Patterns in historical climate data were analyzed for Ottawa, Ontario, Canada, for the interval 1890–2019. Variables analyzed included records of annual, seasonal, and extreme temperature and ...precipitation, diurnal temperature range, and various environmental responses. Using LOWESS regressions, it was found that annual and seasonal temperatures in Ottawa have generally increased through this interval, precipitation has shifted to a less snowy, rainier regime, and diurnal temperature variation has decreased. Furthermore, the annual growing season has lengthened by 23 days to ~163 days, and the annual number of frost-free days increased by 13 days to ~215 days. Despite these substantial climatic shifts, some variables (e.g., extreme weather events per year) have remained largely stable through the interval. Time-series analyses (including multitaper spectral analysis and continuous and cross wavelet transforms) have revealed the presence of several strong cyclical patterns in the instrumental record attributable to known natural climate phenomena. The strongest such influence on Ottawa’s climate has been the 11-year solar cycle, while the influence of the El Niño-Southern Oscillation, Arctic Oscillation, North Atlantic Oscillation, and Quasi-Biennial Oscillation were also observed and linked with the trends in annual, seasonal, and extreme weather. The results of this study, particularly the observed linkages between temperature and precipitation variables and cyclic climate drivers, will be of considerable use to policymakers for the planning, development, and maintenance of city infrastructure as Ottawa continues to rapidly grow under a warmer, wetter climate regime.
Breast density, or the amount of fibroglandular tissue (FGT) relative to the overall breast volume, increases the risk of developing breast cancer. Although previous studies have utilized deep ...learning to assess breast density, the limited public availability of data and quantitative tools hinders the development of better assessment tools. Our objective was to (1) create and share a large dataset of pixel-wise annotations according to well-defined criteria, and (2) develop, evaluate, and share an automated segmentation method for breast, FGT, and blood vessels using convolutional neural networks. We used the Duke Breast Cancer MRI dataset to randomly select 100 MRI studies and manually annotated the breast, FGT, and blood vessels for each study. Model performance was evaluated using the Dice similarity coefficient (DSC). The model achieved DSC values of 0.92 for breast, 0.86 for FGT, and 0.65 for blood vessels on the test set. The correlation between our model's predicted breast density and the manually generated masks was 0.95. The correlation between the predicted breast density and qualitative radiologist assessment was 0.75. Our automated models can accurately segment breast, FGT, and blood vessels using pre-contrast breast MRI data. The data and the models were made publicly available.
The purpose of this study was to assess the interobserver variability of users of the MRI lexicon in the fifth edition of the BI-RADS atlas.
Three breast imaging specialists reviewed 280 routine ...clinical breast MRI findings reported as BI-RADS category 3. Lesions reported as BI-RADS 3 were chosen because variability in the use of BI-RADS descriptors may influence which lesions are classified as probably benign. Each blinded reader reviewed every study and recorded breast features (background parenchymal enhancement) and lesion features (lesion morphology, mass shape, mass margin, mass internal enhancement, nonmass enhancement distribution, nonmass enhancement internal enhancement, enhancement kinetics) according to the fifth edition of the BI-RADS lexicon and provided a final BI-RADS assessment. Interobserver variability was calculated for each breast and lesion feature and for the final BI-RADS assessment.
Interobserver variability for background parenchymal enhancement was fair (ĸ = 0.28). There was moderate agreement on lesion morphology (ĸ = 0.53). For masses, there was substantial agreement on shape (ĸ = 0.72), margin (ĸ = 0.78), and internal enhancement (ĸ = 0.69). For nonmass enhancement, there was substantial agreement on distribution (ĸ = 0.69) and internal enhancement (ĸ = 0.62). There was slight agreement on lesion kinetics (ĸ = 0.19) and final BI-RADS assessment (ĸ = 0.11).
There is moderate to substantial agreement on most MRI BI-RADS lesion morphology descriptors, particularly mass and nonmass enhancement features, which are important predictors of malignancy. Considerable disagreement remains, however, among experienced readers whether to follow particular findings.
Breast cancer screening is among the most common radiological tasks, with more than 39 million examinations performed each year. While it has been among the most studied medical imaging applications ...of artificial intelligence, the development and evaluation of algorithms are hindered by the lack of well-annotated, large-scale publicly available data sets.
To curate, annotate, and make publicly available a large-scale data set of digital breast tomosynthesis (DBT) images to facilitate the development and evaluation of artificial intelligence algorithms for breast cancer screening; to develop a baseline deep learning model for breast cancer detection; and to test this model using the data set to serve as a baseline for future research.
In this diagnostic study, 16 802 DBT examinations with at least 1 reconstruction view available, performed between August 26, 2014, and January 29, 2018, were obtained from Duke Health System and analyzed. From the initial cohort, examinations were divided into 4 groups and split into training and test sets for the development and evaluation of a deep learning model. Images with foreign objects or spot compression views were excluded. Data analysis was conducted from January 2018 to October 2020.
Screening DBT.
The detection algorithm was evaluated with breast-based free-response receiver operating characteristic curve and sensitivity at 2 false positives per volume.
The curated data set contained 22 032 reconstructed DBT volumes that belonged to 5610 studies from 5060 patients with a mean (SD) age of 55 (11) years and 5059 (100.0%) women. This included 4 groups of studies: (1) 5129 (91.4%) normal studies; (2) 280 (5.0%) actionable studies, for which where additional imaging was needed but no biopsy was performed; (3) 112 (2.0%) benign biopsied studies; and (4) 89 studies (1.6%) with cancer. Our data set included masses and architectural distortions that were annotated by 2 experienced radiologists. Our deep learning model reached breast-based sensitivity of 65% (39 of 60; 95% CI, 56%-74%) at 2 false positives per DBT volume on a test set of 460 examinations from 418 patients.
The large, diverse, and curated data set presented in this study could facilitate the development and evaluation of artificial intelligence algorithms for breast cancer screening by providing data for training as well as a common set of cases for model validation. The performance of the model developed in this study showed that the task remains challenging; its performance could serve as a baseline for future model development.
Abstract Purpose The aim of this study was to better understand the relationship between digital breast tomosynthesis (DBT) difficulty and radiology trainee performance. Methods Twenty-seven ...radiology residents and fellows and three expert breast imagers reviewed 60 DBT studies consisting of unilateral craniocaudal and medial lateral oblique views. Trainees had no prior DBT experience. All readers provided difficulty ratings and final BI-RADS® scores. Expert breast imager consensus interpretations were used to determine the ground truth. Trainee sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated for low- and high-difficulty subsets of cases as assessed by each trainee him or herself (self-assessed difficulty) and consensus expert-assessed difficulty. Results For self-assessed difficulty, the trainee AUC was 0.696 for high-difficulty and 0.704 for low-difficulty cases ( P = .753). Trainee sensitivity was 0.776 for high-difficulty and 0.538 for low-difficulty cases ( P < .001). Trainee specificity was 0.558 for high-difficulty and 0.810 for low-difficulty cases ( P < .001). For expert-assessed difficulty, the trainee AUC was 0.645 for high-difficulty and 0.816 for low-difficulty cases ( P < .001). Trainee sensitivity was 0.612 for high-difficulty and .784 for low-difficulty cases ( P < .001). Trainee specificity was 0.654 for high-difficulty and 0.765 for low-difficulty cases ( P = .021). Conclusions Cases deemed difficult by experts were associated with decreases in trainee AUC, sensitivity, and specificity. In contrast, for self-assessed more difficult cases, the trainee AUC was unchanged because of increased sensitivity and compensatory decreased specificity. Educators should incorporate these findings when developing educational materials to teach interpretation of DBT.
Recent studies showed preliminary data on associations of MRI-based imaging phenotypes of breast tumours with breast cancer molecular, genomic, and related characteristics. In this study, we present ...a comprehensive analysis of this relationship.
We analysed a set of 922 patients with invasive breast cancer and pre-operative MRI. The MRIs were analysed by a computer algorithm to extract 529 features of the tumour and the surrounding tissue. Machine-learning-based models based on the imaging features were trained using a portion of the data (461 patients) to predict the following molecular, genomic, and proliferation characteristics: tumour surrogate molecular subtype, oestrogen receptor, progesterone receptor and human epidermal growth factor status, as well as a tumour proliferation marker (Ki-67). Trained models were evaluated on the set of the remaining 461 patients.
Multivariate models were predictive of Luminal A subtype with AUC = 0.697 (95% CI: 0.647-0.746, p < .0001), triple negative breast cancer with AUC = 0.654 (95% CI: 0.589-0.727, p < .0001), ER status with AUC = 0.649 (95% CI: 0.591-0.705, p < .001), and PR status with AUC = 0.622 (95% CI: 0.569-0.674, p < .0001). Associations between individual features and subtypes we also found.
There is a moderate association between tumour molecular biomarkers and algorithmically assessed imaging features.
Abstract Background The value of breast self-examination (BSE) to detect early breast cancer is controversial. Methods Within an institutional review board–approved prospective study, 147 high-risk ...women were enrolled from 2004 to 2007. Yearly clinical examination, BSE teaching, and mammography were performed simultaneously followed by interval breast magnetic resonance imaging (MRI). Women underwent additional BSE teaching at 6 months. Women reporting a mass on BSE underwent clinical evaluation. Results Fourteen breast cancers were detected in 12 women. BSE detected 6/14 breast cancers versus 6/14 detected by MRI and 2/14 by mammography. Of 24 masses detected by BSE, 6/24 were malignant. The sensitivity, specificity, and predictive value of BSE to detect breast cancer were 58.3%, 87.4%, and 29.2%, respectively. The sensitivity, specificity, and predictive value of a Breast Image Reporting and Data System (BI-RADS) score of ≥4 on MRI were 66.7%, 88.9%, and 34.8%, respectively. Conclusions BSE detects new breast cancers in high-risk women undergoing screening mammogram, CBE, and yearly breast MRI.
To compare the speed and accuracy of the interpretations of digital mammograms by radiologists by using printed-film versus soft-copy display.
After being trained in interpretation of digital ...mammograms, eight radiologists interpreted 63 digital mammograms, all with old studies for comparison. All studies were interpreted by all readers in soft-copy and printed-film display, with interpretations of images in the same cases at least 1 month apart. Mammograms were interpreted in cases that included six biopsy-proved cancers and 20 biopsy-proved benign lesions, 20 cases of probably benign findings in patients who underwent 6-month follow-up, and 17 cases without apparent findings. Area under the receiver operating characteristic curve (A(z)), sensitivity, and specificity were calculated for soft-copy and printed-film display.
There was no significant difference in the speed of interpretation, but interpretations with soft-copy display were slightly faster. The differences in A(z), sensitivity, and specificity were not significantly different; A(z) and sensitivity were slightly better for interpretations with printed film, and specificity was slightly better for interpretations with soft copy.
Interpretation with soft-copy display is likely to be useful with digital mammography and is unlikely to significantly change accuracy or speed.
Rare copy-number variants (CNVs) associated with neurodevelopmental disorders (NDDs), i.e., ND-CNVs, provide an insight into the neurobiology of NDDs and, potentially, a link between biology and ...clinical outcomes. However, ND-CNVs are characterised by incomplete penetrance resulting in heterogeneous carrier phenotypes, ranging from non-affected to multimorbid psychiatric, neurological, and physical phenotypes. Recent evidence indicates that other variants in the genome, or ‘other hits’, may partially explain the variable expressivity of ND-CNVs. These may be other rare variants or the aggregated effects of common variants that modify NDD risk. Here we discuss the recent findings, current questions, and future challenges relating to other hits research in the context of ND-CNVs and their potential for improved clinical diagnostics and therapeutics for ND-CNV carriers.
Several large-scale studies report additional variants that exacerbate NDD risk in patients carrying pathogenic ND-CNVs.GWAS-derived polygenic risk scores (PRSs) for cognitive and psychiatric phenotypes may influence clinical outcomes by increasing the risk for a given NDD in ND-CNV carriers.Exome sequencing studies of NDD patient cohorts demonstrated the presence of additional, rare, pathogenic single-nucleotide variants (SNVs) in the genomes of ND-CNV carriers.A variety of analytical approaches to integrate genomic variants, including burden testing, variance-component testing, and machine learning, can be applied to interrogate ND-CNV pathogenicity and variable phenotype expressivity in carriers.
Equity of access to higher education (HE) has been a priority for the Irish Government over the last fifteen years. Since 2005, the Higher Education Authority (HEA) in Ireland has introduced three ...successive national strategic plans for equity of access, which demonstrates its importance in HE policy. The aim of these initiatives is to improve equity of access, participation and success in HE for disadvantaged students. The findings generated from the third of these strategic plans, indicate that although some progress has been made to support this integration from further education (FE) to HE, challenges remain with an acknowledgement that there is a need to establish transparent supporting structures for building coherent pathways from FE to HE. Higher-level qualifications are now a common expectation among the general population and in industry, reflecting increased ambition, labour market demand for higher-level skills, and the need to continually upskill and/or reskill. The objective of this empirical investigation is to identify the perceived barriers preventing students of FE from progressing to HE in Ireland. Like others who have investigated this topic this investigation adopted an interpretivist approach to enquiry and a case study methodological approach. Primary data was generated from in depth focus groups and qualitative surveys. Data was analysed using thematic analysis. The preliminary findings of the first phase of this investigation illustrate that student transition from FE to HE is multidimensional. Findings have supported the development of a draft Transitions Framework, which is currently being piloted with case study students. The ultimate goal is to pursue the effective implementation of the Transitions Framework.
PDF
