Breast Imaging

doi:10.1007/978-3-642-31271-7

Cited by 30 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A test object of 45 mm × 25 mm × 120 mm, made of Perspex, was used to assess the CNR in the stitching blocks. The object was placed above the detector with the 45 mm side parallel to the beam axis, in order to deliver a breastequivalent thickness of 53 mm (Dance et al 2012). Since the object only covers 20% of the detector area (its projection on the detector area is 25 mm × 120 mm), it was stepped across the detector reticle and images acquired at each step.…”

Section: Contrast-to-noise Ratiomentioning

confidence: 99%

Performance of a novel wafer scale CMOS active pixel sensor for bio-medical imaging

Esposito

Anaxagoras²,

Konstantinidis

et al. 2014

Phys. Med. Biol.

View full text Add to dashboard Cite

Abstract:Recently CMOS Active Pixels Sensors (APSs) have become a valuable alternative to amorphous Silicon and Selenium Flat Panel Imagers (FPIs) in bio-medical imaging applications. CMOS APSs can now be scaled up to the standard 20 cm diameter wafer size by means of a reticle stitching block process. However despite wafer scale CMOS APS being monolithic, sources of non-uniformity of response and regional variations can persist representing a significant challenge for wafer scale sensor response. Nonuniformity of stitched sensors can arise from a number of factors related to the manufacturing process, including variation of amplification, variation between readout components, wafer defects and process variations across the wafer due to manufacturing processes. This paper reports on an investigation into the spatial non-uniformity and regional variations of a wafer scale stitched CMOS APS. For the first time a per-pixel analysis of the electro-optical performance of a wafer CMOS APS is presented, to address inhomogeneity issues arising from the stitching techniques used to manufacture wafer scale sensors. A complete model of the signal generation in the pixel array has been provided and proved capable of accounting for noise and gain variations across the pixel array. This novel analysis leads to readout noise and conversion gain being evaluated at pixel level, stitching block level and in regions of interest, resulting in a coefficient of variation ≤ 1.9%. The uniformity of the image quality performance has been further investigated in a typical X-ray application, i.e. mammography, showing a uniformity in terms of CNR among the highest when compared with mammography detectors commonly used in clinical practise. Finally, in order to compare the CONFIDENTIAL -FOR REVIEW ONLY PMB-100279.R1detection capability of this novel APS with the currently used technology (i.e. FPIs), theoretical evaluation of the Detection Quantum Efficiency (DQE) at zero-frequency has been performed, resulting in a higher DQE for this detector compared to FPIs. Optical characterization, X-ray contrast measurements and theoretical DQE evaluation suggest that a trade off can be found between the need of a large imaging area and the requirement of a uniform imaging performance, making the DynAMITe large area CMOS APS suitable for a range of bio-medical applications. in bio-medical imaging applications. CMOS APSs can now be scaled up to the standard 20 cm diameter wafer size by means of a reticle stitching block process. However despite wafer scale CMOS APS being monolithic, sources of non-uniformity of response and regional variations can persist representing a significant challenge for wafer scale sensor response. Non-uniformity of stitched sensors can arise from a number of factors related to the manufacturing process, including variation of amplification, variation between readout components, wafer defects and process variations across the wafer due to manufacturing processes. This paper reports on an investigation into the spatial non-uni...

show abstract

Section: Contrast-to-noise Ratiomentioning

confidence: 99%

Performance of a novel wafer scale CMOS active pixel sensor for bio-medical imaging

Esposito

Anaxagoras²,

Konstantinidis

et al. 2014

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

“…6 A number of methods have been used to measure MD. These include visual area-based methods, for example, BI-RADS breast composition categories, 7 Boyd categories, 8 percent density recorded on visual analogue scales (VAS), 9 and semiautomated thresholding (Cumulus). 10 The automated Densitas software 11 operates in an area-based fashion on processed (for presentation) full-field digital mammograms (FFDM), while methods including Volpara 12 and Quantra 13 use raw (for processing) mammograms to estimate volumes of dense fibroglandular and fatty tissue in the breast.…”

Section: Introductionmentioning

confidence: 99%

Prediction of reader estimates of mammographic density using convolutional neural networks

et al. 2019

View full text Add to dashboard Cite

Mammographic density is an important risk factor for breast cancer. In recent research, percentage density assessed visually using visual analogue scales (VAS) showed stronger risk prediction than existing automated density measures, suggesting readers may recognize relevant image features not yet captured by hand-crafted algorithms. With deep learning, it may be possible to encapsulate this knowledge in an automatic method. We have built convolutional neural networks (CNN) to predict density VAS scores from full-field digital mammograms. The CNNs are trained using whole-image mammograms, each labeled with the average VAS score of two independent readers. Each CNN learns a mapping between mammographic appearance and VAS score so that at test time, they can predict VAS score for an unseen image. Networks were trained using 67,520 mammographic images from 16,968 women and for model selection we used a dataset of 73,128 images. Two case-control sets of contralateral mammograms of screen detected cancers and prior images of women with cancers detected subsequently, matched to controls on age, menopausal status, parity, HRT and BMI, were used for evaluating performance on breast cancer prediction. In the case-control sets, odd ratios of cancer in the highest versus lowest quintile of percentage density were 2.49 (95% CI: 1.59 to 3.96) for screendetected cancers and 4.16 (2.53 to 6.82) for priors, with matched concordance indices of 0.587 (0.542 to 0.627) and 0.616 (0.578 to 0.655), respectively. There was no significant difference between reader VAS and predicted VAS for the prior test set (likelihood ratio chi square, p ¼ 0.134). Our fully automated method shows promising results for cancer risk prediction and is comparable with human performance.

show abstract

“…Spatial resolution, as characterized by the modulation transfer function (MTF), is an important detector metric; and the literature is rich with published Monte Carlo detector optimizations which focus primarily on MTF. [1][2][3][4][5][6][7] However, optimization of the MTF fails to account for the contribution of noise to imaging performance. A better figure-of-merit is the frequency-dependent detective quantum efficiency [DQE(f)], which is the spectral representation in Fourier domain of the signal-to-noise characteristics of a given detector configuration.…”

Section: Introductionmentioning

confidence: 99%