External Images Research Articles

Learning Curvelet-based Directional Dictionaries for Single Image Super Resolution

Learning and reconstruction-based methods are the two main approaches to the solve single image super resolution (SISR) problem. In this paper, to exploit the advantages of both learning based and reconstruction based approaches, we propose a new SISR framework by combining them, which can effectively utilize their benefits. The external directional dictionaries (EDD) are learned from external high quality images. Additionally, we embeded the nonlocal means (NLM) filter and an isotropic total variation (TV) scheme in the reconstruction based method. We suggest a new supervised clustering scheme via curvelet based direction extraction method (CCDE) to learn the external directional dictionaries from candidate patches with sharp edges. Each input patch is coded by all the EDD. Each of the reconstructed patches under different EDD is applied with a weighted penalty to characterize the given input patch. To disclose new details, the local smoothness and nonlocal self-similarity priors are added on the recovered patch by TV scheme and NLM filter. Extensive experimental results validate the effectiveness and robustness of the proposed method comparing with the state-of- the-art algorithms in SISR methods. Our proposed schemes can retrieve more fine structures and obtain superior results than the competing methods with the scaling factors of 2 and 3.

Building a model of image formation of a higher education institution in Ukraine

The article determines that to create a positive image, higher education institutions need to have a clear idea of what image they need, in this regard, the model of image formation is determined. The concept of the model of image formation of a higher education institution, which is a system of interconnected components of internal and external images of higher education, as well as the means of their formation and implementation to match the image markers of the audience that has a direct (static) impact on performance HEI for a long time. To create a model of image formation of the Ukrainian institution of higher education, three models were presented and characterized: model of the process of image formation of A. Rotovsky; model of corporate image formation B. Ji; model of image formation of I. Vikentyev. As a result of the analysis of the specified models the general recommendations on their improvement are presented.
 The expectations of first-year higher education students of Ukrainian higher education institutions from the provided educational services in the 2017—2018 and 2018—2019 academic years were monitored. It was found that the main features that applicants paid attention to when choosing a higher education institution are: the quality of educational services with the necessary specialties, the possibility of internships abroad, employment with high wages, as well as the presence of well-known graduates and teachers. The most popular sources of information about the activities of higher education institutions for higher education seekers (acquaintances (relatives) studying at universities and official representations of the institution on the Internet (official website, social networks) are identified. A six-component model of image formation of a Ukrainian higher education institution is schematically presented; possible options for determining the expected effect (calculation of efficiency) of the implementation of image strategies in the activities of higher education institutions are considered.

Cascaded deep learning-based auto-segmentation for head and neck cancer patients: Organs at risk on T2-weighted magnetic resonance imaging.

To investigate multiple deep learning methods for automated segmentation (auto-segmentation) of the parotid glands, submandibular glands, and level II and level III lymph nodes on magnetic resonance imaging (MRI). Outlining radiosensitive organs on images used to assist radiation therapy (radiotherapy) of patients with head and neck cancer (HNC) is a time-consuming task, in which variability between observers may directly impact on patient treatment outcomes. Auto-segmentation on computed tomography imaging has been shown to result in significant time reductions and more consistent outlines of the organs at risk. Three convolutional neural network (CNN)-based auto-segmentation architectures were developed using manual segmentations and T2-weighted MRI images provided from the American Association of Physicists in Medicine (AAPM) radiotherapy MRI auto-contouring (RT-MAC) challenge dataset (n=31). Auto-segmentation performance was evaluated with segmentation similarity and surface distance metrics on the RT-MAC dataset with institutional manual segmentations (n=10). The generalizability of the auto-segmentation methods was assessed on an institutional MRI dataset (n=10). Auto-segmentation performance on the RT-MAC images with institutional segmentations was higher than previously reported MRI methods for the parotid glands (Dice: 0.860 ± 0.067, mean surface distance [MSD]: 1.33 ± 0.40mm) and the first report of MRI performance for submandibular glands (Dice: 0.830 ± 0.032, MSD: 1.16 ± 0.47mm). We demonstrate that high-resolution auto-segmentations with improved geometric accuracy can be generated for the parotid and submandibular glands by cascading a localizer CNN and a cropped high-resolution CNN. Improved MSDs were observed between automatic and manual segmentations of the submandibular glands when a low-resolution auto-segmentation was used as prior knowledge in the second-stage CNN. Reduced auto-segmentation performance was observed on our institutional MRI dataset when trained on external RT-MAC images; only the parotid gland auto-segmentations were considered clinically feasible for manual correction (Dice: 0.775 ± 0.105, MSD: 1.20 ± 0.60mm). This work demonstrates that CNNs are a suitable method to auto-segment the parotid and submandibular glands on MRI images of patients with HNC, and that cascaded CNNs can generate high-resolution segmentations with improved geometric accuracy. Deep learning methods may be suitable for auto-segmentation of the parotid glands on T2-weighted MRI images from different scanners, but further work is required to improve the performance and generalizability of these methods for auto-segmentation of the submandibular glands and lymph nodes.

건축공간의 형상과 경계의 변화에 관한 연구

This study macroscopically examines the modification in architectural space and attempts to systematically suggest what kind of process the contemporary architectural space has developed in terms of shape and boundary. Furthermore, in order to analyze the categories of architectural systems that stimulated and induced changes in the shape and boundaries of architectural spaces in terms of architectural structures, forms, materials, experiences, design tools, various theories, and artistic needs, the flow of change in the shape of space is presented in diagrams. Through important examples, the geometrical shape and effect of architectural space were presented. The case was presented as a plan, elevation, cross-section, and internal and external images, and the change of space was visually refined. The shape and boundary change of these spaces were integrated into a conceptual language and diagrammed. Finally, the modern architectural space was reconstructed through heterogeneity, diversity, and dynamism, and it developed into the trends of folding, landscape, and metamorphosis, which is an effort to define space organically, and the value of space through communication with nature and living energy.

Work life quality of organisation's employees as a condition for establishing an internal image of the organisation

One of the most important tasks of corporate governance is to create a positive image of their organisation both in the external environment and in-house. In recent years, a large number of studies aimed at the study and assessment of external corporate images is complemented by studies of internal corporate image, as only their simultaneous improvement can achieve a high competitiveness level of an organisation in the external market and reach stability and high potential for the development of the organisation itself. One of the most important factors used in the formation of a corporate internal image is employee satisfaction with the quality of their work life, namely, working conditions, motivation system, the prevailing social and psychological climate, etc. All these internal image factors are the indicators of work life quality and are reflected in corporate loyalty of employees, which to a great extent is a key indicator of company's internal image. The presented paper is devoted to sociological analysis reviewing the impact caused by work life quality of employees on the formation and maintenance of the internal image their companies have, based on the author's specific sociological research conducted with the use of quantitative analysis methods.

Self-Supervised Learning-Based High-Resolution Ultrasound Imaging for Prostate Brachytherapy

Ultrasound (US) imaging has been widely used in image-guided needle/seed placement in prostate brachytherapy. For 3D US images with large slice thickness, high frequency information in the slice direction is missing and cannot be resolved through interpolation. As an ill-posed problem, most high-resolution generation methods rely on the presence of external/training atlases to learn the transform from low resolution to high resolution images. The anatomical variations among individuals bring unavoidable uncertainties in this atlas-based learning mechanism. In this study, we propose a self-supervised learning method, which does not use any external high-resolution atlas images, yet can resolve high resolution images from the 3D images with a large slice thickness.In 3D US imaging, in-plane spatial resolution is generally much higher than through-plane resolution. Considering the natural isotropy of US imaging on biological tissue, for a certain patient, the mapping from low-resolution to high-resolution images can be established by learning the mapping from down-sampled in-plane images (low-resolution) to original in-plane US images (high-resolution), which enables the generation of high-resolution through-plane images. Two independent cycle consistent generative adversarial networks (CycleGANs) are trained with paired original in-plane US images and two sets of in-plane down-sampled US images. Finally, high-resolution 3D US images are reconstructed by combining the generated 2D images obtained via feeding through-plane images into the two CycleGAN models. A validation study was conducted to quantitatively assess the proposed method using 3D high in-/through-plane resolution breast US images of 50 breast cancer patients. The feasibility of the proposed method for obtaining high-resolution US images for prostate brachytherapy was tested using 3D US images with 2-mm slice thickness from 45 prostate cancer patients.In the breast US images validation test using a three-times spatial resolution enhancement, the proposed method achieved the mean absolute error value of 0.90 (range 0.57-1.35), the peak signal-to-noise ratio value of 37.88 (range 35.84-40.48) dB, and the visual information fidelity value of 0.69 (range 0.68-0.72), which significantly outperforms bicubic interpolation. For the prostate cases, the proposed method outperformed the bicubic interpolation for the spatial resolution enhancement factors of 5 and 10.A novel deep learning-based algorithm for reconstructing high-resolution 3D US images from sparely acquired 2D images was developed and validated. Significant improvement on through-plane resolution has been achieved by using the acquired 2D images without any external atlas images. Its self-supervision capability could accelerate high-resolution US imaging for prostate brachytherapy.

Deep learning can predict lymph node status directly from histology in colorectal cancer

BackgroundLymph node status is a prognostic marker and strongly influences therapeutic decisions in colorectal cancer (CRC). ObjectivesThe objective of the study is to investigate whether image features extracted by a deep learning model from routine histological slides and/or clinical data can be used to predict CRC lymph node metastasis (LNM). MethodsUsing histological whole slide images (WSIs) of primary tumours of 2431 patients in the DACHS cohort, we trained a convolutional neural network to predict LNM. In parallel, we used clinical data derived from the same cases in logistic regression analyses. Subsequently, the slide-based artificial intelligence predictor (SBAIP) score was included in the regression. WSIs and data from 582 patients of the TCGA cohort were used as the external test set. ResultsOn the internal test set, the SBAIP achieved an area under receiver operating characteristic (AUROC) of 71.0%, the clinical classifier achieved an AUROC of 67.0% and a combination of the two classifiers yielded an improvement to 74.1%. Whereas the clinical classifier's performance remained stable on the TCGA set, performance of the SBAIP dropped to an AUROC of 61.2%. Performance of the clinical classifier depended strongly on the T stage. ConclusionDeep learning–based image analysis may help predict LNM of patients with CRC using routine histological slides. Combination with clinical data such as T stage might be useful. Strategies to increase performance of the SBAIP on external images should be investigated.

The Influence of External Images and Service Quality on Customer Satisfaction and Revisit Intention in Beauty Service Workers

This study attempted to investigate the influence of external images and service quality on customer satisfaction and revisit intention among beauty service workers and use the results as basic data for the growth of the beauty service industry by maximizing customer satisfaction. The results found the following: In terms of relationships between external images and service quality factors, external images revealed a positive correlation with ‘kindness,’ ‘activeness,’ ‘professionalism’ and ‘accuracy’ with statistical significance (p<.001). They also showed a positive correlation with ‘customer satisfaction’ and ‘revisit intention’ (p<.001) with statistical significance. Furthermore, ‘kindness,’ ‘activeness,’ ‘professionalism’ and ‘accuracy’ revealed a positive correlation with ‘customer satisfaction’ and ‘revisit intention’ with statistical significance (p<.001). As external images improved, customer satisfaction increased. Also, as ‘kindness,’ ‘activeness,’ ‘professionalism,’ and ‘accurateness’ improved, customer satisfaction increased. Furthermore, with improvements in external images, revisit intention also increased. As ‘kindness,’ ‘activeness,’ ‘professionalism’ and ‘accurateness’ were increased, revisit intention also increased. When customer satisfaction improved, revisit intention also increased.

China’s Image in Russia: Historical Background

Studying the cultural specificity of the neighboring countries is an important research question for philosophical anthropology and philosophy of culture as it gives a perspective of new possibilities of interaction and partnership in various areas. People’s understanding of a different culture originates from a complex interrelation of consciously chosen strategies and unconsciously projected feelings. The resulting image of the Other culture often plays a significant role in policy-making and helps determine the quality and extent of interrelations, including cooperation, both at the governmental and non-governmental levels. There are multiple ways of constructing such an image. Intensified contacts, that include direct interaction, diplomatic and scientific contacts, lead to the increase of knowledge about neighboring countries’ lifestyle, household organization, and traditions. Art is one of the key means of getting to know another culture. The author pays special attention to visual images, namely political caricature, as interesting yet poorly explored ways of forming the image of China in Russian society’s perception. Herein political caricature, i.e. political visual satire, not only translates the atmosphere of those times but also greatly affects the audience and influences public opinion by creating strong and vivid impressions. Both external and internal images of the self-perception of the Chinese played their role in the development of the perception of China. Their comparative analysis proves that these images underwent a certain evolution in Russia: the image of the neighboring state dynamically altered over the years. Emotional and affective evaluations of China several times drastically changed from positive to negative, and back. The paper explores peculiar features of the political caricature of the 20th century and looks into the ways the imagery shows the Chinese in the light of major political events. It is noted that the images represent both Chinese and Russian historical and cultural context of the epoch. During the period under consideration, the Soviet way of thinking about China mainly came from the logical approach of seeing China as the Other. At the same time, this period is characterized by an increased interest in universal human values of freedom and political independence, the unity of the working population, and the positive outlook on the workpeople’s future. The Chinese middle kingdom in the vein of proletarian internationalism appears a leader fighting for the above-mentioned values. This friendly approach towards China and the possibility of a partnership between the countries remains unquestioned till the 1950s.

Self-supervised learning for accelerated 3D high-resolution ultrasound imaging.

Ultrasound (US) imaging has been widely used in diagnosis, image-guided intervention, and therapy, where high-quality three-dimensional (3D) images are highly desired from sparsely acquired two-dimensional (2D) images. This study aims to develop a deep learning-based algorithm to reconstruct high-resolution (HR) 3D US images only reliant on the acquired sparsely distributed 2D images. We propose a self-supervised learning framework using cycle-consistent generative adversarial network (cycleGAN), where two independent cycleGAN models are trained with paired original US images and two sets of low-resolution (LR) US images, respectively. The two sets of LR US images are obtained through down-sampling the original US images along the two axes, respectively. In US imaging, in-plane spatial resolution is generally much higher than through-plane resolution. By learning the mapping from down-sampled in-plane LR images to original HR US images, cycleGAN can generate through-plane HR images from original sparely distributed 2D images. Finally, HR 3D US images are reconstructed by combining the generated 2D images from the two cycleGAN models. The proposed method was assessed on two different datasets. One is automatic breast ultrasound (ABUS) images from 70 breast cancer patients, the other is collected from 45 prostate cancer patients. By applying a spatial resolution enhancement factor of 3 to the breast cases, our proposed method achieved the mean absolute error (MAE) value of 0.90±0.15, the peak signal-to-noise ratio (PSNR) value of 37.88±0.88dB, and the visual information fidelity (VIF) value of 0.69±0.01, which significantly outperforms bicubic interpolation. Similar performances have been achieved using the enhancement factor of 5 in these breast cases and using the enhancement factors of 5 and 10 in the prostate cases. We have proposed and investigated a new deep learning-based algorithm for reconstructing HR 3D US images from sparely acquired 2D images. Significant improvement on through-plane resolution has been achieved by only using the acquired 2D images without any external atlas images. Its self-supervision capability could accelerate HR US imaging.

지역축제 참가자의 외적이미지에 대한 인식 차이의 연구: 거주지와 연령에 따른 인식 차이를 중심으로

This study studied the differences in perception of external images of local festival participants. Based on the results of this study, research was conducted to help organize or plan festivals. Studies have shown that there is no difference in perception of external images depending on their residence. However, there are differences in perception of external images depending on age. This study has three implications. First, the general idea of exclusive and conservative local residents is that there is no difference in perception with urban people. Second, the older the elderly, the older the school uniform generation is, so there is a limit to understanding the diversity of appearance expressions. Also, young people tend to think horizontally and not understand being harmed. Therefore, it is possible to enhance the satisfaction of the festival participants by reflecting this on the planning and spatial composition of the festival. Third, when planning festival programs based on research results, it is necessary to organize festival spaces and plan programs according to age groups. This is because it is more important to reduce the contact point of conflict through separation of space and participation events than to ask groups with different ideas to understand others.

RetinexDIP: A Unified Deep Framework for Low-Light Image Enhancement

Low-light images suffer from low contrast and unclear details, which not only reduces the available information for humans but limits the application of computer vision algorithms. Among the existing enhancement techniques, Retinex-based and learning-based methods are under the spotlight today. In this paper, we bridge the gap between the two methods. First, we propose a novel &#x201C;generative&#x201D; strategy for Retinex decomposition, by which the decomposition is cast as a generative problem. Second, based on the strategy, a unified deep framework is proposed to estimate the latent components and perform low-light image enhancement. Third, our method can weaken the coupling relationship between the two components while performing Retinex decomposition. Finally, the RetinexDIP performs Retinex decomposition without any external images, and the estimated illumination can be easily adjusted and is used to perform enhancement. The proposed method is compared with ten state-of-the-art algorithms on seven public datasets, and the experimental results demonstrate the superiority of our method. Code is available at: <uri>https://github.com/zhaozunjin/RetinexDIP</uri>.

Automated Image Denoising Model: Contribution Towards Optimized Internal and External Basis

For handling digital images for various applications, image denoising is considered as a fundamental pre-processing step. Diverse image denoising algorithms have been introduced in the past few decades. The main intent of this proposal is to develop an effective image denoising model on the basis of internal and external patches. This model adopts Non-local means (NLM) for performing the denoising, which uses redundant information of the image in pixel or spatial domain to reduce the noise. While performing the image denoising using NLM, “denoising an image patch using the other noisy patches within the noisy image is done for internal denoising and denoising a patch using the external clean natural patches is done for external denoising”. Here, the selection of optimal block from the entire datasets including internal noisy images and external clean natural images is decided by a new hybrid optimization algorithm. The two renowned optimization algorithms Chicken Swarm Optimization (CSO), and Dragon Fly Algorithm (DA) are merged, and the new hybrid algorithm Rooster-based Levy Updated DA (RLU-DA) is adopted. The experimental results in terms of some relevant performance measures show the promising results of the proposed model with remarkable stability and high accuracy.

Generic frames (Climate and Environment Coverage)

Generic frames represent typical layers of contextualization in stories and are broadly applicable to a range of different news topics (i.e., across topics). Examples of generic frames are the human interest or responsibility frames (e.g., Dirikx &amp; Gelders, 2010; Semetko &amp; Valkenburg, 2000). Field of application/theoretical foundation: Generic frames are used in different traditions of framing: (a) sociological tradition, such as frames in external news or images (e.g., Entman, 1993; Gamson &amp; Modigliani, 1989), (b) psychological tradition, such as frames in people's minds (e.g., Tversky &amp; Kahneman, 1981), and (3) communication science tradition, such as frame production (communicators or journalists develop frames), frame setting or frame building (journalists adopt frames from communicators) (e.g., Matthes, 2014; see Borah, 2011 for a systematic examination of framing research). References/combination with other methods of data collection: Research has conducted experimental studies to investigate how generic frames (e.g., conflict, morality or economics) affect selective exposure to climate change news (e.g., Feldman &amp; Hart, 2018) or what impact gain and loss frames have on the perception of the threat posed by the consequences of climate change (Bilandzic et al., 2017). Example studies: Bilandzic et al. (2017); Dirikx &amp; Gelders (2010); Feldman &amp; Hart (2018) Information on Dirikx &amp; Gelders (2010) Authors: Astrid Dirikx &amp; Dave Gelders Research question: This study examines the way Dutch and French newspapers frame climate change during the annual United Nations Conferences of the Parties (COPs) Object of analysis: The study analyzed a total of 257 news articles in Dutch and French quality newspapers: De Volkskrant (N = 52) and NRC Handelsblad (N = 61) for the Netherlands and Le Monde (N = 77) and Le Figaro (N = 67) for France. Time frame of analysis: The analysis covers the annual meetings of the United Nations Conferences of the Parties (COPs) from 2001 until 2007 Info about variable Variables: Frames Attribution of responsibility: This frame presents an issue or problem in such a way that the responsibility or blame for the cause or the solution is placed on political authorities, individuals or groups Conflict frame: This frame emphasizes conflicts between parties/individuals and stresses the points of divergence between the opponents (Economic) consequences frame: This frame emphasizes the manner in which an issue will (economically) affect people Human interest frame: This frame presents an issue from a more emotional point of view; it personalizes a problem Morality frame: This frame presents situations from a religious/moral angle Level of analysis: Each of the four newspapers was first screened entirely for articles mentioning “climate change,” “global warming” or “greenhouse effect” in their title or lead. In an additional screening articles not matching these criteria but whose core content was related to climate change were selected. Different article types were taken into account: columns, interviews, commentaries and “classic” news articles. The unit of analysis was the whole article. Variables and values: The authors used a standard set of content analytic indicators to measure the prevalence of the five generic frames developed by Semetko and Valkenburg (2000). They developed a series of 20 questions that can be answered with “don’t agree”, “largely agree” and “completely agree”. Each question was meant to measure one of five news media frames. Attribution of responsibility Does the story suggest that some level of government has the ability to alleviate the problem? Does the story suggest that some level of the government is responsible for the issue/problem? Does the story suggest solution(s) to the problem/issue? Does the story suggest that an individual (or group of people in society) is responsible for the issue-problem? Does the story suggest the problem requires urgent action? Human interest frame Does the story provide a human example or “human face” on the issue? Does the story employ adjectives or personal vignettes that generate feelings of outrage, empathy-caring, sympathy, or compassion? Does the story emphasize how individuals and groups are affected by the issue/problem? Does the story go into the private or personal lives of the actors? Does the story contain visual information that might generate feelings of outrage, empathy-caring, sympathy, or compassion? Conflict frame Does the story reflect disagreement between parties/individuals/groups/countries? Does one party/individual/group/country reproach another? Does the story refer to two sides or to more than two sides of the problem or issue? Does the story refer to winners and losers? Morality frame Does the story contain any moral message? Does the story make reference to morality, God, and other religious tenets? Does the story offer specific social prescriptions about how to behave? (Economic) consequences frame Is there a mention of financial losses or gains now or in the future? Is there a mention of the costs/degree of expense involved? Is there a reference to economic consequences of pursuing or not pursuing a course of action? Reliability: - Codebook: Table 1 in Dirikx &amp; Gelders (2010)

Quantitative comparison of data-driven gating and external hardware gating for 18F-FDG PET-MRI in patients with esophageal tumors

BackgroundRespiratory motion during PET imaging reduces image quality. Data-driven gating (DDG) based on principal component analysis (PCA) can be used to identify respiratory signals. The use of DDG, without need for external devices, would greatly increase the feasibility of using respiratory gating in a routine clinical setting. The objective of this study was to evaluate data-driven gating in relation to external hardware gating and regular static image acquisition on PET-MRI data with respect to SUVmax and lesion volumes.MethodsSixteen patients with esophageal or gastroesophageal cancer (Siewert I and II) underwent a 6-min PET scan on a Signa PET-MRI system (GE Healthcare) 1.5–2 h after injection of 4 MBq/kg 18F-FDG. External hardware gating was done using a respiratory bellow device, and DDG was performed using MotionFree (GE Healthcare). The DDG raw data files and the external hardware-gating raw files were created on a Matlab-based toolbox from the whole 6-min scan LIST-file. For comparison, two 3-min static raw files were created for each patient. Images were reconstructed using TF-OSEM with resolution recovery with 2 iterations, 28 subsets, and 3-mm post filter. SUVmax and lesion volume were measured in all visible lesions, and noise level was measured in the liver. Paired t-test, linear regression, Pearson correlation, and Bland-Altman analysis were used to investigate difference, correlation, and agreement between the methods.ResultsA total number of 30 lesions were included in the study. No significant differences between DDG and external hardware-gating SUVmax or lesion volumes were found, but the noise level was significantly reduced in the DDG images. Both DDG and external hardware gating demonstrated significantly higher SUVmax (9.4% for DDG, 10.3% for external hardware gating) and smaller lesion volume (− 5.4% for DDG, − 6.6% for external gating) in comparison with non-gated static images.ConclusionsData-driven gating with MotionFree for PET-MRI performed similar to external device gating for esophageal lesions with respect to SUVmax and lesion volume. Both gating methods significantly increased the SUVmax and reduced the lesion volume in comparison with non-gated static acquisition. DDG resulted in reduced image noise compared to external device gating and static images.

Fooling the size\u2013weight illusion\u2014Using augmented reality to eliminate the effect of size on perceptions of heaviness and sensorimotor prediction

Augmented reality, whereby computer-generated images are overlaid onto the physical environment, is becoming significant part of the world of education and training. Little is known, however, about how these external images are treated by the sensorimotor system of the user – are they fully integrated into the external environmental cues, or largely ignored by low-level perceptual and motor processes? Here, we examined this question in the context of the size–weight illusion (SWI). Thirty-two participants repeatedly lifted and reported the heaviness of two cubes of unequal volume but equal mass in alternation. Half of the participants saw semi-transparent equally sized holographic cubes superimposed onto the physical cubes through a head-mounted display. Fingertip force rates were measured prior to lift-off to determine how the holograms influenced sensorimotor prediction, while verbal reports of heaviness after each lift indicated how the holographic size cues influenced the SWI. As expected, participants who lifted without augmented visual cues lifted the large object at a higher rate of force than the small object on early lifts and experienced a robust SWI across all trials. In contrast, participants who lifted the (apparently equal-sized) augmented cubes used similar force rates for each object. Furthermore, they experienced no SWI during the first lifts of the objects, with a SWI developing over repeated trials. These results indicate that holographic cues initially dominate physical cues and cognitive knowledge, but are dismissed when conflicting with cues from other senses.

Abstract PO-003: Deep learning identifies conserved pan-cancer tumor features

Abstract Histopathological images are an integral data type for studying cancer. We show pre-trained convolutional neural networks (CNNs) can be systematically applied across cancer types, enabling comparisons to reveal shared spatial behaviors. We develop CNNs with a common architecture trained on 19 cancer types of The Cancer Genome Atlas (TCGA), analyzing 14459 hematoxylin and eosin scanned frozen tissue images. Our CNNs are based on the Inception-V3 network and classify TCGA pathologist-annotated tumor/normal status of whole slide images in all 19 cancer types with consistently high AUCs (0.995±0.008). Remarkably, CNNs trained on one tissue are effective in others (AUC 0.88±0.11), with classifier relationships recapitulating known adenocarcinoma, carcinoma, and developmental biology. Moreover, classifier comparisons reveal intra-slide spatial similarities, with an average tile-level correlation of 0.45±0.16 between classifier pairs on the TCGA test sets. In particular, the TCGA-trained classifiers had average tile-level correlation of 0.52±0.09 and 0.58±0.08 on hold-out TCGA lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) test sets, respectively. These relations are reflected on two external datasets, i.e., LUAD and LUSC whole slide images of Clinical Proteomic Tumor Analysis Consortium. The CNNs trained on TCGA achieved cross-classification AUCs of 0.75±0.12 and 0.73±0.13 on LUAD and LUSC external validation sets, respectively. These CNNs had average tile-level correlations of 0.38±0.09 and 0.39±0.08 on LUAD and LUSC validation sets, respectively. Breast cancers, bladder cancers, and uterine cancers have spatial patterns that are particularly easy to detect, suggesting these cancers can be canonical types for image analysis. This study illustrates pre-trained CNNs can detect tumor features across a wide range of cancers, suggesting presence of pan-cancer tumor features. These shared features allow combining datasets when analyzing small samples to narrow down the parameter search space of CNN models. Citation Format: Javad Noorbakhsh, Saman Farahmand, Ali Foroughi pour, Sandeep Namburi, Dennis Caruana, David Rimm, Mohammad Soltanieh-ha, Kourosh Zarringhalam, Jeffrey H. Chuang. Deep learning identifies conserved pan-cancer tumor features [abstract]. In: Proceedings of the AACR Virtual Special Conference on Artificial Intelligence, Diagnosis, and Imaging; 2021 Jan 13-14. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(5_Suppl):Abstract nr PO-003.

Who designs the classroom’s images? Study of visual culture diversity at three Valencian schools

The research reported on here elaborated on the effects that images displayed on the walls have in schools as a variable in evaluating the educational quality. Mixed methods were used to analyse the images displayed on the walls of 27 classrooms to determine their provenance and authorship. The results show quantitative, qualitative and aesthetical values of the school visual culture, thanks to the numerous photographic series (resembling in graphs) which are a testimony of each classroom. After the typologies of the displayed images at the schools had been documented, it was demonstrated that images mostly come from the school products industry and from faculty members.i The kind of analysis and the conclusions that were obtained could be considered in other contexts of schools with cultural diversity. Four alternatives are proposed in order to improve schools’ visual culture diversity: 1) consensus between the entire educative community – both adults and children selecting which images should adorn the classroom walls; 2) an increasing of external images, but generated by students and teachers about their personal experiences to increase the sense of belonging in the classroom; 3) a greater presence of internal images created by specialised agents, that enrich the learning processes, diversity, iconography and style of the images; 4) the emphasis of elementary teachers’ visual education to integrate images pedagogically.

Reassessing External Images of the EU: Evolving Narratives in Times of Crisis

The past decade has challenged the EU and its international image. The Eurozone sovereign debt crisis, the Ukraine crisis, the so-called irregular migration crisis and the Covid-19 pandemic have all put the EU under severe strain. This article explores if and how the EU’s performance in such crises has impacted upon the external image of the EU. The analysis shows that external images have closely followed the EU’s actual performance, although filtering it through the powerful lenses of local and regional concerns and sensibilities. While some traditional images have proved to be resilient in the longer run (as in the case of the EU as an economic powerhouse or a frequently divided community), others have been severely weakened by the EU’s crisis responses (such as the EU as a bastion of human rights). Our findings contribute to the discussion on the public diplomacy and information strategy of the European External Action Service (EEAS) in shaping locally-resonating positive images of the EU worldwide. external images of the EU, multiple crises of the EU, EU public diplomacy

Artificial intelligence for automatic cerebral ventricle segmentation and volume calculation: a clinical tool for the evaluation of pediatric hydrocephalus.

Imaging evaluation of the cerebral ventricles is important for clinical decision-making in pediatric hydrocephalus. Although quantitative measurements of ventricular size, over time, can facilitate objective comparison, automated tools for calculating ventricular volume are not structured for clinical use. The authors aimed to develop a fully automated deep learning (DL) model for pediatric cerebral ventricle segmentation and volume calculation for widespread clinical implementation across multiple hospitals. The study cohort consisted of 200 children with obstructive hydrocephalus from four pediatric hospitals, along with 199 controls. Manual ventricle segmentation and volume calculation values served as "ground truth" data. An encoder-decoder convolutional neural network architecture, in which T2-weighted MR images were used as input, automatically delineated the ventricles and output volumetric measurements. On a held-out test set, segmentation accuracy was assessed using the Dice similarity coefficient (0 to 1) and volume calculation was assessed using linear regression. Model generalizability was evaluated on an external MRI data set from a fifth hospital. The DL model performance was compared against FreeSurfer research segmentation software. Model segmentation performed with an overall Dice score of 0.901 (0.946 in hydrocephalus, 0.856 in controls). The model generalized to external MR images from a fifth pediatric hospital with a Dice score of 0.926. The model was more accurate than FreeSurfer, with faster operating times (1.48 seconds per scan). The authors present a DL model for automatic ventricle segmentation and volume calculation that is more accurate and rapid than currently available methods. With near-immediate volumetric output and reliable performance across institutional scanner types, this model can be adapted to the real-time clinical evaluation of hydrocephalus and improve clinician workflow.