NuCLS: A scalable crowdsourcing approach and dataset for nucleus classification and segmentation in breast cancer.

BACKGROUND: Deep learning enables accurate high-resolution mapping of cells and tissue structures that can serve as the foundation of interpretable machine-learning models for computational pathology. However, generating adequate labels for these structures is a critical barrier, given the time and effort required from pathologists. RESULTS: This article describes a novel collaborative framework for engaging crowds of medical students and pathologists to produce quality labels for cell nuclei. We used this approach to produce the NuCLS dataset, containing >220,000 annotations of cell nuclei in breast cancers. This builds on prior work labeling tissue regions to produce an integrated tissue region- and cell-level annotation dataset for training that is the largest such resource for multi-scale analysis of breast cancer histology. This article presents data and analysis results for single and multi-rater annotations from both non-experts and pathologists. We present a novel workflow that uses algorithmic suggestions to collect accurate segmentation data without the need for laborious manual tracing of nuclei. Our results indicate that even noisy algorithmic suggestions do not adversely affect pathologist accuracy and can help non-experts improve annotation quality. We also present a new approach for inferring truth from multiple raters and show that non-experts can produce accurate annotations for visually distinctive classes. CONCLUSIONS: This study is the most extensive systematic exploration of the large-scale use of wisdom-of-the-crowd approaches to generate data for computational pathology applications.

Combined use of electroconvulsive therapy and antipsychotics (both clozapine and non-clozapine) in treatment resistant schizophrenia: A comparative meta-analysis.

AIM: To assess the relative efficacies of clozapine plus Electroconvulsive Therapy (ECT) compared against non-clozapine typical and atypical antipsychotics plus ECT for the treatment of "Treatment Resistant Schizophrenia" (TRS). Primarily to assess if clozapine delivers a significant improvement over other antipsychotics when combined with ECT. DESIGN: Major electronic databases were searched between 1990 and March 2017 for trials measuring the effects of either clozapine augmented ECT, other antipsychotic-augmented ECT, or both. After the systematic review of the data, a random-effects meta-analysis was conducted measuring the relative effect sizes of the different treatment regimens. SUBJECTS: 1179 patients in 23 studies reporting the usage of ECT augmentation with antipsychotics. A total of 95 patients were tested with clozapine, and ECT (9 studies) and 1084 patients were tested with non-clozapine antipsychotics (14 studies) such as flupenthixol, chlorpromazine, risperidone, sulpiride, olanzapine, and loxapine with concurrent ECT treatment considered for systematic review. Of these, 13 studies reported pre and post-treatment scores were included in the meta-analysis. MAIN OUTCOME MEASURES: The main outcome measure was the presence and degree of both positive and negative psychotic symptoms, as measured by either of two standardized clinician administered tests, the Brief Psychiatric Rating Scale (BPRS), and the Positive and Negative Symptom Scale (PANSS). RESULTS: The comparison of the different antipsychotics established the supremacy of ECT-augmented clozapine treatment against other typical and atypical antipsychotics. The Forest Plot revealed that the overall standard mean difference was 0.891 for non-clozapine studies and 1.504 for clozapine studies, at a 95% interval. Furthermore, the heterogeneity plots showed that while clozapine studies showed no significant heterogeneity, non-clozapine studies showed an I2 statistic value at 42.19%, suggesting moderate heterogeneity. Lastly, publication bias showed asymmetrical plots and significant values of Kendal's tau and Egger's rank test. CONCLUSION: ECT augmentation technique was found to be effective in the reduction of psychometric scale scores, and the resulting improvement was significant. Clozapine maintained its stance as the most effective treatment for Treatment-Resistant Schizophrenia, followed by flupenthixol.