Interoperable slide microscopy viewer and annotation tool for imaging data science and computational pathology.

The exchange of large and complex slide microscopy imaging data in biomedical research and pathology practice is impeded by a lack of data standardization and interoperability, which is detrimental to the reproducibility of scientific findings and clinical integration of technological innovations. We introduce Slim, an open-source, web-based slide microscopy viewer that implements the internationally accepted Digital Imaging and Communications in Medicine (DICOM) standard to achieve interoperability with a multitude of existing medical imaging systems. We showcase the capabilities of Slim as the slide microscopy viewer of the NCI Imaging Data Commons and demonstrate how the viewer enables interactive visualization of traditional brightfield microscopy and highly-multiplexed immunofluorescence microscopy images from The Cancer Genome Atlas and Human Tissue Atlas Network, respectively, using standard DICOMweb services. We further show how Slim enables the collection of standardized image annotations for the development or validation of machine learning models and the visual interpretation of model inference results in the form of segmentation masks, spatial heat maps, or image-derived measurements.

Highdicom: a Python Library for Standardized Encoding of Image Annotations and Machine Learning Model Outputs in Pathology and Radiology.

Machine learning (ML) is revolutionizing image-based diagnostics in pathology and radiology. ML models have shown promising results in research settings, but the lack of interoperability between ML systems and enterprise medical imaging systems has been a major barrier for clinical integration and evaluation. The DICOM® standard specifies information object definitions (IODs) and services for the representation and communication of digital images and related information, including image-derived annotations and analysis results. However, the complexity of the standard represents an obstacle for its adoption in the ML community and creates a need for software libraries and tools that simplify working with datasets in DICOM format. Here we present the highdicom library, which provides a high-level application programming interface (API) for the Python programming language that abstracts low-level details of the standard and enables encoding and decoding of image-derived information in DICOM format in a few lines of Python code. The highdicom library leverages NumPy arrays for efficient data representation and ties into the extensive Python ecosystem for image processing and machine learning. Simultaneously, by simplifying creation and parsing of DICOM-compliant files, highdicom achieves interoperability with the medical imaging systems that hold the data used to train and run ML models, and ultimately communicate and store model outputs for clinical use. We demonstrate through experiments with slide microscopy and computed tomography imaging, that, by bridging these two ecosystems, highdicom enables developers and researchers to train and evaluate state-of-the-art ML models in pathology and radiology while remaining compliant with the DICOM standard and interoperable with clinical systems at all stages. To promote standardization of ML research and streamline the ML model development and deployment process, we made the library available free and open-source at https://github.com/herrmannlab/highdicom .

Childhood encephalitis in the Greater Mekong region (the SouthEast Asia Encephalitis Project): a multicentre prospective study.

BACKGROUND: Encephalitis is a worldwide public health issue, with a substantially high burden among children in southeast Asia. We aimed to determine the causes of encephalitis in children admitted to hospitals across the Greater Mekong region by implementing a comprehensive state-of-the-art diagnostic procedure harmonised across all centres, and identifying clinical characteristics related to patients' conditions. METHODS: In this multicentre, observational, prospective study of childhood encephalitis, four referral hospitals in Cambodia, Vietnam, Laos, and Myanmar recruited children (aged 28 days to 16 years) who presented with altered mental status lasting more than 24 h and two of the following minor criteria: fever (within the 72 h before or after presentation), one or more generalised or partial seizures (excluding febrile seizures), a new-onset focal neurological deficit, cerebrospinal fluid (CSF) white blood cell count of 5 per mL or higher, or brain imaging (CT or MRI) suggestive of lesions of encephalitis. Comprehensive diagnostic procedures were harmonised across all centres, with first-line testing was done on samples taken at inclusion and results delivered within 24 h of inclusion for main treatable causes of disease and second-line testing was done thereafter for mostly non-treatable causes. An independent expert medical panel reviewed the charts and attribution of causes of all the included children. Using multivariate analyses, we assessed risk factors associated with unfavourable outcomes (ie, severe neurological sequelae and death) at discharge using data from baseline and day 2 after inclusion. This study is registered with ClinicalTrials.gov, NCT04089436, and is now complete. FINDINGS: Between July 28, 2014, and Dec 31, 2017, 664 children with encephalitis were enrolled. Median age was 4·3 years (1·8-8·8), 295 (44%) children were female, and 369 (56%) were male. A confirmed or probable cause of encephalitis was identified in 425 (64%) patients: 216 (33%) of 664 cases were due to Japanese encephalitis virus, 27 (4%) were due to dengue virus, 26 (4%) were due to influenza virus, 24 (4%) were due to herpes simplex virus 1, 18 (3%) were due to Mycobacterium tuberculosis, 17 (3%) were due to Streptococcus pneumoniae, 17 (3%) were due to enterovirus A71, 74 (9%) were due to other pathogens, and six (1%) were due to autoimmune encephalitis. Diagnosis was made within 24 h of admission to hospital for 83 (13%) of 664 children. 119 (18%) children had treatable conditions and 276 (42%) had conditions that could have been preventable by vaccination. At time of discharge, 153 (23%) of 664 children had severe neurological sequelae and 83 (13%) had died. In multivariate analyses, risk factors for unfavourable outcome were diagnosis of M tuberculosis infection upon admission (odds ratio 3·23 [95% CI 1·04-10·03]), coma on day 2 (2·90 [1·78-4·72]), supplementary oxygen requirement (1·89 [1·25-2·86]), and more than 1 week duration between symptom onset and admission to hospital (3·03 [1·68-5·48]). At 1 year after inclusion, of 432 children who were discharged alive from hospital with follow-up data, 24 (5%) had died, 129 (30%) had neurological sequelae, and 279 (65%) had completely recovered. INTERPRETATION: In southeast Asia, most causes of childhood encephalitis are either preventable or treatable, with Japanese encephalitis virus being the most common cause. We provide crucial information that could guide public health policy to improve diagnostic, vaccination, and early therapeutic guidelines on childhood encephalitis in the Greater Mekong region. FUNDING: Institut Pasteur, Institut Pasteur International Network, Fondation Merieux, Aviesan Sud, INSERM, Wellcome Trust, Institut de Recherche pour le Développement (IRD), and Fondation Total.

Hopfield networks as a model of prototype-based category learning: A method to distinguish trained, spurious, and prototypical attractors.

We present an investigation of the potential use of Hopfield networks to learn neurally plausible, distributed representations of category prototypes. Hopfield networks are dynamical models of autoassociative memory which learn to recreate a set of input states from any given starting state. These networks, however, will almost always learn states which were not presented during training, so called spurious states. Historically, spurious states have been an undesirable side-effect of training a Hopfield network and there has been much research into detecting and discarding these unwanted states. However, we suggest that some of these states may represent useful information, namely states which represent prototypes of the categories instantiated in the network's training data. It would be desirable for a memory system trained on multiple instance tokens of a category to extract a representation of the category prototype. We present an investigation showing that Hopfield networks are in fact capable of learning category prototypes as strong, stable, attractors without being explicitly trained on them. We also expand on previous research into the detection of spurious states in order to show that it is possible to distinguish between trained, spurious, and prototypical attractors.

The mood disorders association of british columbia psychiatric urgent care program: a preliminary evaluation of a suggested alternative model of outpatient psychiatric care.

OBJECTIVE: To describe an alternative model of psychiatric outpatient care for patients with mood and anxiety disorders (the Mood Disorders Association of British Columbia Psychiatric Urgent Care Program or the MDA Program) using group medical visits (GMV) and (or) email communications in lieu of individual follow-up appointments. METHOD: Annual costs of the MDA Program were compared with average costs of private psychiatrists offering outpatient care and patients being treated in a mental health centre. In addition, questionnaires as to patient satisfaction with the MDA Program intake, GMV experience, and family physician satisfaction with the MDA Program were administered. RESULTS: The MDA Program model of care is significantly more cost effective than individual psychiatric outpatient care or health authority mental health centre care for patients with moderate or severe illness. Patients and family physicians were very satisfied with the model of care and GMVs offered. CONCLUSIONS: The MDA Program model of care appears to be efficient and cost-effective, and patients and referring physicians appear satisfied with the care offered in this program.

Hyaluronan synthesis and myogenesis: a requirement for hyaluronan synthesis during myogenic differentiation independent of pericellular matrix formation.

Exogenous hyaluronan is known to alter muscle precursor cell proliferation, migration, and differentiation, ultimately inhibiting myogenesis in vitro. The aim of the current study was to investigate the role of endogenous hyaluronan synthesis during myogenesis. In quantitative PCR studies, the genes responsible for synthesizing hyaluronan were found to be differentially regulated during muscle growth, repair, and pathology. Although all Has genes (Has1, Has2, and Has3) were differentially regulated in these models, only Has2 gene expression consistently associated with myogenic differentiation. During myogenic differentiation in vitro, Has2 was the most highly expressed of the synthases and increased after induction of differentiation. To test whether this association between Has2 expression and myogenesis relates to a role for Has2 in myoblast differentiation and fusion, C2C12 myoblasts were depleted of Has2 by siRNA and induced to differentiate. Depletion of Has2 inhibited differentiation and caused a loss of cell-associated hyaluronan and the hyaluronan-dependent pericellular matrix. The inhibition of differentiation caused by loss of hyaluronan was confirmed with the hyaluronan synthesis inhibitor 4-methylumbelliferone. In hyaluronan synthesis-blocked cultures, restoration of the pericellular matrix could be achieved through the addition of exogenous hyaluronan and the proteoglycan versican, but this was not sufficient to restore differentiation to control levels. These data indicate that intrinsic hyaluronan synthesis is necessary for myoblasts to differentiate and form syncytial muscle cells, but the hyaluronan-dependent pericellular matrix is not sufficient to support differentiation alone; additional hyaluronan-dependent cell functions that are yet unknown may be required for myogenic differentiation.

Measuring the severity of depression and remission in primary care: validation of the HAMD-7 scale.

BACKGROUND: Symptomatic remission is the optimal outcome in depression. A brief, validated tool for symptom measurement that can indicate when remission has occurred in mental health and primary care settings is unavailable. We evaluated a 7-item abbreviated version (HAMD-7) of the 17-item Hamilton Depression Rating Scale (HAMD-17) in a randomized controlled clinical trial of patients with major depressive disorder being cared for in primary care settings. METHODS: We enrolled 454 patients across 47 primary care settings who met DSM-IV-TR criteria for a major depressive disorder. Of these, 410 patients requiring antidepressant medication were randomized to have their symptoms rated with either HAMD-7 (n = 205) or HAMD-17 (n = 205) as the primary measurement tool. The primary outcome was the proportion of patients who achieved a-priori defined responses to 8 weeks of therapy using each instrument. RESULTS: Of the 205 participants per group, 67% of those evaluated with HAMD-7 were classified as having responded to therapy (defined as a > or = 50% reduction from the pretreatment score), compared with 74% of those evaluated with HAMD-17 (p = 0.43). The difference between the groups' changes in scores from baseline (pretreatment) to endpoint was significant (p < 0.001), without a main effect of group (p = 0.84) or group-by-time (p = 0.83) interaction. The HAMD-7 test was brief to administer (e.g., 3-4 min for 85% of the primary care physicians evaluated), which facilitated the efficient and structured evaluation of salient depressive symptoms. INTERPRETATION: The abbreviated HAMD-7 depression scale is equivalent to the HAMD-17 in assessing remission in patients with a major depressive disorder undergoing drug therapy.

Canadian Network for Mood and Anxiety Treatments (CANMAT) guidelines for the management of patients with bipolar disorder: consensus and controversies.

Since the previous publication of Canadian Network for Mood and Anxiety Treatments (CANMAT) guidelines in 1997, there has been a substantial increase in evidence-based treatment options for bipolar disorder. The present guidelines review the new evidence and use criteria to rate strength of evidence and incorporate effectiveness, safety, and tolerability data to determine global clinical recommendations for treatment of various phases of bipolar disorder. The guidelines suggest that although pharmacotherapy forms the cornerstone of management, utilization of adjunctive psychosocial treatments and incorporation of chronic disease management model involving a healthcare team are required in providing optimal management for patients with bipolar disorder. Lithium, valproate and several atypical antipsychotics are first-line treatments for acute mania. Bipolar depression and mixed states are frequently associated with suicidal acts; therefore assessment for suicide should always be an integral part of managing any bipolar patient. Lithium, lamotrigine or various combinations of antidepressant and mood-stabilizing agents are first-line treatments for bipolar depression. First-line options in the maintenance treatment of bipolar disorder are lithium, lamotrigine, valproate and olanzapine. Historical and symptom profiles help with treatment selection. With the growing recognition of bipolar II disorders, it is anticipated that a larger body of evidence will become available to guide treatment of this common and disabling condition. These guidelines also discuss issues related to bipolar disorder in women and those with comorbidity and include a section on safety and monitoring.