Explore the vast range of titles available.

International challenges have become the de facto standard for comparative assessment of image analysis algorithms. Although segmentation is the most widely investigated medical image processing task, the various challenges have been organized to focus only on specific clinical tasks. We organized the Medical Segmentation Decathlon (MSD)—a biomedical image analysis challenge, in which algorithms compete in a multitude of both tasks and modalities to investigate the hypothesis that a method capable of performing well on multiple tasks will generalize well to a previously unseen task and potentially outperform a custom-designed solution. MSD results confirmed this hypothesis, moreover, MSD winner continued generalizing well to a wide range of other clinical problems for the next two years. Three main conclusions can be drawn from this study: (1) state-of-the-art image segmentation algorithms generalize well when retrained on unseen tasks; (2) consistent algorithmic performance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the training of accurate AI segmentation models is now commoditized to scientists that are not versed in AI model training. International challenges have become the de facto standard for comparative assessment of image analysis algorithms. Here, the authors present the results of a biomedical image segmentation challenge, showing that a method capable of performing well on multiple tasks will generalize well to a previously unseen task.

Severe mental illnesses have been linked to white matter abnormalities, documented by postmortem studies. However, cause and effect have remained difficult to distinguish. CNP (2′,3′‐cyclic nucleotide 3′‐phosphodiesterase) is among the oligodendrocyte/myelin‐associated genes most robustly reduced on mRNA and protein level in brains of schizophrenic, bipolar or major depressive patients. This suggests that CNP reduction might be critical for a more general disease process and not restricted to a single diagnostic category. We show here that reduced expression of CNP is the primary cause of a distinct behavioural phenotype, seen only upon aging as an additional ‘pro‐inflammatory hit’. This phenotype is strikingly similar in Cnp heterozygous mice and patients with mental disease carrying the AA genotype at CNP SNP rs2070106. The characteristic features in both species with their partial CNP ‘loss‐of‐function’ genotype are best described as ‘catatonia‐depression’ syndrome. As a consequence of perturbed CNP expression, mice show secondary low‐grade inflammation/neurodegeneration. Analogously, in man, diffusion tensor imaging points to axonal loss in the frontal corpus callosum. To conclude, subtle white matter abnormalities inducing neurodegenerative changes can cause/amplify psychiatric diseases.

Bell’s mania (mania with delirium) is an acute neurobehavioral syndrome of unknown etiology that is characterized by the rapid onset of grandiosity, psychomotor excitement, emotional lability, psychosis, and sleep disruption consistent with mania, coupled with alterations in sensorium, and disorientation characteristic of delirium. Catatonia is a common feature of the syndrome. The authors describe a case of recurrent delirium/mania with prominent catatonic features after a cerebellar and pontine stroke, and subsequent successful treatment with lorazepam. Symptoms quickly resolved after antipsychotics were discontinued, with continuation of valproate and lorazapam treatment. Failure to recognize this patient’s syndrome as a form of catatonia could have had severe, even life-threatening, consequences. The use of neuroleptic medications in cases of delirium/mania with catatonic signs may result in marked clinical deterioration, whereas high-dose lorazepam can ameliorate catatonic signs.

International challenges have become the de facto standard for comparative assessment of image analysis algorithms given a specific task. Segmentation is so far the most widely investigated medical image processing task, but the various segmentation challenges have typically been organized in isolation, such that algorithm development was driven by the need to tackle a single specific clinical problem. We hypothesized that a method capable of performing well on multiple tasks will generalize well to a previously unseen task and potentially outperform a custom-designed solution. To investigate the hypothesis, we organized the Medical Segmentation Decathlon (MSD) - a biomedical image analysis challenge, in which algorithms compete in a multitude of both tasks and modalities. The underlying data set was designed to explore the axis of difficulties typically encountered when dealing with medical images, such as small data sets, unbalanced labels, multi-site data and small objects. The MSD challenge confirmed that algorithms with a consistent good performance on a set of tasks preserved their good average performance on a different set of previously unseen tasks. Moreover, by monitoring the MSD winner for two years, we found that this algorithm continued generalizing well to a wide range of other clinical problems, further confirming our hypothesis. Three main conclusions can be drawn from this study: (1) state-of-the-art image segmentation algorithms are mature, accurate, and generalize well when retrained on unseen tasks; (2) consistent algorithmic performance across multiple tasks is a strong surrogate of algorithmic generalizability; (3) the training of accurate AI segmentation models is now commoditized to non AI experts.

Semantic segmentation of medical images aims to associate a pixel with a label in a medical image without human initialization. The success of semantic segmentation algorithms is contingent on the availability of high-quality imaging data with corresponding labels provided by experts. We sought to create a large collection of annotated medical image datasets of various clinically relevant anatomies available under open source license to facilitate the development of semantic segmentation algorithms. Such a resource would allow: 1) objective assessment of general-purpose segmentation methods through comprehensive benchmarking and 2) open and free access to medical image data for any researcher interested in the problem domain. Through a multi-institutional effort, we generated a large, curated dataset representative of several highly variable segmentation tasks that was used in a crowd-sourced challenge - the Medical Segmentation Decathlon held during the 2018 Medical Image Computing and Computer Aided Interventions Conference in Granada, Spain. Here, we describe these ten labeled image datasets so that these data may be effectively reused by the research community.

Catatonia, characterized by staring, immobility, mutism, and unusual postures, has diverse causes. Treatment of an underlying disorder and intravenous lorazepam, sometimes at high doses, are usually successful.

Introduction Compared to usual care, specialty services for first‐episode psychosis (FES) have superior patient outcomes. The Early Psychosis Intervention Network (EPINET), comprised of eight U.S. regional clinical networks, aims to advance the quality of FES care within the ethos of learning healthcare systems (LHS). Among these, the Academic Community (AC) EPINET was established to provide FES care, collect common data elements, leverage informatics, foster a culture of continuous learning and quality improvement, and engage in practice‐based research. Methods We designed and implemented a novel LHS of university‐affiliated FES programs within a hub (academic leadership team) and spoke (FES clinics) model. A series of site implementation meetings engaged stakeholders, setting the stage for a culture that values data collection and shared learning. We built clinical workflows to collect common data elements at enrollment and at consecutive 6‐month intervals in parallel to an informatics workflow to deliver outcome visualizations and drive quality improvement efforts. Results All six clinical sites successfully implemented data capture workflows and engaged in the process of designing the informatics platform. Upon developing the structure, processes, and initial culture of the LHS, a total of 614 patients enrolled in AC‐EPINET, with the most common primary diagnoses of schizophrenia (32.1%) and unspecified psychotic disorders (23.6%). Visualized outcomes were delivered to clinical teams who began to consider locally relevant quality improvement projects. Conclusions AC‐EPINET is a novel LHS, with a simultaneous focus on science, informatics, incentives, and culture. The work of developing AC‐EPINET thus far has highlighted the need for future LHS’ to be mindful of the complexities of data security issues, develop more automated informatic workflows, resource quality assurance efforts, and attend to building the cultural infrastructure with the input of all stakeholders.

Negative symptom severity predicts functional outcome and quality life in people with psychosis. However, negative symptoms are poorly responsive to antipsychotic medication and existing literature has not converged on their neurobiological basis. Previous work in small schizophrenia samples has observed that lower cerebellar-prefrontal connectivity is associated with higher negative symptom severity and demonstrated in a separate neuromodulation experiment that increasing cerebellar-prefrontal connectivity reduced negative symptom severity. We sought to expand this finding to test associations between cerebellar-prefrontal connectivity with negative symptom severity and cognitive performance in a large, transdiagnostic sample of individuals with psychotic disorders. In this study, 260 individuals with psychotic disorders underwent resting-state MRI and clinical characterization. Negative symptom severity was measured using the Positive and Negative Symptoms Scale, and cognitive performance was assessed with the Screen for Cognitive Impairment in Psychiatry. Using a previously identified cerebellar region as a seed, we performed seed to whole brain analyses and regressed connectivity against negative symptom severity, using age and sex as covariates. Consistent with prior work, we identified relationships between higher cerebellar-prefrontal connectivity and lower negative symptom severity (r=-0.17, p=.007). Higher cerebellar-prefrontal connectivity was also associated with better delayed verbal learning (r=.13, p=.034). Our results provide further evidence supporting the relationship between cerebellar-prefrontal connectivity and negative symptom severity and cognitive performance. Larger, randomized, sham-controlled neuromodulation studies should test if increasing cerebellar-prefrontal connectivity leads to reductions in negative symptoms in psychosis.

Experts from academia and industry discuss how to create a new, more effective translational neuroscience drawing on novel technology and recent discoveries. Today, translational neuroscience faces significant challenges. Available therapies to treat brain and nervous system disorders are extremely limited and dated, and further development has effectively ceased. Disinvestment by the private sector occurred just as promising new technologies in genomics, stem cell biology, and neuroscience emerged to offer new possibilities. In this volume, experts from both academia and industry discuss how novel technologies and reworked translation concepts can create a more effective translational neuroscience. The contributors consider such topics as using genomics and neuroscience for better diagnostics and biomarker identification; new approaches to disease based on stem cell technology and more careful use of animal models; and greater attention to human biology and what it will take to make new therapies available for clinical use. They conclude with a conceptual roadmap for an effective and credible translational neuroscience—one informed by a disease-focused knowledge base and clinical experience. Contributors Tobias M. Böckers, Thomas Bourgeron, Karl Broich, Nils Brose, Bruce N. Cuthbert, Ilka Diester, Gül Dölen, Guoping Feng, Richard Frackowiak, Raquel E. Gur, Stephan Heckers, Franz Hefti, David M. Holtzman, Steven E. Hyman, Nancy Ip, Cynthia Joyce, Tobias Kaiser, Edward H. Koo, Walter J. Koroshetz, Katja S. Kroker, Robert C. Malenka, Isabelle Mansuy, Eliezer, Masliah, Yuan Mei, Andreas Meyer-Lindenberg, Lennart Mucke, Pierluigi Nicotera, Karoly Nikolich, Michael J. Owen, Menelas N. Pangalos, Alvaro Pascual-Leone, Joel S. Perlmutter, Trevor W. Robbins, Lee L. Rubin, Akira Sawa, Mareike Schnaars, Bernd Sommer, Maria Grazia Spillantini, Laura Spinney, Matthew W. State, Marius Wernig