Graph learning for particle accelerator operations.

Particle accelerators play a crucial role in scientific research, enabling the study of fundamental physics and materials science, as well as having important medical applications. This study proposes a novel graph learning approach to classify operational beamline configurations as good or bad. By considering the relationships among beamline elements, we transform data from components into a heterogeneous graph. We propose to learn from historical, unlabeled data via our self-supervised training strategy along with fine-tuning on a smaller, labeled dataset. Additionally, we extract a low-dimensional representation from each configuration that can be visualized in two dimensions. Leveraging our ability for classification, we map out regions of the low-dimensional latent space characterized by good and bad configurations, which in turn can provide valuable feedback to operators. This research demonstrates a paradigm shift in how complex, many-dimensional data from beamlines can be analyzed and leveraged for accelerator operations.

Stool characteristics and colonic transit in irritable bowel syndrome: evaluation at two time points.

OBJECTIVE: Information on the relationships between stool characteristics and colonic transit time (CTT) in irritable bowel syndrome (IBS) is limited. Our aims were: (i) to relate stool frequency and consistency to total and segmental CTTs, (ii) to correlate changes in these stool characteristics with changes in CTTs between a baseline assessment and a 12-week assessment, and (iii) to examine the confounding effects of mood on these relationships, in patients with IBS. MATERIALS AND METHODS: Twenty-one female patients with IBS underwent, on two occasions 12 weeks apart, a colonic transit study and completed at these times Bristol Stool Form Scale (BSFS) and Bowel Symptoms Severity Rating Scale (BSSRS). All patients also completed the Hospital Anxiety and Depression scale. RESULTS: Between baseline and the 12-week assessment, an increase in the number of days over the past week without a bowel motion correlated with prolonged total CTT (r = 0.54, p = 0.01). An increase in the number of days with more than three bowel motions per day correlated with a shorter right CTT (r = -0.52, p = 0.02). Only after adjusting for anxiety and depression, did an increase in loose or watery bowel motions (for BSSRS but not for BSFS) correlate with a shorter right CTT (r = -0.47, p = 0.03). CONCLUSIONS: Stool frequency, as well as stool consistency, correlates with CTT. Correlations between stool consistency and CTT are more robust for BSSRS than for BSFS. An effect of mood appears to be important in the relationship between stool consistency and CTT.

Deep Cellular Recurrent Network for Efficient Analysis of Time-Series Data With Spatial Information.

Efficient processing of large-scale time-series data is an intricate problem in machine learning. Conventional sensor signal processing pipelines with hand-engineered feature extraction often involve huge computational costs with high dimensional data. Deep recurrent neural networks have shown promise in automated feature learning for improved time-series processing. However, generic deep recurrent models grow in scale and depth with the increased complexity of the data. This is particularly challenging in presence of high dimensional data with temporal and spatial characteristics. Consequently, this work proposes a novel deep cellular recurrent neural network (DCRNN) architecture to efficiently process complex multidimensional time-series data with spatial information. The cellular recurrent architecture in the proposed model allows for location-aware synchronous processing of time-series data from spatially distributed sensor signal sources. Extensive trainable parameter sharing due to cellularity in the proposed architecture ensures efficiency in the use of recurrent processing units with high-dimensional inputs. This study also investigates the versatility of the proposed DCRNN model for the classification of multiclass time-series data from different application domains. Consequently, the proposed DCRNN architecture is evaluated using two time-series data sets: a multichannel scalp electroencephalogram (EEG) data set for seizure detection, and a machine fault detection data set obtained in-house. The results suggest that the proposed architecture achieves state-of-the-art performance while utilizing substantially less trainable parameters when compared to comparable methods in the literature.

The attachments of 'autonomous' vehicles.

The ideal of the self-driving car replaces an error-prone human with an infallible, artificially intelligent driver. This narrative of autonomy promises liberation from the downsides of automobility, even if that means taking control away from autonomous, free-moving individuals. We look behind this narrative to understand the attachments that so-called 'autonomous' vehicles (AVs) are likely to have to the world. Drawing on 50 interviews with AV developers, researchers and other stakeholders, we explore the social and technological attachments that stakeholders see inside the vehicle, on the road and with the wider world. These range from software and hardware to the behaviours of other road users and the material, social and economic infrastructure that supports driving and self-driving. We describe how innovators understand, engage with or seek to escape from these attachments in three categories: 'brute force', which sees attachments as problems to be solved with more data, 'solve the world one place at a time', which sees attachments as limits on the technology's reach and 'reduce the complexity of the space', which sees attachments as solutions to the problems encountered by technology developers. Understanding attachments provides a powerful way to anticipate various possible constitutions for the technology.

Deep Learning Based Superconducting Radio-Frequency Cavity Fault Classification at Jefferson Laboratory.

This work investigates the efficacy of deep learning (DL) for classifying C100 superconducting radio-frequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. CEBAF is a large, high-power continuous wave recirculating linac that utilizes 418 SRF cavities to accelerate electrons up to 12 GeV. Recent upgrades to CEBAF include installation of 11 new cryomodules (88 cavities) equipped with a low-level RF system that records RF time-series data from each cavity at the onset of an RF failure. Typically, subject matter experts (SME) analyze this data to determine the fault type and identify the cavity of origin. This information is subsequently utilized to identify failure trends and to implement corrective measures on the offending cavity. Manual inspection of large-scale, time-series data, generated by frequent system failures is tedious and time consuming, and thereby motivates the use of machine learning (ML) to automate the task. This study extends work on a previously developed system based on traditional ML methods (Tennant and Carpenter and Powers and Shabalina Solopova and Vidyaratne and Iftekharuddin, Phys. Rev. Accel. Beams, 2020, 23, 114601), and investigates the effectiveness of deep learning approaches. The transition to a DL model is driven by the goal of developing a system with sufficiently fast inference that it could be used to predict a fault event and take actionable information before the onset (on the order of a few hundred milliseconds). Because features are learned, rather than explicitly computed, DL offers a potential advantage over traditional ML. Specifically, two seminal DL architecture types are explored: deep recurrent neural networks (RNN) and deep convolutional neural networks (CNN). We provide a detailed analysis on the performance of individual models using an RF waveform dataset built from past operational runs of CEBAF. In particular, the performance of RNN models incorporating long short-term memory (LSTM) are analyzed along with the CNN performance. Furthermore, comparing these DL models with a state-of-the-art fault ML model shows that DL architectures obtain similar performance for cavity identification, do not perform quite as well for fault classification, but provide an advantage in inference speed.

Personality, gender and medico-legal matters in medical practice.

OBJECTIVES: The aim of this paper was to explore the relationship between the personality traits of Australian General Practitioners (GPs) and their gender, work practice arrangements, and history of medico-legal matters. METHODS: A cross-sectional self report survey was mailed to 1239 GPs. There were 566 respondents (45.7% response rate to survey). The survey assessed personality traits (using the Eysenck Personality Questionnaire), demographic and practice information, and history of medico-legal matters with any medical defence organization. The number and type of medico-legal matters was also extracted from the UNITED Medical Protection database. RESULTS: Male respondents had significantly higher psychoticism scores than females (p<0.001), and females had significantly higher neuroticism scores than males (p<0.01), as in community samples. However, for GPs who worked more than 48 hours per week, there were no gender differences in personality trait scores. Solo practitioners and non-solo practitioners did not differ on personality scores. Proceduralists and non-proceduralists did not differ on personality scores. However, a higher proportion of proceduralists experienced a medico-legal matter than non-proceduralists (p<0.001). There was a positive correlation between extraversion scores and doctors who attended peer review (p<0.001). There was no difference in the numbers of medico-legal matters for doctors who attended peer review. Males who self reported a medico-legal matter had higher neuroticism scores than the males who did not report medico-legal matters. This was not the case for females. For males, this pattern was not replicated when considering data from UNITED. CONCLUSIONS: The known demographic and practice factors that differ for doctors having a medico-legal matter are replicated here--being male, a proceduralist and working longer hours. There is not a consistent pattern regarding personality traits and medico-legal matters.

GPs' concerns about medicolegal issues - How it affects their practice.

BACKGROUND: General practitioners' concerns about medicolegal issues have been shown to influence the practice of medicine. This research looks at GPs' beliefs about medicolegal issues and how medicolegal concerns affect their practice. METHODS: A descriptive comparative design was used. A cross sectional self report survey was sent to 1239 GPs, 566 responded (46% response rate). Responses were considered as a group, and then comparisons were made between those who had experienced a medicolegal matter and those who had not. This data was sourced from surveys and medicolegal insurer records. RESULTS: General practitioners with previous medicolegal experiences were more likely than their colleagues to report believing the law required them to make perfect decisions and that medicolegal factors made them consider early retirement from medicine. They were also less likely to believe that inadequate communication is a factor in most complaints. More than half the GPs reported having made practice changes due to medicolegal concerns in the following areas: test ordering (73%); specialist referrals (66%); systems to track test results (70%); and communication of risk to patients (68%). Other changes were reported less frequently. DISCUSSION: This study found that GPs' concerns about medicolegal matters impact on their practise of medicine. While greater awareness of medicolegal issues may lead to positive impacts, the negative impact of their concerns is that some changes arise from anxiety about medicolegal matters rather than from the exercise of good clinical judgment.

Factors associated with psychiatric morbidity and hazardous alcohol use in Australian doctors.

OBJECTIVE: To identify factors associated with psychiatric morbidity and hazardous alcohol use in Australian doctors. DESIGN, SETTING AND PARTICIPANTS: Cross-sectional postal survey of 2999 doctors (including all major specialty groups, trainees and general practitioners) insured with an Australian medical insurance company. The potential for psychiatric morbidity was measured by the General Health Questionnaire (GHQ), and the potential for hazardous alcohol use by the Alcohol Use Disorders Identification Test (AUDIT). The survey was conducted in 2007. MAIN OUTCOME MEASURES: Demographic, work-related and personality factors associated with a GHQ score > 4 and an AUDIT score > or = 8. RESULTS: Factors significantly associated with psychiatric morbidity in doctors were: having a current medicolegal matter, not taking a holiday in the previous year, working long hours, type of specialty, and having personality traits of neuroticism and introversion. Factors significantly associated with potentially hazardous alcohol use were being male, being Australian-trained, being between 40 and 49 years of age, having personality traits of neuroticism and extroversion, failing to meet Continuing Medical Education requirements, and being a solo practitioner. CONCLUSIONS: The mental health of medical practitioners is crucial to the quality of care their patients receive. Doctors should reflect on their hours of work and need for holidays. Involvement with medicolegal processes, such as lawsuits, complaints and inquiries, is a stressful part of medical practice today. Doctors need to be educated about these processes and understand how the experience may affect their health, work and loved ones.

The temporal dynamics of relationships between cannabis, psychosis and depression among young adults with psychotic disorders: findings from a 10-month prospective study.

BACKGROUND: The aim was to examine the temporal relationships over 10 months between cannabis use and symptoms of psychosis and depression in people with schizophrenia and related disorders. The design was a prospective study of 101 patients with schizophrenia and related disorders who were assessed monthly over 10 months on medication compliance, cannabis and other drug use, symptoms of depression and symptoms of psychosis. METHOD: Linear regression methods to assess relationships between cannabis use and symptoms of psychosis and depression while adjusting for serial dependence, medication compliance and other demographic and clinical variables. RESULTS: Cannabis use predicted a small but statistically significant increase in symptoms of psychosis, but not depression, after controlling for other differences between cannabis users and non-users. Symptoms of depression and psychosis did not predict cannabis use. CONCLUSION: Continued cannabis use by persons with schizophrenia predicts a small increase in psychotic symptom severity but not vice versa.

Psychological morbidity in Australian doctors who have and have not experienced a medico-legal matter: cross-sectional survey.

OBJECTIVE: To describe the differences in psychological morbidity between Australian general practitioners (GPs) who have experienced a medico-legal matter and those who have not. METHODS: A total of 1,499 GPs were initially invited to participate in the study. Two hundred and sixty requested not to participate, with 1,239 subsequently being sent a survey. There were 566 respondents (45.7% response rate to survey). There were two sources of data. First, a cross-sectional survey sought demographic information, personality traits via the Eysenck Personality Questionnaire (EPQ), history of a medico-legal matter with any medical defence organization, and measures of psychological morbidity, including the General Health Questionnaire (GHQ), Sheehan Disability Scale (SDS), and Alcohol Use Disorders Identification Test. Second, information was extracted from the United Medical Protection database on medico-legal matters. RESULTS: Fifty-nine per cent of respondents to the survey reported ever having a medico-legal matter, with 13% having a current medico-legal matter. Those with a current matter reported increased levels of disability (in work, social or family life) and higher prevalence of psychiatric morbidity (45% vs 27% GHQ 'case identification' rates), compared to those with no current matter. Those respondents with a history of past medico-legal matters reported increased levels of disability (SDS) and depression subscores (GHQ). Male respondents drank significantly more alcohol than female respondents, and male respondents with current or past medico-legal matters had significantly higher levels of alcohol use than male respondents with no experience of medico-legal matters. CONCLUSIONS: Doctors who have current and past medico-legal matters have a higher level of psychological morbidity. The study design was unable to distinguish cause or effect. A longitudinal study is planned to investigate this. The findings have significant implications for medical training, doctor support systems and medical insurance groups.

Reasons for cannabis use in psychosis.

OBJECTIVE: To examine the reasons for cannabis use among individuals with psychotic disorders. METHOD: Forty-nine people with psychotic disorders in treatment with community health centres in Northern Sydney were interviewed to collect information about their experience of antipsychotic side-effects and their influence on cannabis use. Other information collected on cannabis use included: amount and frequency, effects of use and other general reasons given for use. RESULTS: It was found that boredom, social motives, improving sleep, anxiety and agitation and symptoms associated with negative psychotic symptoms or depression were the most important motivators of cannabis use. Positive symptoms of psychosis and antipsychotic side-effects that were not associated with anxiety, were not important motivators of cannabis use. CONCLUSIONS: As cannabis use may precipitate relapse in this population, it is important to reduce these motivators of use. Clinician's must assess and treat these problems, thus reducing the need for patients to self-medicate with cannabis, and therefore reducing the risk of relapse.

Let's get physical: an audit of medical practice in first episode psychosis.

OBJECTIVES: To assess aspects of medical examination, diagnosis and side-effect monitoring, and to consider the role of routine investigations in this group as recommended by national guidelines. METHOD: A retrospective file audit was performed on young people presenting with first episode psychosis (n = 117) over 12 months of treatment contact. RESULTS: Diagnoses were: first episode psychosis (43%), schizophrenia (16%), drug-induced psychosis (12%), affective psychosis (13%) and brief reactive psychosis (2%). Only four of the 52 (8%) subjects undergoing neuroimaging had any abnormality, with only two of these requiring referral. Three of the 33 (9%) electroencephalograms were obviously abnormal, but without epileptiform activity. There was little documentation of the assessment of involuntary movements (4% of sample) or weight (15% of sample). CONCLUSIONS: The low rates of clinically important abnormal findings in computed tomography/magnetic resonance imaging and electroencephalogram re-open debate about the need for routine neuroimaging and electrophysiology in this population.

The development of a mental health service in East Timor: an Australian mental health relief project.

East Timor (the Democratic Republic of Timor-Leste) occupies the eastern half of the island of Timor, which lies between North Western Australia and the Indonesian archipelago. East Timor has a population of around 860 000. It is predominantly rural and there are few large towns. The country has a largely subsistence agricultural economy; coffee is the principal cash crop. The population is extremely poor, and transport and communications are primitive.

Psychological trauma: psychiatry and the law in conflict.

OBJECTIVE: To contrast the psychiatric and legal construct of psychological trauma, with posttraumatic stress disorder (PTSD) as the reference disorder. METHOD: Literature relevant to DSM-IV PTSD definition of a traumatic event (Criterion A) was assessed. This construct was compared with the current legal status for psychological trauma arising from recent judgements of the Australian High Court. CONCLUSIONS: The current legal construct of psychological trauma is not only practical and sensible, but is far more evidence-based than the DSM-IV construct of trauma.

Changes in early psychosis service provision: a file audit.

OBJECTIVE: To measure change in services provided to young people with first-episode psychosis following the introduction of specialized early psychosis teams and staff training. METHOD: A standardized tool was developed to audit services provided to young people with first-episode psychosis. The tool initially comprised 27 clinical indicators measuring aspects of optimal care derived from the Australian clinical guidelines for early psychosis. The first 12 months of treatment, as documented in the case records, were audited for all young people receiving their first treatment for psychosis during a 6-month period prior to the introduction of these service developments (n = 47). These subjects were compared with those who received treatment after the implementation of service development strategies (n = 70). A comparison was also made within the second group, between those receiving some treatment from a specialized early psychosis team and those being exclusively treated by other services. RESULTS: Inter-rater reliability was achieved for 24 of the 27 indicators. Improvements were found on 10 indicators which measured psychosocial interventions, prescribing practices, family interventions and continuity of care. There was no significant deterioration on any of the indicators. Clients who attended early psychosis teams were significantly more likely to receive psychoeducation. CONCLUSIONS: The services increased their provision of "guideline concordant" care for early psychosis. The audit proved useful to monitor performance, to demonstrate improvements in care and to identify those areas of service provision and documentation in need of improvement.

Clinical outcome of an early psychosis intervention program: evaluation in a real-world context.

OBJECTIVE: To examine whether staff training and service restructuring to provide specialized early psychosis services results in improved clinical outcomes for young people with first-episode psychosis. METHOD: Staff attended workshops on the treatment of early psychosis throughout 1997-2000 and specialized early psychosis teams began operating between 1998 and 2000 following service restructure. There was no additional funding provided for clinical services, but through the restructure, there was a shift in resources. During this period a comprehensive package including the Brief Psychiatric Rating Scale, Scale for the Assessment of Negative Symptoms and the Health of the Nation Outcome Scale was introduced for clinicians to assess patients at intake, 3 months and 12 months into treatment. Symptom scores of patients treated earlier in the project are compared with those patients treated later, after more training and service developments had occurred. RESULTS: Ninety-four of 215 potential first-episode patients consented to take part in the project. They provided data variously at intake, 3 months and 12 months into treatment. Regardless of the year of treatment, significant improvement in psychiatric symptomatology was found over the three assessment periods. Individuals who entered the service in the latter phase of the project experienced fewer negative symptoms (after 12 months of treatment) compared with patients who entered the service in the early phase of the project. CONCLUSIONS: Improvements in both pharmacological and possibly psychosocial treatment may have led to a greater improvement in negative symptoms. Benefits and limitations of conducting research in a "real-world" context are discussed.