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.

Efficacy and Safety of Musculoskeletal Ultrasound Guided Aspiration and Intra-Lesional Corticosteroids Injection of Ruptured Baker's Cyst: A Retrospective Observational Study.

OBJECTIVE: The aim of the work was to demonstrate the effectiveness and safety of ultrasonographic guided aspiration performed with corticosteroid injection intra-lesional for ruptured Baker cysts (BCs). METHODS: Single-center retrospective study that included 42 patients with knee joint disorder associated with ruptured BCs were treated by ultrasonographic guided aspiration of fluid from the cyst and different points from the calf then intra-lesional injection of corticosteroids once or twice, 1 week apart. Follow up were done weekly until complete resolution of symptoms. Visual analog scale (VAS) and Rauschning-Lindgren and Lysholm Knee Scoring Scales (RLC) were used for assessment. RESULTS: Clinical parameters (VAS and RLC) improved significantly in all patients at both post injection evaluation visits (1 week and 12 weeks). Ultrasonographic features improved significantly with complete disappearance of free fluid in the calf in 35 (83.3%) cases 1 week after the injection, and in 41 (97.6%) after 12 weeks. As regards BCs only 4 (9.5%) cases showed complete disappearance after 1 week and there was recurrent BCs in 38 (90.5%) cases which required reaspiration. While after 12 weeks, BCs were completely disappeared in 23 (54.8%) cases, most of the relapsed BCs were complex BCs. No side effects were reported in all cases. CONCLUSION: Ultrasonographic guided aspiration followed by injection of corticosteroids intra-lesional is an efficient and safe method for managing ruptured BCs.