Automated Screening for Abdominal Aortic Aneurysm in CT Scans under Clinical Conditions Using Deep Learning

Abdominal aortic aneurysms (AAA) may remain clinically silent until they enlarge and patients present with a potentially lethal rupture. This necessitates early detection and elective treatment. The goal of this study was to develop an easy-to-train algorithm which is capable of automated AAA screening in CT scans and can be applied to an intra-hospital environment. Three deep convolutional neural networks (ResNet, VGG-16 and AlexNet) were adapted for 3D classification and applied to a dataset consisting of 187 heterogenous CT scans. The 3D ResNet outperformed both other networks. Across the five folds of the first training dataset it achieved an accuracy of 0.856 and an area under the curve (AUC) of 0.926. Subsequently, the algorithms performance was verified on a second data set containing 106 scans, where it ran fully automated and resulted in an accuracy of 0.953 and an AUC of 0.971. A layer-wise relevance propagation (LRP) made the decision process interpretable and showed that the network correctly focused on the aortic lumen. In conclusion, the deep learning-based screening proved to be robust and showed high performance even on a heterogeneous multi-center data set. Integration into hospital workflow and its effect on aneurysm management would be an exciting topic of future research.

Achieving Precision Medicine in Allergic Disease: Progress and Challenges

Allergic diseases (atopic dermatitis, food allergy, eosinophilic esophagitis, asthma and allergic rhinitis), perhaps more than many other traditionally grouped disorders, share several overlapping inflammatory pathways and risk factors, though we are still beginning to understand how the relevant patient and environmental factors uniquely shape each disease. Precision medicine is the concept of applying multiple levels of patient-specific data to tailor diagnoses and available treatments to the individual; ideally, a patient receives the right intervention at the right time, in order to maximize effectiveness but minimize morbidity, mortality and cost. While precision medicine in allergy is in its infancy, the recent success of biologics, development of tools focused on large data set integration and improved sampling methods are encouraging and demonstrates the utility of refining our understanding of allergic endotypes to improve therapies. Some of the biggest challenges to achieving precision medicine in allergy are characterizing allergic endotypes, understanding allergic multimorbidity relationships, contextualizing the impact of environmental exposures (the “exposome”) and ancestry/genetic risks, achieving actionable multi-omics integration, and using this information to develop adequately powered patient cohorts and refined clinical trials. In this paper, we highlight several recently developed tools and methods showing promise to realize the aspirational potential of precision medicine in allergic disease. We also outline current challenges, including exposome sampling and building the “knowledge network” with multi-omics integration.

Panoramic stitching of heterogeneous single-cell transcriptomic data

Researchers are generating single-cell RNA sequencing (scRNA-seq) profiles of diverse biological systems1–4 and every cell type in the human body.5 Leveraging this data to gain unprecedented insight into biology and disease will require assembling heterogeneous cell populations across multiple experiments, laboratories, and technologies. Although methods for scRNA-seq data integration exist6,7, they often naively merge data sets together even when the data sets have no cell types in common, leading to results that do not correspond to real biological patterns. Here we present Scanorama, inspired by algorithms for panorama stitching, that overcomes the limitations of existing methods to enable accurate, heterogeneous scRNA-seq data set integration. Our strategy identifies and merges the shared cell types among all pairs of data sets and is orders of magnitude faster than existing techniques. We use Scanorama to combine 105,476 cells from 26 diverse scRNA-seq experiments across 9 different technologies into a single comprehensive reference, demonstrating how Scanorama can be used to obtain a more complete picture of cellular function across a wide range of scRNA-seq experiments.

Conflation and Integration of Archived Geologic Maps and Associated Uncertainties

Old, archived geologic maps are often available with little or no associated metadata. This creates special problems in terms of extracting their data to use with a modern database. This research focuses on some problems and uncertainties associated with conflating older geologic maps in regions where modern geologic maps are, as yet, non-existent as well as vertically integrating the conflated maps with layers of modern GIS data (in this case, The National Map of the U.S. Geological Survey). Ste. Genevieve County, Missouri was chosen as the test area. It is covered by six archived geologic maps constructed in the years between 1928 and 1994. Conflating these maps results in a map that is internally consistent with these six maps, is digitally integrated with hydrography, elevation and orthoimagery data, and has a 95% confidence interval useful for further data set integration.