field digital Mammography System (Novation™) generates digital mammographic images that can be used for screening and diagnosis of breast cancer and is intended for use in the same clinical applications as traditional film-based mammographic systems. Mammographic images can be interpreted by either hardcopy film or by softcopy at a workstation.
Device Story
Novation™ is a full-field digital mammography system; utilizes Siemens Mammomat 3000 Nova x-ray stand for breast compression and x-ray production; incorporates Hologic Selenia™ amorphous selenium (a-Se) digital image receptor for direct conversion of x-ray photons to digital data. System includes acquisition workstation for patient data entry, image processing, and preview; images forwarded to MammoReportPlus for softcopy review or hardcopy printing. Operated by technologists in clinical environments; interpreted by physicians. Features Automatic Exposure Control (AEC) and multiple anode/filter combinations (Mo/Mo, Mo/Rh, W/Rh) to optimize image quality and dose. Device provides digital images for breast cancer screening and diagnosis, enabling clinical decision-making equivalent to traditional film-screen mammography.
Clinical Evidence
No new clinical trials were conducted. Equivalence was established via non-clinical bench testing and blinded reads of mammography phantoms (RMI 156 and CDMAM 3.4) by three independent physicists. Bench testing compared MTF, DQE, SNR, and sensitivity against the Hologic Selenia™ system. Results showed performance equivalence across all metrics. FDA permitted reliance on clinical data from Hologic PMA P010025 based on demonstrated technical equivalence.
Technological Characteristics
Full-field digital mammography system; amorphous selenium (a-Se) direct-conversion detector (24x29 cm active area). Anode/filter combinations: Mo/Mo, Mo/Rh, W/Rh. Features Automatic Exposure Control (AEC). Connectivity: acquisition workstation, softcopy display (MammoReportPlus), and hardcopy printer. Software includes image capture, processing, and display algorithms. Compliant with MQSA requirements.
Indications for Use
Indicated for screening and diagnosis of breast cancer in patients requiring mammographic imaging. Intended for use in clinical settings by trained personnel (physicians, technologists, medical physicists) in MQSA-certified facilities.
Regulatory Classification
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.
Special Controls
*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.
Predicate Devices
Submission Summary (Full Text)
{0}
# SUMMARY OF SAFETY AND EFFECTIVENESS DATA
## I. GENERAL INFORMATION
Device Generic Name: Full Field Digital Mammography System
Device Trade Name: Siemens Mammomat *Novation*™
Applicant: Siemens Medical Solutions USA, Inc.
51 Valley Stream Parkway
Malvern, PA 19355
PMA Number: P030010
Date of Panel Recommendation: None
Date of Good Manufacturing Inspection: June 1-4, 2004
Date of Notice of Approval to Applicant: TBD
## II. INDICATIONS FOR USE
The Siemens Mammomat *Novation*™ Full Field digital Mammography System (*Novation*™) generates digital mammographic images that can be used for screening and diagnosis of breast cancer and is intended for use in the same clinical applications as traditional film-based mammographic systems. Mammographic images can be interpreted by either hardcopy film or by softcopy at a workstation.
## III. DEVICE DESCRIPTION
The *Novation*™ consists of an image acquisition system, hardcopy display, and softcopy workstation. It utilizes the Siemens Mammomat 3000 Nova mammography x-ray system (K932672) for the production of x-rays, for supporting compression of the breast, and for the physical support of the digital image receptor licensed from the Hologic Lorad Selenia™ (Selenia™) system (P010025). It also uses the same image capture algorithms, image processing algorithms, and image display algorithms for softcopy display and hardcopy printouts. The *Novation*™ maintains the same features as the Siemens Mammomat 3000 Nova including the Automatic Exposure Control (AEC) system and anode/filter combinations of Molybdenum/Molybdenum (Mo/Mo), Molybdenum/Rhodium (Mo/Rh), and Tungsten/Rhodium (W/Rh).
PMA P030010
{1}
The Siemens Mammomat 3000 Nova x-ray stand holds the Hologic amorphous selenium (a-Se) digital image receptor with an active area of $24 \times 29 \, \text{cm}$ which directly converts incoming x-ray photons to digital image data. At the acquisition workstation, the user enters the patient identification data (or receives it from a work list), acquires, processes, and displays the images for image preview. These images are then forwarded either for hardcopy printing or softcopy display to the MammoReportPlus for review and diagnosis.
Users must ensure that they have completed the Siemens Novation$^{\text{DR}}$ training program prior to conducting patient exams. The Siemens training program will address the personnel training requirements under MQSA regulations in product labeling to assure that the prospective users are aware of the required eight hours of training for any medical physicist, technologist, or interpreting physician.
## IV. CONTRAINDICATIONS
None known.
## V. WARNINGS AND PRECAUTIONS
The warnings and precautions can be found in the Mammomat Novation$^{\text{DR}}$ labeling.
## VI. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH
Potential adverse effects of mammography include:
- Excessive breast compression
- Excessive x-ray exposure
- Electric shock
- Infection and skin irritation
- Abrasion or puncture wound
## VII. ALTERNATIVE PRACTICES AND PROCEDURES
Various methods are available for screening or diagnosing of breast cancer. These include a clinical breast examination, screen-film mammography, ultrasound, and magnetic resonance imaging. Biopsy of a detected abnormality is often obtained to diagnose or rule out cancer.
## VIII. MARKETING HISTORY
The system components (X-ray tube, support assembly, and compression paddle) of the Mammomat 3000 Nova mammography x-ray system (K932672) have been marketed since 1993 and have never been withdrawn for any reason related to safety or effectiveness. Since
PMA P030010
{2}
March 2004, the Novation$^{\text{DR}}$ has been marketed in the European Union and to date the Novation$^{\text{DR}}$ system has never been withdrawn either from marketing for any reason related to safety or effectiveness.
## IX. SUMMARY OF NON-CLINICAL STUDIES
Siemens performed a series of quantitative measurements meant to characterize the physics aspects of the licensed digital receptor and image display as implemented in the Novation$^{\text{DR}}$ system. It was pivotal to establish the technical equivalence of the Novation$^{\text{DR}}$ and the Selenia$^{\text{TM}}$ systems in order to legitimize the use of Hologic’s clinical data. Technical testing encompassed characteristic curves, automatic exposure control (AEC), modulation transfer function (MTF), detector quantum efficiency (DQE), signal-to-noise ratio (SNR), noise power spectrum (NPS), sensitometric response, and phantom scoring.
## 1. Characteristic Curves
A characteristic curve is a plot of the image pixel intensity measured in analog-to-digital units (ADUs) versus the x-ray exposure level at the image receptor cover. Pixel intensity is the digital value measured in ADUs ranging from 0 to $2^{14}$.
The Novation$^{\text{DR}}$ detector has a linear response with a linearity of $>0.999$ in a specified range. Figure 1.1 illustrates the characteristic curves of the Novation$^{\text{DR}}$ image receptor.
Figure 1.1 Characteristic curves for the Novation$^{\text{DR}}$ image receptor.
PMA P030010
{3}
# 2. Automatic Exposure Control (AEC)
Automatic exposure control (AEC) is a technology that is widely used in standard x-ray imaging and digital imaging systems. The objective of an AEC system is to optimize image quality while minimizing patient dose in an effort to produce consistent radiology images from patient to patient regardless of size or presence of pathology.
Approximately 1000 images were generated for the standard and magnification modes using the Novation$^{\text{DR}}$ AEC system. In order to document the limits of the AEC performance, data was gathered for a range of phantom thicknesses simulating the population at large and consequently exercising all the anode/filter combinations (Mo/Mo, Mo/Rh and W/Rh).
Additional testing was performed to compare image quality between images obtained in the manual mode and those generated in the AEC mode, and to compare doses delivered by the film-screen system (Nova 3000) with doses delivered by the digital system (Novation$^{\text{DR}}$).
## 2.1 Automatic Exposure Control (AEC)-Enabled Novation$^{\text{DR}}$
The performance of the AEC using the Novation$^{\text{DR}}$ was evaluated by acquiring data sets of varying kVs and varying breast equivalent thicknesses. AEC performance was measured in both standard and magnification imaging modes.
Figure 2.1 represents the digital signal fluctuation for each technique. The plot displays the $\pm 15\%$ from the average of each technique (solid line) and the measured standard deviation from the mean (dotted line). All data are within the MQSA requirement of $\pm 15\%$.
Figure 2.1 Digital signal fluctuation for each technique (target/filter combination) in standard mode.
## 2.2 Comparison of AEC and Manual Mode Between Siemens Film-Screen (Nova 3000) and Digital (Novation$^{\text{DR}}$) Systems
PMA P030010
{4}
Testing was performed to compare the digital image quality generated by the manual mode to the AEC mode and the dose delivered by the film-screen system to the dose delivered by the digital system.
## 2.2.1 Manual Mode vs. AEC
A total of 90 images were acquired on the Novation$^{\text{DR}}$ system using the target-filter combinations of Mo/Mo, Mo/Rh, and W/Rh.
Exposures were initially made in the AEC mode which were then followed by the manual mode. For the manual mode techniques, dose was chosen from the selected values of the AEC mode minus the 5 mAs scout pulse. The 5 mAs scout pulse was subtracted since this pre-exposure dose does not contribute to image quality. Percent deviation between the manual mode and the AEC modes for CNR and SNR was calculated for each combination of phantom thickness, kVp, and filtration in H mode (low dose) and D mode (high dose).
The results indicate that the percent deviations between SNR and CNR when comparing the AEC and Manual Modes of the Novation$^{\text{DR}}$ for H and D mode were within an acceptance criteria of +/- 10%.
## 2.2.2 Dose Comparison for AEC Mode Between the Nova 3000 and the Novation$^{\text{DR}}$
Measurements were done to compare the dose delivered from the analog system (Nova 3000) to the dose delivered from the digital system (Novation$^{\text{DR}}$). Figure 2.2 shows the dose levels for the Nova 3000 and the Novation$^{\text{DR}}$ using the AEC mode. The dose delivered by the Novation$^{\text{DR}}$ (H mode – low dose) is approximately the same or slightly lower than the dose delivered by the Nova 3000.
Figure 2.2 Dose comparison between Nova 3000 and Novation$^{\text{DR}}$ Using AEC Mode
PMA P030010
{5}
# 3. Modulation Transfer Function
Image sharpness of an imaging receptor is usually quantified by its modulation transfer function (MTF). The pre-sampling MTF was acquired using a $10\mu \mathrm{m}$ slit oriented at about a $1^{\circ}$ angle to the sampling grid of the detector with the anti-scatter grid removed. Figure 3.1 shows the MTF curves for the Selenia™ and NovationDR. Both systems have very similar MTF characteristics and demonstrate equivalence over the range 0 to $7\mathrm{lp / mm}$.
Figure 3.1 Novation$^{\text{DR}}$ and Selenia (MTF) Modular Transfer Function
# 4. Detective Quantum Efficiency (DQE)
DQE characterizes the efficiency with which a receptor uses radiation. Image sharpness was characterized by measuring the image receptor modulation transfer function (MTF) and its limiting spatial resolution. The detective quantum efficiency (DQE) is computed using the following equation where $G$ is the image receptor gain, $X$ is the X-ray exposure (mR), MTF is the modulation transfer function, $\Phi$ is the incident X-ray fluence (photons/mm²/mR), and NPS is the noise power spectrum:
$$
DQE = \frac{G^{2} \bullet X \bullet MTF^{2}}{\Phi \bullet NPS}
$$
DQE images were acquired at $28\mathrm{kV}$ for four different exposure levels (3.5, 7.1, 14.9, & 29.2 mR) on both the Selenia™ and the NovationDR systems. The DQE for an exposure level of 7.1 mR which would be clinically relevant is shown in Figure 4.1. Both systems have very similar DQE characteristics and demonstrate equivalence.
PMA P030010
{6}
Figure 4.1 DQE at exposure of $7.1\,\mathrm{mR}$ for both the Selenia and Novation$^{\mathrm{DR}}$ systems.
## 5. Signal-to-Noise Ratio (SNR)
The signal-to-noise ratio (SNR) is the ratio of the signal level relative to the noise level (standard deviation) and can be expressed as:
$$
SNR = \frac{S}{\sigma}, \text{ where } S = \mathrm{ADU} \text{ Value; } \sigma = \text{Standard Deviation.}
$$
SNR images were acquired at varying techniques (kVs ranging from 25 to 32 and mAs ranging from 26 to 75) using different acrylic phantom thicknesses for the anode filter combinations of Mo/Mo and Mo/Rh. All exposures were normalized to the same ADU at the detector surface of the Novation$^{\mathrm{DR}}$ system. The entrance surface air kerma (ESAK) was measured for the Novation$^{\mathrm{DR}}$ system and then applied to the Selenia$^{\mathrm{TM}}$ system. With deviations less than $+/- 10\%$, both systems demonstrate equivalence as illustrated in Figure 5.1
PMA P030010
{7}
Figure 5.1 Comparison of SNR and percent deviation between the Selenia and Novation $^{\text{DR}}$ using the anode/filter combination Mo/Mo and Mo/Rh.
## 6. Noise Power Spectrum (NPS)
Noise power spectrum (NPS) is a characterization of the noise distribution over the spatial frequency which is important for assessing image degradation. The NPS was determined using the synthesized slit method (Dainty and Shaw 1974) and plotted as a function of frequency (lp/mm) for four exposure levels. Figure 6.1 shows the relative differences in NPS of the Novation $^{\text{DR}}$ and Selenia $^{\text{TM}}$ systems for an entrance exposure of $7.1 \, \text{mR}$. The relative differences in NPS for these systems are a reflection of the gain factor utilized by each system, and a more accurate comparison should be based on the DQE values.
Figure 6.1 Measured NPS of the Novation $^{\text{DR}}$ and Selenia image receptor for an entrance exposure of $7.1 \, \text{mR}$.
PMA P030010
14
{8}
# 7. System Sensitivity
Sensitivity is plotted as a function of the exposure level (mR). The formula used is:
$$
Sensitivity = \frac {Counts}{mR} \cdot \frac {S}{\sigma^{2}}
$$
where S is the ADU value, $\sigma$ the standard deviation and the counts/mR is the slope of the characteristic curve. Figure 7.1 compares the relative sensitivity for the Novation$^{\text{DR}}$ and the Selenia$^{\text{TM}}$ systems. The maximum deviation of these two systems for different exposure levels is 9%.
Figure 7.1 Relative Sensitivity of the Novation$^{\text{DR}}$ (+) and Selenia (*) systems.
# 8. Image Ghosting
Image ghosting is a phenomenon whereby residual images are visible in subsequent images as a result of trapped bulk charges. The Novation$^{\text{DR}}$ image receptor utilizes an erasing and conditioning process to eliminate or reduce the trapped bulk charges sufficiently so image ghosting is not a problem. To quantify the effectiveness of the erasing and conditioning process two test cases were set up using a 50/50 (glandular/adipose tissue) phantom of tissue-equivalent (TE) breast material in such a way that a first exposure delivered a normal dose to the area under the breast phantom while the area adjacent to the breast phantom received an X-ray flux which saturated the detector. The specifics of the two test cases are described below.
In Test Case 1, a 6 cm thick 50/50 tissue equivalent phantom was positioned on the Novation$^{\text{DR}}$ image receptor and exposed at 30kV, 80mAs using the Mo/Mo anode/filter combination. This technique produced an exposure level under the breast of approximately
PMA P030010
{9}
70μGy which is a typical x-ray exposure level for a screen film system using a Kodak Min-R 2000 film. The 6 cm phantom was subsequently replaced by a 2 cm thick 50/50 tissue equivalent phantom that was smaller in breadth and width. Another image was taken at 28 kV, 7.1mAs and Mo/Mo anode/filter combination 1 minute after the first image. This technique also produced an exposure level under the breast of approximately 70μGy.
In Test Case 2, a lead bar was positioned on the *Novation* DR image receptor and exposed at 23kV, 2 mAs for the Mo/Mo anode/filter combination. This technique produced an exposure level under the lead bar corresponding to a zero exposure level. The lead bar phantom was replaced with a 4 cm thick 50/50 tissue equivalent phantom which was sufficiently large that it overlapped the lead bar. A second exposure was made after 3 minutes at 28 kV, 32 mAs and Mo/Mo anode/filter combination. This technique produced an exposure level under the breast of approximately 70μGy.
In the saturated area of the detector many electron-hole pairs are generated and transported in the applied field although some of the electrons can be trapped. During subsequent image acquisition some of the trapped charges that would normally be collected by the charge collection electrodes recombine with trapped electrons/holes and do not contribute to the signal charge. This especially occurs with low exposure X-ray images.
For both test cases signal levels inside the smaller of the phantoms and in the overlap region were measured in digital counts over a 128x128 region of interest. For test case 1 the saturation area received about 3 counts lower than the area directly under the TE breast phantom. This represents a 1.5% difference that will not affect the quality of a clinical image. Similarly, for test case 2 the area under the lead bar receiving a zero exposure level had about 4 counts lower than the saturation area beyond. This represents a 2.9% difference which would not degrade the quality of a clinical image although this test case does not represent normal clinical conditions.
## 9. Phantom Imaging and Blinded Read
## 9.1 Study Procedures
Two mammography phantoms, the RMI 156 and the CDMAM 3.4, were used in this non-clinical study to acquire images according to the manufacturer’s recommendation. The RMI 156 phantom consists of 6 different size nylon fibers which simulate fibrous structures, 5 groups of simulated microcalcifications, and 5 different sized tumor-like masses.
The CDMAM 3.4 consists of an aluminum base containing gold discs of various thicknesses and diameters which are arranged in a matrix of 16 rows and 16 columns. Each square contains two identical discs (same diameter and thickness), one in the center and one in a randomly chosen corner.
PMA P030010
10
{10}
## 9.2 Image Acquisition
A range of sample imaging configurations was used for this study based on varying phantom thicknesses, kV settings, and target-filter combinations. These were selected based on the settings where the device would be expected to be used.
## 9.3 Blinded Read
Three (3) independent physicists experienced in scoring mammography phantoms evaluated the phantom images. The readers were not aware of any information regarding the protocol-specific image parameters or the type of system on which the images were acquired. Images were viewed as both hardcopy and softcopy display and the three readers, with normal or corrected vision, scored the images independently. There were no restrictions on viewing distance or the use of magnification while reading the image.
## 9.4 Overall Results for the RMI 156 Phantom
The differences in blinded read results between Selenia™ and the Novation™ were evaluated through confidence bounds and statistical analyses. Figure 9.1 presents the mean scores by each feature type and device, and Table 1 contains a summary of the overall scores by device and display mode.
Figure 9.1 Mean of fibers, masses and specks by device.
PMA P030010
{11}
PMA P030010
| Table 9.1 Summary of Overall Results for the RMI 156 Phantom by Device and Display Mode | | | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| Overall | Novation^{DR} Hardcopy | Novation^{DR} Softcopy | Selenia Hardcopy | | | |
| | Mean | Std | Mean | Std | Mean | Std |
| | 12.81 | 1.03 | 12.97 | 0.96 | 13.06 | 0.73 |
## 9.5 Overall Results for the CDMAM 3.4 Phantom
For each diameter size of gold disc in the phantom, the minimally visualized disc thickness was also determined to obtain a response curve. Figure 9.2 shows the contrast-detail curve.
Figure 9.2 Contrast detail curve of minimum thickness by object diameter.
The total percent correct indicates the number of correctly identified cells noted on the phantom image. Table 9.2 lists the mean scores of all 3 readers by device.
| Table 9.2 Summary of Correct Cells Identified on CDMAM Phantom by Device | | | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| Overall | Novation^{DR} Hardcopy | Novation^{DR} Softcopy | Selenia Hardcopy | | | |
| | Mean | Std | Mean | Std | Mean | Std |
| | 43.48 | 7.60 | 43.90 | 8.80 | 44.93 | 8.02 |
## CONCLUSION:
This study has demonstrated that the imaging performance of the Novation$^{\text{DR}}$ is equivalent to that of the Selenia$^{\text{TM}}$ system. The inclusion of the CDMAM phantom improved the
12
{12}
evaluation of both systems and added a second level of confirmation to the study. Overall imaging performance of both systems was affected by phantom thickness in a similar fashion.
Equivalence was supported for all mean comparisons across both phantoms, and the results of this study provide good support for the use of the Novation$^{\text{DR}}$ system in both hardcopy and softcopy display. With the establishment of technical equivalence in terms of imaging performance between the Novation$^{\text{DR}}$ and Selenia$^{\text{TM}}$ systems through the generation of data from objective testing and the blinded read of hardcopy and softcopy phantom images, utilization of the clinical data obtained by Hologic is valid for the Novation$^{\text{DR}}$ system.
## X. SUMMARY OF CLINICAL STUDIES
The Mammomat Novation$^{\text{DR}}$ is based on the Siemens Mammomat 3000 Nova mammography x-ray system cleared in Pre-Market Notification, K932672. It uses the same image acquisition subsystem and image display subsystem as the Hologic Selenia$^{\text{TM}}$ system for mammography screening and diagnosis approved in Pre-Market Application, P010025. The conclusions based on data from the objective testing and blinded reads of phantom images in Section IX Summary of Nonclinical Studies demonstrates that the imaging performance characteristics of these systems are sufficiently equivalent to justify the use of data collected during a clinical study conducted by Hologic for PMA P010025. FDA and Hologic agreed that Siemens could reference the clinical studies performed by Hologic, Inc. (PMA Number P010025 and PMA supplement P010025/S1).
## XI. CONCLUSIONS DRAWN FROM NONCLINICAL AND CLINICAL STUDIES
The results of the nonclinical study conducted by Siemens and described above demonstrate equivalence between the Novation$^{\text{DR}}$ and the Selenia$^{\text{TM}}$ digital mammography systems and provide reasonable assurance of the safety and effectiveness of the Novation$^{\text{DR}}$ for screening and diagnostic breast imaging.
## XII. PANEL RECOMMENDATIONS
In accordance with the provisions of section 515(c)(2) of the act as amended by the Safe Medical Devices Act of 1990, this PMA was not referred to the Radiological Devices Advisory Panel, an FDA advisory committee, for review and recommendation because the information in the PMA substantially duplicates information previously reviewed by this panel.
## XIII. REGULATORY REQUIREMENTS
In compliance with the Mammography Quality Standards Act of 1992 (MQSA), Siemens has developed a comprehensive quality control program which will be the responsibility of the
PMA P030010
{13}
user to conduct and review once acceptance testing has been completed. Use of the Novation$^{\text{DR}}$ will only be permissible in MQSA certified facilities.
Users must ensure that they have completed the Siemens Novation$^{\text{DR}}$ training program prior to conducting patient exams. The Siemens training program will address the personnel training requirements under MQSA regulations in product labeling to assure that the prospective users are aware of the required eight hours of training for any medical physicist, technologist, or interpreting physician.
## XIV. FDA DECISION
The applicant’s manufacturing was inspected during June 1 to June 4, 2004 and was not found to be completely in compliance with the Quality Systems Regulations. FDA issued an approval on XXXXXX XX, 2004.
## XIV. APPROVAL SPECIFICATIONS
Directions for Use: See the attached labeling.
Hazards to Health from Use of the Device: See Warnings and Precautions in Essential Prescribing Information (Tab 2), and Potential Adverse Effects of the Device on Health in Section VI.
Post-Approval Requirements and Restrictions: see Approval order
PMA P030010
14
2c
