Measuring the Sciences Contribution to Socioeconomic Development: Russian Approaches in the Global Space of Ideas

Purpose: this review aims to present the ‘state-of-the-art’ on the theory and practice of measuring the contribution of sciences to socioeconomic progress and trace the Russian approaches in the global space of ideas.Methods: for studying a multidisciplinary sample of academic publications and gray literature includes multifactor systematization, critical analysis, and synthetic generalization in a large context («helicopter view»).Results: a brief history of the subject is presented in the context of the state's use of science for socio-economic development. The review outlines the focus areas of scientific thought, the conceptual frameworks state, current approaches to measuring the contribution of sciences, the limitations and risks of the evaluation practice.Conclusions and Relevance: key areas of scientific thought in the subject under review for 100 years: (1) development of econometric approaches for measuring the «profitability» of science; (2) attempts to identify and measure the societal impacts of science; (3) scientific support of government experiments to use the evaluation technologies for increasing the return of budget investment to science; (4) conceptualizing and universalizing the terms. There are no indisputable solutions in any of these areas. Disappointment with the catch-all indicators and universal metrics encourages the development of case-oriented methods and process-oriented approaches.Russian research covers the full range of issues, but their world recognition (except some achievements of Soviet time) is low. The lack of institutional memory leads to the rediscovery of the ideas of the 20th century. Establishing the equivalence of terms and conceptual approaches used by scientists from different countries and periods will help to effectively use the scientific heritage, avoid duplicate research, provide a fair place to the Russian achievements in world science.

Medical Image Segmentation using Squeeze-and-Expansion Transformers

Medical image segmentation is important for computer-aided diagnosis. Good segmentation demands the model to see the big picture and fine details simultaneously, i.e., to learn image features that incorporate large context while keep high spatial resolutions. To approach this goal, the most widely used methods -- U-Net and variants, extract and fuse multi-scale features. However, the fused features still have small "effective receptive fields" with a focus on local image cues, limiting their performance. In this work, we propose Segtran, an alternative segmentation framework based on transformers, which have unlimited "effective receptive fields" even at high feature resolutions. The core of Segtran is a novel Squeeze-and-Expansion transformer: a squeezed attention block regularizes the self attention of transformers, and an expansion block learns diversified representations. Additionally, we propose a new positional encoding scheme for transformers, imposing a continuity inductive bias for images. Experiments were performed on 2D and 3D medical image segmentation tasks: optic disc/cup segmentation in fundus images (REFUGE'20 challenge), polyp segmentation in colonoscopy images, and brain tumor segmentation in MRI scans (BraTS'19 challenge). Compared with representative existing methods, Segtran consistently achieved the highest segmentation accuracy, and exhibited good cross-domain generalization capabilities.

RECONSTRUCTED TEETH IMAGE FROM BRACES WITH GAN

Recent advances in deep learning models have shown promising potential in object removal, which is the task of replacing undesired objects with appropriate pixel values using known context. Object removal-based deep learning can commonly be solved by modeling it as the Img2Img (image to image) translation or Inpainting. Instead of dealing with a large context, this paper aims at a specific application of object removal, that is, erasing braces trace out of an image having teeth with braces (called braces2teeth problem). We solved the problem by three methods corresponding to different datasets. Firstly, we use the CycleGAN model to deal with the problem that paired training data is not available. In the second case, we try to create pseudo-paired data to train the Pix2Pix model. In the last case, we utilize GraphCut combining generative inpainting model to build a user-interactive tool that can improve the result in case the user is not satisfied with previous results. To our best knowledge, this study is one of the first attempts to take the braces2teeth problem into account by using deep learning techniques and it can be applied in various fields, from health care to entertainment.

Fungal metabarcoding data integration framework for the MycoDiversity DataBase (MDDB)

Fungi have crucial roles in ecosystems, and are important associates for many organisms. They are adapted to a wide variety of habitats, however their global distribution and diversity remains poorly documented. The exponential growth of DNA barcode information retrieved from the environment is assisting considerably the traditional ways for unraveling fungal diversity and detection. The raw DNA data in association to environmental descriptors of metabarcoding studies are made available in public sequence read archives. While this is potentially a valuable source of information for the investigation of Fungi across diverse environmental conditions, the annotation used to describe environment is heterogenous. Moreover, a uniform processing pipeline still needs to be applied to the available raw DNA data. Hence, a comprehensive framework to analyses these data in a large context is still lacking. We introduce the MycoDiversity DataBase, a database which includes public fungal metabarcoding data of environmental samples for the study of biodiversity patterns of Fungi. The framework we propose will contribute to our understanding of fungal biodiversity and aims to become a valuable source for large-scale analyses of patterns in space and time, in addition to assisting evolutionary and ecological research on Fungi.

Accelerated EM Connectome Reconstruction using 3D Visualization and Segmentation Graphs

Recent advances in automatic image segmentation and synapse prediction in electron microscopy (EM) datasets of the brain enable more efficient reconstruction of neural connectivity. In these datasets, a single neuron can span thousands of images containing complex tree-like arbors with thousands of synapses. While image segmentation algorithms excel within narrow fields of views, the algorithms sometimes struggle to correctly segment large neurons, which require large context given their size and complexity. Conversely, humans are comparatively good at reasoning with large objects. In this paper, we introduce several semi-automated strategies that combine 3D visualization and machine guidance to accelerate connectome reconstruction. In particular, we introduce a strategy to quickly correct a segmentation through merging and cleaving, or splitting a segment along supervoxel boundaries, with both operations driven by affinity scores in the underlying segmentation. We deploy these algorithms as streamlined workflows in a tool called Neu3 and demonstrate superior performance compared to prior work, thus enabling efficient reconstruction of much larger datasets. The insights into proofreading from our work clarify the trade-offs to consider when tuning the parameters of image segmentation algorithms.