Search

Purpose: To determine if pancreas radiomics-based AI model can detect the CT imaging signature of type 2 diabetes (T2D). Methods: Total 107 radiomic features were extracted from volumetrically segmented normal pancreas in 422 T2D patients and 456 age-matched controls. Dataset was randomly split into training (300 T2D, 300 control CTs) and test subsets (122 T2D, 156 control CTs). An XGBoost model trained on 10 features selected through top-K-based selection method and optimized through threefold cross-validation on training subset was evaluated on test subset. Results: Model correctly classified 73 (60%) T2D patients and 96 (62%) controls yielding F1-score, sensitivity, specificity, precision, and AUC of 0.57, 0.62, 0.61, 0.55, and 0.65, respectively. Model's performance was equivalent across gender, CT slice thicknesses, and CT vendors (p values > 0.05). There was no difference between correctly classified versus misclassified patients in the mean (range) T2D duration [4.5 (0–15.4) versus 4.8 (0–15.7) years, p = 0.8], antidiabetic treatment [insulin (22% versus 18%), oral antidiabetics (10% versus 18%), both (41% versus 39%) (p > 0.05)], and treatment duration [5.4 (0–15) versus 5 (0–13) years, p = 0.4]. Conclusion: Pancreas radiomics-based AI model can detect the imaging signature of T2D. Further refinement and validation are needed to evaluate its potential for opportunistic T2D detection on millions of CTs that are performed annually. [ABSTRACT FROM AUTHOR]

[Display omitted] • 189 patients relevant to BP were collected in our study, • the world's first non-traumatic plexopathy dataset was constructed, • 107 quantitative features of BP were extracted, • 54 machine learning models were trained and evaluated, • the identification performance achieves the accuracy of 0.912 ± 0.016. Automatic abnormality identification of brachial plexus (BP) from normal magnetic resonance imaging to localize and identify a neurologic injury in clinical practice (MRI) is still a novel topic in brachial plexopathy. This study developed and evaluated an approach to differentiate abnormal BP with artificial intelligence (AI) over three commonly used MRI sequences, i.e. T1, FLUID sensitive and post-gadolinium sequences. A BP dataset was collected by radiological experts and a semi-supervised artificial intelligence method was used to segment the BP (based on nnU-net). Hereafter, a radiomics method was utilized to extract 107 shape and texture features from these ROIs. From various machine learning methods, we selected six widely recognized classifiers for training our Brachial plexus (BP) models and assessing their efficacy. To optimize these models, we introduced a dynamic feature selection approach aimed at discarding redundant and less informative features. Our experimental findings demonstrated that, in the context of identifying abnormal BP cases, shape features displayed heightened sensitivity compared to texture features. Notably, both the Logistic classifier and Bagging classifier outperformed other methods in our study. These evaluations illuminated the exceptional performance of our model trained on FLUID-sensitive sequences, which notably exceeded the results of both T1 and post-gadolinium sequences. Crucially, our analysis highlighted that both its classification accuracies and AUC score (area under the curve of receiver operating characteristics) over FLUID-sensitive sequence exceeded 90%. This outcome served as a robust experimental validation, affirming the substantial potential and strong feasibility of integrating AI into clinical practice. [ABSTRACT FROM AUTHOR]

Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combining classifications from volunteers participating in a citizen science project with those from a convolutional neural network. The unique aspect of this work is the deployment, in combination, of both human and machine classifications for near real-time discovery in an astronomical project. We show that the combination of the two methods outperforms either one used individually. This result has important implications for the future development of transient searches, especially in the era of Large Synoptic Survey Telescope and other large-throughput surveys. [ABSTRACT FROM AUTHOR]

Issue addressed: To examine the factors associated with preventing regular smoking among Aboriginal adolescents. Methods: Cross‐sectional analysis of data from 106 Aboriginal adolescents aged 12‐17 years, and their caregivers, from four Aboriginal Community Controlled Health Services in urban New South Wales, 2008‐2012. The relation of individual, social, environmental and cultural factors to having 'never' smoked tobacco regularly was examined using Poisson regression. Results: Overall, 83% of adolescents had never smoked regularly; 13 reported current and five past smoking. Most lived in smoke‐free homes (60%) despite 75% reporting at least one current smoker caregiver. Participants were significantly more likely to have never smoked regularly if they had good mental health (PR = 1.4, 95% CI: 1.1‐1.9), their mother as their primary caregiver (1.3, 1.0‐1.6), good family relationships (1.2, 1.0‐1.5), stable housing (1.3, 1.1‐1.7), had never used alcohol (1.8, 1.3‐2.4), were not sexually active (3.1, 1.3‐ 7.2) and had no criminal justice interactions (1.8, 1.2‐2.8). Conclusions: Most participants lived in smoke‐free homes and the vast majority had never smoked regularly. Promoting good mental health and strengthening social connections may be protective against smoking as those experiencing less social disruption were more likely to have never smoked regularly. Smoking may be an indicator of psychosocial conditions and a prompt for screening and simultaneous treatment. So what?: Organisations should be resourced to deliver holistic adolescent health promotion programs. Programs and policies should support positive family relationships and stable housing as this may protect against the uptake of regular smoking. [ABSTRACT FROM AUTHOR]

Machine learning, deep learning, and artificial intelligence (AI) are terms that have made their way into nearly all areas of medicine. In the case of medical imaging, these methods have become the state of the art in nearly all areas from image reconstruction to image processing and automated analysis. In contrast to other areas, such as brain and breast imaging, the impacts of AI have not been as strongly felt in gynecologic imaging. In this review article, we: (i) provide a background of clinically relevant AI concepts, (ii) describe methods and approaches in computer vision, and (iii) highlight prior work related to image classification tasks utilizing AI approaches in gynecologic imaging. A comprehensive search of several databases from each database's inception to March 18th, 2021, English language, was conducted. The databases included Ovid MEDLINE(R) and Epub Ahead of Print, In-Process & Other Non-Indexed Citations, and Daily, Ovid EMBASE, Ovid Cochrane Central Register of Controlled Trials, and Ovid Cochrane Database of Systematic Reviews and ClinicalTrials.gov. We performed an extensive literature review with 61 articles curated by three reviewers and subsequent sorting by specialists using specific inclusion and exclusion criteria. We summarize the literature grouped by each of the three most common gynecologic malignancies: endometrial, cervical, and ovarian. For each, a brief introduction encapsulating the AI methods, imaging modalities, and clinical parameters in the selected articles is presented. We conclude with a discussion of current developments, trends and limitations, and suggest directions for future study. This review article should prove useful for collaborative teams performing research studies targeted at the incorporation of radiological imaging and AI methods into gynecological clinical practice. • A background of clinically relevant AI concepts and computer vision methods are presented in the context of gynecologic cancer imaging. • Prior work in the area combining AI, Radiology, and Gynecologic Oncology is given, providing a comprehensive resource for future studies. • The challenges to clinical practice integration for AI for the studies reviewed here are discussed. [ABSTRACT FROM AUTHOR]

Purpose: Total kidney volume (TKV) is the most important imaging biomarker for quantifying the severity of autosomal-dominant polycystic kidney disease (ADPKD). 3D ultrasound (US) can accurately measure kidney volume compared to 2D US; however, manual segmentation is tedious and requires expert annotators. We investigated a deep learning-based approach for automated segmentation of TKV from 3D US in ADPKD patients. Method: We used axially acquired 3D US-kidney images in 22 ADPKD patients where each patient and each kidney were scanned three times, resulting in 132 scans that were manually segmented. We trained a convolutional neural network to segment the whole kidney and measure TKV. All patients were subsequently imaged with MRI for measurement comparison. Results: Our method automatically segmented polycystic kidneys in 3D US images obtaining an average Dice coefficient of 0.80 on the test dataset. The kidney volume measurement compared with linear regression coefficient and bias from human tracing were R2 = 0.81, and − 4.42%, and between AI and reference standard were R2 = 0.93, and − 4.12%, respectively. MRI and US measured kidney volumes had R2 = 0.84 and a bias of 7.47%. Conclusion: This is the first study applying deep learning to 3D US in ADPKD. Our method shows promising performance for auto-segmentation of kidneys using 3D US to measure TKV, close to human tracing and MRI measurement. This imaging and analysis method may be useful in a number of settings, including pediatric imaging, clinical studies, and longitudinal tracking of patient disease progression. [ABSTRACT FROM AUTHOR]

Issue Addressed: Dental decay is prevalent among Australian Aboriginal children, yet little is known about their oral health-related behaviours. This study explored the oral health status, behaviours, food and beverage consumption of Aboriginal school children aged 7-9 years in Sydney, Australia.Methods: Parents who were part of an existing longitudinal birth cohort ("Gudaga") were surveyed when their child was between 7 and 9 years. Children (n = 110) also received oral health screening by a trained nurse.Results: A number of children (62%-91%) had at least one visible oral health problem across the 2 years. Around two thirds (62%-67%) of parents rated their child's oral health as excellent/very good and less than half the children (32%-45%) had received dental check-ups. Most children (79%-90%) brushed their teeth and drank water (97%) but more than half (57%-70%) also drank sugar sweetened beverages daily.Conclusions: Parents are instilling good oral health behaviours, however, the oral health screening suggests children are experiencing oral health issues of which parents may be unaware. Parents also seem to be unaware of beverage consumption practices that can increase the risk of childhood decay. SO WHAT?: The findings highlight the need for greater oral health awareness among Aboriginal families on how to recognise early symptoms dental decay and risk factors like sugar sweetened beverages among school going children. This suggests that existing health promotion strategies may not be reaching many Aboriginal families in the urban areas and more culturally appropriate programs may be needed. [ABSTRACT FROM AUTHOR]

Introduction: A community-based project was developed to support quit attempts and denormalise smoking in Aboriginal communities. This qualitative study evaluates the effectiveness of project strategies, messages and the partnership process, and explores impacts observed by Aboriginal community members, complementing findings from a quantitative evaluation.Methods: The study comprised five focus groups (40 participants) and 14 in-depth interviews (with 15 community members). Data were analysed thematically by manual and NVivo software methods.Results: Results demonstrate that the project attracted community attention, was well recalled and messages were considered convincing and persuasive. Participants reported being more likely to quit and to discuss smoking with family and friends, and noticed that many people were quitting. Participants observed an increase in asking people not to smoke in the home and fewer people smoking at events, but noted that many smokers struggled to stay quit. The partnership and participation of Aboriginal Health Workers were viewed as crucial.Conclusion: The qualitative findings reinforce quantitative evaluation findings suggesting that the project contributed to denormalising smoking and motivating quit attempts.So What: The evaluation provided insight into how the project changed attitudes and motivated community members to make quit attempts and provided ideas to meet the ongoing challenge. [ABSTRACT FROM AUTHOR]

Background: Circulatory diseases continue to be the greatest cause of mortality for Australian Aboriginal and Torres Strait Islander people, and a major cause of persistently lower life expectancy compared with non-Aboriginal Australians. The limited information that exists on atrial fibrillation (AF) prevalence in Aboriginal and Torres Strait Islander communities is mostly based on hospital admission data. This shows AF as principal or additional admission diagnosis was 1.4 times higher compared to non-Aboriginal Australians, a higher incidence of AF across the adult life span after age 20 years and a significantly higher prevalence among younger patients. Our study estimates the first national community prevalence and age distribution of AF (including paroxysmal) in Australian Aboriginal people. A handheld single-lead electrocardiograph (ECG) device (iECG), known to be acceptable in this population, was used to record participant ECGs.Methods: This co-designed, descriptive cross-sectional study was conducted in partnership with 16 Aboriginal Community Controlled Health organisations at their facilities and/or with their services delivered elsewhere. The study was also conducted at one state community event. Three (3) Australian jurisdictions were involved: New South Wales, Western Australia and the Northern Territory. Study sites were located in remote, regional and urban areas. Opportunistic recruitment occurred between June 2016 and December 2017. People <45 years of age were excluded.Results: Thirty (30) of 619 Aboriginal people received a 'Possible AF' and 81 an 'Unclassified' result from a hand-held smartphone ECG device. A final diagnosis of AF was made in 29 participants (4.7%; 95%CI 3.0-6.4%), 25 with known AF (five paroxysmal), and four with previously unknown AF. Three (3) of the four with unknown AF were aged between 55-64 years, consistent with a younger age of AF onset in Aboriginal people. Estimated AF prevalence increased with age and was higher in those aged >55 years than the general population (7.2% compared with 5.4%). Slightly more men than women were diagnosed with AF.Conclusions: This study is a significant contribution to the evidence which supports screening for AF in Aboriginal and Torres Strait Islander people commencing at a younger age than as recommended in the Australian guidelines (>65 years). We recommend the age of 55 years. Consideration should be given to the inclusion of AF screening in the Australian Government Department of Health annual 'Aboriginal and Torres Strait Islander Health Assessment'.Clinical Trial Registration: ACTRN12616000459426. [ABSTRACT FROM AUTHOR]

We use Bayesian convolutional neural networks and a novel generative model of Galaxy Zoo volunteer responses to infer posteriors for the visual morphology of galaxies. Bayesian CNN can learn from galaxy images with uncertain labels and then, for previously unlabelled galaxies, predict the probability of each possible label. Our posteriors are well-calibrated (e.g. for predicting bars, we achieve coverage errors of 11.8 per cent within a vote fraction deviation of 0.2) and hence are reliable for practical use. Further, using our posteriors, we apply the active learning strategy BALD to request volunteer responses for the subset of galaxies which, if labelled, would be most informative for training our network. We show that training our Bayesian CNNs using active learning requires up to 35–60 per cent fewer labelled galaxies, depending on the morphological feature being classified. By combining human and machine intelligence, Galaxy zoo will be able to classify surveys of any conceivable scale on a time-scale of weeks, providing massive and detailed morphology catalogues to support research into galaxy evolution. [ABSTRACT FROM AUTHOR]

Introduction: A partnership between three Aboriginal Community Controlled Organisations and a mainstream health service was formed to develop, implement and evaluate a comprehensive and culturally appropriate social marketing project which aimed to encourage smokers to quit smoking. The project also supported quit attempts and promoted denormalisation of smoking.Methods: The project was evaluated through baseline (n = 427) and follow-up (n = 611) surveys carried out face-to-face with Aboriginal and/or Torres Strait Islander participants 18 years and older recruited through convenience sampling at community events and venues during 2010-2011 and 2015.Results: The proportion of participants who had made one or more quit attempts increased significantly between baseline and follow-up surveys (54%, 101 out of 187; vs 64%, 189 out of 297; P < 0.05). Participants who had intended to quit within 6 months (AOR, 3.29; 95% CI 1.90-5.68; P < 0.01); and participants disagreeing with the statement "I don't mind if people smoke inside my home" (AOR, 1.74; 95% CI 1.06-2.84; P < 0.05) were significantly more likely to have made one or more quit attempts compared to the respective reference groups.Conclusion: Study findings demonstrate that the project was associated with increased quit attempts. Intention to quit and attitude were found to be the predictors of making a quit attempt. SO WHAT?: Many studies suggest the need to denormalise smoking; this study demonstrated both change in attitudes and an increase in quit attempts. It is recognised that many quit attempts may be needed for long-term smoking cessation. [ABSTRACT FROM AUTHOR]

Quantifying galaxy morphology is a challenging yet scientifically rewarding task. As the scale of data continues to increase with upcoming surveys, traditional classification methods will struggle to handle the load. We present a solution through an integration of visual and automated classifications, preserving the best features of both human and machine.We demonstrate the effectiveness of such a system through a re-analysis of visual galaxy morphology classifications collected during the Galaxy Zoo 2 (GZ2) project. We reprocess the top-level question of the GZ2 decision tree with a Bayesian classification aggregation algorithm dubbed SWAP, originally developed for the Space Warps gravitational lens project. Through a simple binary classification scheme, we increase the classification rate nearly 5-fold classifying 226 124 galaxies in 92 d of GZ2 project time while reproducing labels derived from GZ2 classification data with 95.7 per cent accuracy. We next combine this with a Random Forest machine learning algorithm that learns on a suite of non-parametric morphology indicators widely used for automated morphologies. We develop a decision engine that delegates tasks between human and machine and demonstrate that the combined system provides at least a factor of 8 increase in the classification rate, classifying 210 803 galaxies in just 32 d of GZ2 project time with 93.1 per cent accuracy. As the Random Forest algorithm requires a minimal amount of computational cost, this result has important implications for galaxy morphology identification tasks in the era of Euclid and other large-scale surveys. [ABSTRACT FROM AUTHOR]

Issue Addressed: Australian Aboriginal children have a higher risk of dental caries yet there is limited focus on oral health risk factors for urban Aboriginal preschool children. This study examined the oral health behaviours and fluid consumption practices of young children from an urban Aboriginal community in south-western Sydney, Australia.Methods: In total, 157 Aboriginal children who were recruited to the "Gudaga" longitudinal birth cohort participated in this study. A survey design was employed and parents responded to the oral health questions when their child was between 18 and 60 months.Results: Few parents (20%) were concerned about their child's oral health across the time period. By 60 months, only 20% of children had seen a dentist while 80% were brushing their teeth at least once daily. High levels of bottle use were seen up to 30 months. Consumption of sugary drinks was also very high in the early years, although this was replaced by water by 36 months.Conclusions: While there are some encouraging findings, such as the rates of tooth brushing and increasing rates of water consumption, the findings do highlight the poor uptake of dental services and high levels of bottle usage among urban aboriginal children during their early years. SO WHAT?: Targeted oral health promotional programs are needed in the urban Aboriginal community to better support parents understanding of good oral health practices in the early years and engagement with dental health services. [ABSTRACT FROM AUTHOR]

Background: Housing is a key determinant of the poor health of Aboriginal Australians. Most Aboriginal people live in cities and large towns, yet research into housing conditions has largely focused on those living in remote areas. This paper measures the prevalence of housing problems amongst participants in a study of urban Aboriginal families in New South Wales, Australia, and examines the relationship between tenure type and exposure to housing problems.Methods: Cross-sectional survey data was provided by 600 caregivers of 1406 Aboriginal children aged 0-17 years participating in Phase One of the Study of Environment on Aboriginal Resilience and Child Health (SEARCH). Regression modelling of the associations between tenure type (own/mortgage, private rental or social housing) and housing problems was conducted, adjusting for sociodemographic factors.Results: The majority (60%) of SEARCH households lived in social housing, 21% rented privately and 19% either owned their home outright or were paying a mortgage ("owned"). Housing problems were common, particularly structural problems, damp and mildew, vermin, crowding and unaffordability. Physical dwelling problems were most prevalent for those living in social housing, who were more likely to report three or more physical dwelling problems than those in owned (PR 3.19, 95%CI 1.97, 5.73) or privately rented homes (PR 1.49, 1.11, 2.08). However, those in social housing were the least likely to report affordability problems. Those in private rental moved home most frequently; children in private rental were more than three times as likely to have lived in four or more homes since birth than those in owned homes (PR 3.19, 95%CI 1.97, 5.73). Those in social housing were almost half as likely as those in private rental to have lived in four or more homes since birth (PR 0.56, 95%CI 0.14, 0.77). Crowding did not vary significantly by tenure type.Conclusions: The high prevalence of housing problems amongst study participants suggests that urban Aboriginal housing requires further attention as part of efforts to reduce the social and health disadvantage experienced by Aboriginal Australians. Particular attention should be directed to the needs of those renting in the private and social housing sectors, who are experiencing the poorest dwelling conditions. [ABSTRACT FROM AUTHOR]

We present optical spectra of the nearby Type Ia supernova SN 2011fe at 100, 205, 311, 349 and 578 d post-maximum light, as well as an ultraviolet (UV) spectrum obtained with the Hubble Space Telescope at 360 d post-maximum light. We compare these observations with synthetic spectra produced with the radiative transfer code PHOENIX. The day +100 spectrum can be well fitted with models that neglect collisional and radiative data for forbidden lines. Curiously, including these data and recomputing the fit yields a quite similar spectrum, but with different combinations of lines forming some of the stronger features. At day +205 and later epochs, forbidden lines dominate much of the optical spectrum formation; however, our results indicate that recombination, not collisional excitation, is the most influential physical process driving spectrum formation at these late times. Consequently, our synthetic optical and UV spectra at all epochs presented here are formed almost exclusively through recombinationdriven fluorescence. Furthermore, our models suggest that the UV spectrum even as late as day +360 is optically thick and consists of permitted lines from several iron-peak species. These results indicate that the transition to the 'nebular' phase in Type Ia supernovae is complex and highly wavelength dependent. [ABSTRACT FROM AUTHOR]

This paper presents the Gudaga Research Program as a case study describing the practice principles used to implement a successful research partnership with an urban Aboriginal community in south-western Sydney. This is one of few papers that address research issues unique to working with urban Aboriginal communities, in which the Aboriginal culture of the community is not homogenous. The authors argue that the relationships between the researchers and key community members and research participants underpin the research success. Throughout, the authors show that ongoing processes to nurture and reaffirm these relationships are important and require ongoing investment. A proposed practice framework demonstrates the relationship between knowledge and understanding of Aboriginal history and culture, the underpinning values including trust, respect, and reciprocity, and shared skills and communication. Examples of how these were built into the research are provided. These are important skills that have application beyond the research process. [ABSTRACT FROM AUTHOR]

Issue addressed Tobacco consumption contributes to health disparities among Aboriginal Australians who experience a greater burden of smoking-related death and diseases. This paper reports findings from a baseline survey on factors associated with smoking, cessation behaviours and attitudes towards smoke-free homes among the Aboriginal population in inner and south-western Sydney. Methods A baseline survey was conducted in inner and south-western Sydney from October 2010 to July 2011. The survey applied both interviewer-administered and self-administered data collection methods. Multiple logistic regression was performed to determine the factors associated with smoking. Results Six hundred and sixty-three participants completed the survey. The majority were female (67.5%), below the age of 50 (66.6%) and more than half were employed (54.7%). Almost half were current smokers (48.4%) with the majority intending to quit in the next 6 months (79.0%) and living in a smoke-free home (70.4%). Those aged 30-39 years (AOR 3.28; 95% CI: 2.06-5.23) and the unemployed (AOR 1.67; 95% CI: 1.11-2.51) had higher odds for current smoking. Participants who had a more positive attitude towards smoke-free homes were less likely to smoke (AOR 0.79; 95% CI: 0.74-.85). Conclusions A high proportion of participants were current smokers among whom intention to quit was high. Age, work status and attitudes towards smoke-free home were factors associated with smoking. So what? The findings address the scarcity of local evidence crucial for promoting cessation among Aboriginal tobacco smokers. Targeted promotions for socio-demographic subgroups and of attitudes towards smoke-free homes could be meaningful strategies for future smoking-cessation initiatives. [ABSTRACT FROM AUTHOR]