This paper investigates whether two publicly available Artificial Intelligence (AI) models can detect retrospectively identified missed cancers within a double reader breast screening program and determine whether challenging mammographic cases are reflected in the performance of AI models. Transfer learning was conducted on the Globally-aware Multiple Instance Classifier (GMIC) and Global-Local Activation Maps (GLAM) models using an Australian mammographic dataset. Mammograms were enhanced to improve poor contrast using the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. The sensitivity of the two AI models with pre-trained and transfer learning modes was evaluated on four mammographic case groups: ‘missed’ cancers, ‘prior-visible’ cancers, ‘prior-invisible’ cancers and ‘current’ cancers from the archives of a double reader breast screening program. The GMIC model outperformed the GLAM model with pre-trained and transfer learning modes in terms of sensitivity for all four cancer groups. The performance of the GMIC and GLAM models was best in ‘prior-visible’ cancers, followed by ‘prior-invisible’ cancers, ‘current’ cancers and ‘missed’ cancers. The performance of the GMIC and GLAM models on the ‘missed’ cancer cases was 84.2% and 81.5%, respectively while for the ‘prior-visible’ cancer cases, the performance was 92.7% and 89.2%, respectively. After transfer learning, both the GMIC and GLAM models demonstrated statistically significant improvement (>9.4%) in terms of sensitivity for all cancer groups. The AI models with transfer learning showed significant improvement in malignancy detection in challenging mammographic cases. The study also supports the potential of the AI models to identify missed cancers within a double reader breast screening program.
Breast arterial calcifications (BAC) are increasingly recognized as indicative markers for cardiovascular disease (CVD). In this study, we manually annotated BAC areas on 3,330 mammograms, forming the foundational dataset for developing a deep learning model to automate assessment of BAC. Using this annotated data, we propose a semi-supervised deep learning approach to analyze unannotated mammography images, leveraging both labeled and unlabeled data to improve BAC segmentation accuracy. Our approach combines the U-net architecture, a well-established deep learning method for medical image segmentation, with a semi-supervised learning technique. We retrieved mammographic examinations of 6,000 women (3,000 with confirmed CVD and 3,000 without) from the screening archive to allow for a focused study. Utilizing our trained deep learning model, we accurately detected and measured the severity of BAC in these mammograms. Additionally, we examined the time between mammogram screenings and the occurrence of CVD events. Our study indicates that both the presence and severity (grade) of BAC, identified and measured using deep learning for automated segmentation, are crucial for primary CVD prevention. These findings underscore the value of technology in understanding the link between BAC in mammograms and cardiovascular disease, shaping future screening and prevention strategies for women's health.
Silicosis is a type of occupational lung disease or pneumoconiosis that results from the inhalation of crystalline silica dust that can lead to fatal respiratory conditions. This study aims to develop an online platform and benchmark radiologists' performance in diagnosing silicosis. Fifty readers (33 radiologists and 17 radiology trainees) interpreted a test-set of 15 HRCT cases. The median AUROC for all readers combined was 0.92 (0.93 for radiologists and 0.91 for trainees). No statistical differences were observed among the radiologists and trainees for their performance. Moderate agreement was recorded among readers for the correct diagnosis of silicosis (κ=0.57), however, there was considerable variability (κ<0.2) in the accurate detection of irregular opacities and ground glass opacities. Our online platform shows promise in providing tailored education to clinicians and facilitating future works of long-term observer studies and development of educational solutions to enhance the diagnostic accuracy of silicosis detection.
KEYWORDS: Mammography, Current controlled current source, Artificial intelligence, Education and training, Breast density, Cancer, Cancer detection, Breast cancer
This preliminary study investigates the magnitude of concordance, affecting factors and restrictions when radiologists' make annotations on mammographic images. Annotated data is key to the development of artificial intelligence (AI) tools and errors from annotations can reduce the accuracy of these tool. Two highly experienced radiologists (>20 years’ experience) provided annotations as rectangular regions of interest to mark the location of lesions when they read 856 mammographic images with known cancer signs. Mammographic images were resized to same resolution of 1664 × 768 pixels using bilinear interpolation. We calculated Lin’s concordance correlation coefficient (CCC) between the coordinates in x-axis and y-axis of the 4 corners of the overlapped annotations. The two overlapped annotations in different views (cranio-caudal (CC) and medio-lateral oblique (MLO)) were evaluated for agreement between radiologists. The values of Lin’s CCC were classified in four interpretation levels: the ‘almost perfect’, ‘substantial’, ‘moderate’ and ‘poor’ according to McBride's guide (2015). The results demonstrated ‘almost perfect’, ‘substantial’, ‘moderate’ and ‘poor’ concordance in 50.1%, 29.8%, 9.5% and 10.6% of the total overlapped annotations in the MLO view, with 93.1%, 5.6%, 0.3% and 1.0% of the total overlapped annotations in the CC view, respectively. Overall, the radiologists demonstrated stronger concordance when annotating the CC view compared to the MLO. Breast density (BD) also affected the concordance of the radiologists’ annotations with a decrease in the strength of concordance agreement between breast density classifications, from 0-50% BD = higher concordance to 50-100% BD = lower concordance. Our annotation investigation has implications for AI, where delineation of lesions is often the starting point for training data.
KEYWORDS: Digital breast tomosynthesis, Mammography, Education and training, Cancer, Breast cancer, Breast imaging, Diagnostics, Breast, Cancer detection, Radiology
The final stage in the medical imaging diagnostic system is the radiologist’s interpretation of the images, though research on the factors influencing performance in digital breast tomosynthesis (DBT) is inconclusive. This study seeks to understand the performance of radiologists in reading DBT images and the parameters impacting observer performance in three different countries. The study used a DBT mammogram test to compare the performance of radiologists from Australia, China and Iran in reading thirty-five DBT cases. A range of performance metrics including specificity, sensitivity, lesion sensitivity, ROC AUC and JAFROC FOM were generated for each radiologist upon the conclusion of the test set. The radiologists also provided demographic information relating to their experience in reading digital mammograms and DBT. Each country had a greater percentage of radiologists that have completed a breast imaging fellowship compared to those that have not. Australia had a greater percentage of radiologists that have completed training in DBT reading (Australia=88.2%), while China and Iran had a smaller percentage of radiologists that have not completed training in DBT reading (China=37%, Iran=40%). Significant differences were identified between the three countries in specificity (p=.001), lesion sensitivity (p=.016), ROC (p<.001) and JAFROC (p<.001). Australia had the highest mean value for all performance metrics, while China had the lowest mean value for all performance metrics. Australian radiologists have a moderate positive correlation between lesion sensitivity and the number of years reading DBT images (r=.513, p=.042). Iranian radiologists who read more than 20 DBT cases per week obtained significantly higher performance in lesion sensitivity 73.3% vs. 51.8%; p=.032) than the ones who read less than 20 DBT cases per week.
Purpose: This study aims at establishing the optimum x-ray energy for synchrotron acquired propagation-based computed tomography (PB-CT) images to obtain highest radiological image quality of breast mastectomy samples. It also examines the correlation between objective physical measures of image quality with subjective human observer scores to model factors impacting visual determinants of image quality. Approach: Thirty mastectomy samples were scanned at Australian Synchrotron’s Imaging and Medical Beamline. Samples were scanned at energies of 26, 28, 30, 32, 34, and 60 keV at a standard dose of 4mGy. Objective physical measures of image quality were assessed using signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), SNR/resolution (SNR/res), CNR/resolution (CNR/res) and visibility. Additional calculations for each measure were performed against reference absorption-based computer tomography (AB-CT) images scanned at 32 keV and 4mGy. This included differences in SNR (dSNR), CNR (dCNR), SNR/res (dSNR/res), CNR/res (dCNR/res), and visibility (dVis). Physical measures of image quality were also compared with visual grading analysis data to determine a correlation between observer scores and objective metrics. Results: For dSNR, dCNR, dSNR/res, dCNR/res, and dVis, a statistically significant difference was found between the energy levels. The peak x-ray energy for dSNR and dSNR/res was 60 keV. For dCNR and dCNR/res 34 keV produced the highest measure compared to 28 keV for dVis. Visibility and CNR correlate to 56.8% of observer scores. Conclusion: The optimal x-ray energy differs for different objective measures of image quality with 30-34 keV providing optimum image quality for breast PB-CT. Visibility and CNR correlate highest to medical imaging expert scores.
One of the imaging modalities offered by the Imaging and Medical Beamline (IMBL) at the Australian Synchrotron is Xray phase-contrast propagation-based computed tomography (PB-CT). The unique combination of high coherence and high brightness of radiation produced by synchrotron X-ray sources enables phase contrast imaging with excellent sensitivity to small density differences in soft tissues and tumors. The PB-CT images using spatially coherent radiation show high signal-to-noise ratio (SNR) without reducing the spatial resolution. This is due to the combined effect of forward free-space propagation and the advanced step of phase retrieval in the reconstruction processes that allows to accommodate noisier recorded images. This gives an advantage of potentially reducing the radiation dose delivered to the sample whilst preserving the reconstructed image quality. It is expected that the PB-CT technique will be well suited for diagnostic breast imaging in the near future with the advantage that it could provide better tumor detection and characterization/grading than mammography and other breast imaging modalities/techniques in general. The PB-CT technique is expected to reduce false negative and false positive cancer diagnoses that result from overlapping regions of tissue in 2D mammography and avoid patient pain and discomfort that results from breast compression. The present paper demonstrates that PB-CT produces superior results for imaging low-density materials such as breast mastectomy samples, when compared to the conventional absorption-based CT collected at the same radiation dose. The performance was quantified in terms of both the measured objective image characteristics and the subjective scores from radiological assessments. This work is part of the ongoing research project aimed at the introduction of 3D X-ray medical imaging at the IMBL as innovative tomographic methods to improve the detection and diagnosis of breast cancer. Major progress of this project includes the characterization of a large number of mastectomy samples, both normal and cancerous.
Purpose: Breast cancer is the most common cancer in women in developing and developed countries and is responsible for 15% of women’s cancer deaths worldwide. Conventional absorption-based breast imaging techniques lack sufficient contrast for comprehensive diagnosis. Propagation-based phase-contrast computed tomography (PB-CT) is a developing technique that exploits a more contrast-sensitive property of x-rays: x-ray refraction. X-ray absorption, refraction, and contrast-to-noise in the corresponding images depend on the x-ray energy used, for the same/fixed radiation dose. The aim of this paper is to explore the relationship between x-ray energy and radiological image quality in PB-CT imaging.
Approach: Thirty-nine mastectomy samples were scanned at the imaging and medical beamline at the Australian Synchrotron. Samples were scanned at various x-ray energies of 26, 28, 30, 32, 34, and 60 keV using a Hamamatsu Flat Panel detector at the same object-to-detector distance of 6 m and mean glandular dose of 4 mGy. A total of 132 image sets were produced for analysis. Seven observers rated PB-CT images against absorption-based CT (AB-CT) images of the same samples on a five-point scale. A visual grading characteristics (VGC) study was used to determine the difference in image quality.
Results: PB-CT images produced at 28, 30, 32, and 34 keV x-ray energies demonstrated statistically significant higher image quality than reference AB-CT images. The optimum x-ray energy, 30 keV, displayed the largest area under the curve ( AUCVGC ) of 0.754 (p = 0.009). This was followed by 32 keV (AUCVGC = 0.731, p ≤ 0.001), 34 keV (AUCVGC = 0.723, p ≤ 0.001), and 28 keV (AUCVGC = 0.654, p = 0.015).
Conclusions: An optimum energy range (around 30 keV) in the PB-CT technique allows for higher image quality at a dose comparable to conventional mammographic techniques. This results in improved radiological image quality compared with conventional techniques, which may ultimately lead to higher diagnostic efficacy and a reduction in breast cancer mortalities.
KEYWORDS: Digital breast tomosynthesis, Mammography, Breast cancer, Digital mammography, Cancer, Breast, Diagnostics, Tomography, New and emerging technologies, Medicine
The aim of this study is to evaluate the effect of adding digital breast tomosynthesis (DBT) to digital mammography (DM) on sensitivity and specificity scores for readers with different DM and DBT experience compared with that of DM alone. Ethical committee approval was obtained. 41 DM and DBT cases (22 cancer, 19 normal), each containing two views, were reviewed by 18 readers, divided into groups according to level of experience with DBT and DM. Readers were asked to report each case in two modes (DM and DM+DBT) using a 5-point scale (1- Normal, 2- Benign, 3- Equivocal, 4-Suspicious, 5- Malignant). The radiologists’ diagnostic performance was compared between DM and DM + DBT and evaluated by sensitivity and specificity. Readers with no DBT workshop showed higher sensitivity using DM+DBT compared with DM (P-value 0.03). Female readers, readers with less than 5 years of DM experience, readers with more than 20 mammography reads per week, readers who are not using DBT in clinical practice, readers with mammography fellowship, and readers who had a DBT workshop showed a significantly higher specificity using DM+DBT in comparison to DM alone (P-values 0.01, 0.01, 0.02, 0.03, 0.03, 0.03, 0.01 respectively). The current study showed that the addition of DBT to DM might not significantly change the readers performance in terms sensitivity, however it may result in less number of recalls to additional examinations which provides a substantial benefit in the screening and diagnostic settings.
Aim: In recent years Phase Contrast Tomography (PCT) has been rapidly progressing towards clinical translation as an advanced imaging technology for breast cancer diagnosis. Recent optimization of PCT with mastectomy samples has refined imaging protocols and biomedical-engineering prowess is now required to formalize patient table and breast immobilisation requirements. PCT imaging requires women to lie in prone position similar to conventional breast CT, however the imaging couch rotates above the beam allowing exposure of the breast beneath. Motion artefact through involuntary movement of the breast through the rotation cycle has the potential to reduce diagnostic quality of the results. Methods: This paper details the biomedical engineering cycle of breast holder development alongside medical physics considerations. Breast immobilisation via a plastic or silicone supporting material which is sufficiently transparent for X-rays in the targeted energy range is explained, including the two step process of considering single cup versus double cup solutions and how mild-suction to the breast can be implemented in order to maximum breast tissue visualization and assist with dose uniformity. Results: Considering patient comfort, breast positioning and implications upon attenuation and phase shift, a number of models were developed in Australia and Italy. Early prototypes are described here with some preliminary imaging. Considerable work is taking place over the next three months as models undergo imaging with mastectomy samples at the Imaging and Medical Beamline at the Australian Synchrotron and the ELETTRA Synchrotron Italy. Consumer representatives will be rating the immobilisation device for comfort prior to the start of clinical trials in 2020.
Propagation-based phase-contrast CT (PB-CT) is a novel imaging technique that visualises variations in both X-ray attenuation and refraction. This study aimed to compare the clinical image quality of breast PB-CT using synchrotron radiation with conventional absorption-based CT (AB-CT), at the same radiation dose. Seven breast mastectomy specimens were scanned and evaluated by a group of 14 radiologists and medical imaging experts who assessed the images based on seven radiological image quality criteria. Visual grading characteristics (VGC) were used to analyse the results and the area under the VGC curve was obtained to measure the differences between the two techniques. For six image quality criteria (overall quality, perceptible contrast, lesion sharpness, normal tissue interfaces, calcification visibility and image noise), PB-CT images were superior to AB-CT images of the same dose (AUCVGC: 0.704 to 0.914, P≤.05). For the seventh criteria (artefacts), PB-CT images were also rated better than AB-CT images (AUCVGC: 0.647) but the difference was not significant. The results of this study provide a solid basis for future experimental and clinical protocols of breast PB-CT.
This study explored the possibility of using the gist signal (radiologists’ first impression about a case) for improving the performance of two recently developed deep learning-based breast cancer detection tools. We investigated whether by combining the cancer class probability from the networks with the gist signal, higher performance in identifying malignant cases can be achieved. In total, we recruited 53 radiologists, who provided an abnormality score on a scale from 0 to 100 to unilateral mammograms following a 500-millisecond presentation of the image. Twenty cancer cases, 40 benign cases, and 20 normal were included. Two state-ofthe-art deep learning-based tools (M1 and M2) for breast cancer detection were adopted. The abnormality scores from the networks and the gist responses for each observer were fed into a support vector machine (SVM). The SVM was personalized for each radiologist and its performance was evaluated using leave-one-out cross-validation. We also considered the average reader; whose gist responses were the mean abnormality scores given by all 53 readers to each image. The mean and range of AUCs in the gist experiment were 0.643 and 0.492-0.794, respectively. The AUC values for M1 and M2 were 0.789 (0.632-0.892) and 0.814 (0.673-0.897), respectively. For the average reader, the AUC for gist, gist+M1, and gist+M2 were 0.760 (0.617-0.862), 0.847 (0.754-0.928), 0.897 (0.789-0.946). For 45 readers, the performance of at least one of the models improved after aggregating its output with the gist signal. The results showed that the gist signal has the potential to improve the performance of adopted deep learning-based tools.
Phase-contrast imaging of the breast is expected to deliver significantly improved image quality and diagnostic value at a reduced radiation dose compared to present-day 2D X-ray mammography, digital breast tomosynthesis (DBT) and computed tomography (CT) and become a viable method for early diagnosis of breast cancer in women. This paper builds upon the evaluation of a novel protocol to evaluate 3D mammographic phase contrast imaging for the detection of breast cancer undertaken with a purpose designed phantom and selected breast cancer specimens. Following evaluation, propagation-based phase contrast imaging was demonstrated to have high contrast to noise ratio alongside an important reduction in radiation dose. The challenge now is to shift the focus of research to real clinic solutions, with the worldfirst demonstration of X-ray in-line full field phase-contrast mammographic tomography (PCT) with cancer patients through an international collaboration of a multi-disciplinary team.
KEYWORDS: Social networks, Social network analysis, Mammography, Breast, Radiology, Imaging systems, Medical imaging, Breast cancer, Health sciences, Statistical analysis
Rationale and objectives: Observer performance has been widely studied through examining the characteristics of individuals. Applying a systems perspective, while understanding of the system’s output, requires a study of the interactions between observers. This research explains a mixed methods approach to applying a social network analysis (SNA), together with a more traditional approach of examining personal/ individual characteristics in understanding observer performance in mammography. Materials and Methods: Using social networks theories and measures in order to understand observer performance, we designed a social networks survey instrument for collecting personal and network data about observers involved in mammography performance studies. We present the results of a study by our group where 31 Australian breast radiologists originally reviewed 60 mammographic cases (comprising of 20 abnormal and 40 normal cases) and then completed an online questionnaire about their social networks and personal characteristics. A jackknife free response operating characteristic (JAFROC) method was used to measure performance of radiologists. JAFROC was tested against various personal and network measures to verify the theoretical model. Results: The results from this study suggest a strong association between social networks and observer performance for Australian radiologists. Network factors accounted for 48% of variance in observer performance, in comparison to 15.5% for the personal characteristics for this study group. Conclusion: This study suggest a strong new direction for research into improving observer performance. Future studies in observer performance should consider social networks’ influence as part of their research paradigm, with equal or greater vigour than traditional constructs of personal characteristics.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.