Simplified Images Research Articles

Interpretable surrogate models to approximate the predictions of convolutional neural networks in glaucoma diagnosis

Deep learning systems, especially in critical fields like medicine, suffer from a significant drawback, their black box nature, which lacks mechanisms for explaining or interpreting their decisions. In this regard, our research aims to evaluate the use of surrogate models for interpreting convolutional neural network (CNN) decisions in glaucoma diagnosis. Our approach is novel in that we approximate the original model with an interpretable one and also change the input features, replacing pixels with tabular geometric features of the optic disc, cup, and neuroretinal rim. We trained CNNs with two types of images: original images of the optic nerve head and simplified images showing only the disc and cup contours on a uniform background. Decision trees were used as surrogate models due to their simplicity and visualization properties, while saliency maps were calculated for some images for comparison. The experiments carried out with 1271 images of healthy subjects and 721 images of glaucomatous eyes demonstrate that decision trees can closely approximate the predictions of neural networks trained on simplified contour images, with R-squared values near 0.9 for VGG19, Resnet50, InceptionV3 and Xception architectures. Saliency maps proved difficult to interpret and showed inconsistent results across architectures, in contrast to the decision trees. Additionally, some decision trees trained as surrogate models outperformed a decision tree trained on the actual outcomes without surrogation. Decision trees may be a more interpretable alternative to saliency methods. Moreover, the fact that we matched the performance of a decision tree without surrogation to that obtained by decision trees using knowledge distillation from neural networks is a great advantage since decision trees are inherently interpretable. Therefore, based on our findings, we think this approach would be the most recommendable choice for specialists as a diagnostic tool.

MODERN MILITARY PROSE: GENRE AND STYLE TYPOLOGY

The paper deals with the typology of modern Ukrainian military prose based on the analysis of relevant novels. According to the structural and functional features of the works, it is reasonable to classify them using division into high/mass/intermediate literature. The first type includes complex and multidimensional texts (with allusions, literary reminiscences, irony, parody, etc.), and their action takes place mainly in a mental sphere. Heroes are characterized by understanding of social processes and war realities (significant is the presence of two identities, “properly Ukrainian” and “Soviet”); they rethink their mistakes and transform in response to new challenges. These texts are mostly focused on the acquisition of individual subjectivity. They require a trained erudite reader who can discern the writer’s narratives behind their multi-layered nature. Mass literature includes works where action dominates, and such elements as shooting, chase, superman hero, excessive emotions, love triangle, etc. are widely used. The adventure genre, military detective, melodrama, and action novel are currently most fully represented. Intermediate prose may combine, on the one hand, topical social issues, character evolution, logical and consistent presentation of events (as a feature of “high literature”) with a simplified plot and images, the lack of attention to the psychology of the characters, and various means of mass literature. Contemporary Ukrainian writers have managed to create a large-scale cultural discourse, offering works that cover the problem of war, its causes and consequences, and the topic of a person at war from different perspectives and contexts.

Selling the indigenous in Nordic welfare states: examples from Norway and Sweden

ABSTRACT Indigenous peoples’ right to control representations of their own culture and heritage is unquestionable, but in the case of tourism activities other stakeholders’ understandings come into play. The nation-state is still an important organizational foundation for tourism. For the Indigenous Sámi people, who are located in four different nation-states, national destination management organizations (DMOs) have a crucial role in how their culture and traditions are represented. The current study examines the content of Visit Norway and Visit Sweden’s visual marketing of indigenous Sámi tourism products. Using content analysis to sort electronic images and related texts, categories distinguishing natural, human, and other types of relevant symbols were created. The marketing strategies of both countries reinforce the traditional connection of the Sámi people to nature and their reindeer. Visit Sweden uses a distinct notion of what we call the artification of the Sámi, where young female artists contribute to the modern image of this indigenous people. Visit Norway continues to use more stereotypical representations of the Sámi, with a focus on colourful outfits and traditional buildings. Thus, tourism marketing continues to reinforce simplified images of the indigenous populations of the Arctic and their relation to the nation-state.

Public Opinion : Stereotypes

In this text, Walter Lippmann, essayist and columnist, explores the question of the formation of stereotypes in the minds of individuals. He examines the role of perceptions, prejudices, and past experiences in the creation of these stereotypes as well as their importance in the formation of public opinion by American citizens in the aftermath of the First World War. Stereotypes presented here as mental shortcuts allowing citizens to understand the world around them and better understand its complexity. These simplified images, confronted with the reality of events, can also be updated and lead individuals to revise their judgment as well as their understanding of events.

2-D sex images elicit mate copying in fruit flies

Although the environment is three-dimensional (3-D), humans are able to extract subtle information from two-dimensional (2-D) images, particularly in the domain of sex. However, whether animals with simpler nervous systems are capable of such information extraction remains to be demonstrated, as this ability would suggest a functional generalisation capacity. Here, we performed mate-copying experiments in Drosophila melanogaster using 2-D artificial stimuli. Mate copying occurs when naïve females observe the mating success of potential mates and use that social information to build their own mating preference. By replacing live demonstrations with (i) photos or (ii) simplified images of copulating pairs, we found that even crudely simplified images of sexual intercourse still elicit mate copying, suggesting that Drosophila is able to extract sex-related information even from a degraded image. This new method constitutes a powerful tool to further investigate mate copying in that species and sexual preferences in general.

Protests and perceptions

Protests and Perceptions: The Framing of the Modern Protester considers the role ofthe modern mass media in creating and perpetuating simplified images of newsworthyevents and people. Specifically, an in-depth analysis of New York Times_and WashingtonPost coverage of the Summit of the Americas held from April 20, 2001 to April 22, 2001in Quebec, Canada, provides examples of media framing of both Summit participantsand protesters. The analysis shows that newspaper reports on the debate are presentedthrough a pro-capitalist framework, which influences readers' perception of those whooppose globalization and free trade.

Exploiting semantic segmentation to boost reinforcement learning in video game environments

In this work we explore enhancing performance of reinforcement learning algorithms in video game environments by feeding it better, more relevant data. For this purpose, we use semantic segmentation to transform the images that would be used as input for the reinforcement learning algorithm from their original domain to a simplified semantic domain with just silhouettes and class labels instead of textures and colors, and then we train the reinforcement learning algorithm with these simplified images. We have conducted different experiments to study multiple aspects: feasibility of our proposal, and potential benefits to model generalization and transfer learning. Experiments have been performed with the Super Mario Bros video game as the testing environment. Our results show multiple advantages for this method. First, it proves that using semantic segmentation enables reaching higher performance than the baseline reinforcement learning algorithm without modifying the actual algorithm, and in fewer episodes; second, it shows noticeable performance improvements when training on multiple levels at the same time; and finally, it allows to apply transfer learning for models trained on visually different environments. We conclude that using semantic segmentation can certainly help reinforcement learning algorithms that work with visual data, by refining it. Our results also suggest that other computer vision techniques may also be beneficial for data prepossessing. Models and code will be available on github upon acceptance.

Identifying the limits of peripheral visual processing in 9-month-old infants.

Most fundamental aspects of information processing in infancy have been primarily investigated using simplified images centrally presented on computer displays. This approach lacks ecological validity as in reality the majority of visual information is presented across the visual field, over a range of eccentricities. Limited studies are present, however, about the extent and the characteristics of infant peripheral vision after 7 months of age. The present work investigates the limits of infant (9-month-olds) and adult visual fields using a detection task. Gabor patches were presented at one of six eccentricities per hemifield, from 35° up to 60° in the left and right mid-peripheral visual fields. Detection rates at different eccentricities were measured from video recordings (infant sample) or key press responses (adult sample). Infant performance declined below chance level beyond 50°, whereas adults performed at ceiling level across all eccentricities. The performance of 9-month-olds was unequal even within 50°, suggesting regions of differential sensitivity to low-level visual information in the infant's periphery. These findings are key to understanding the limits of visual fields in the infant and, in turn, will inform the design of future infant studies.

METHOD OF "SKELETAL STRUCTURES" TO SIMPLIFY THE PARAMETRIZATION PROCESS IN AUTOCAD

The article considers the method of "skeletal structures", proposed to simplify the process of parameterization. When parameterizing objects whose shape has more than one contour, there may be some complications, namely: auto-overlay dependencies, or sequential parameterization of the relative position of image fragments, can lead to an excessive number of dependencies, which can lead to inconsistencies when performing parameterization . The proposed method allows you to minimize such inconveniences, reduce the number of superimposed parameters and improve the quality of control over the shape of the object. The stages of parameterization of objects by the method of "skeletal structures" are formulated and characterized and recommendations for its use are given.
 The method of "skeletal structures" was developed and applied in the process of studying the topic "Parameterization" of the course "Engineering and Computer Graphics" at the Department of Descriptive Geometry, Engineering and Computer Graphics of the National Technical University of Ukraine "Kyiv Polytechnic Institute named after Igor Sikorsky". The application of the method on the example of flat parts of complex shape and for the structures used in the study of the topic "Assembly drawing" is demonstrated. As a condition of the task in the topic "Assembly drawing" we used a parameterized folding unit, which students adjusted to the parameters corresponding to the variant of the task, and supplemented the drawing with simplified images of fastening threaded elements made by them in the form of dynamic blocks.

ExGenNet: Learning to Generate Robotic Facial Expression Using Facial Expression Recognition.

The ability of a robot to generate appropriate facial expressions is a key aspect of perceived sociability in human-robot interaction. Yet many existing approaches rely on the use of a set of fixed, preprogrammed joint configurations for expression generation. Automating this process provides potential advantages to scale better to different robot types and various expressions. To this end, we introduce ExGenNet, a novel deep generative approach for facial expressions on humanoid robots. ExGenNets connect a generator network to reconstruct simplified facial images from robot joint configurations with a classifier network for state-of-the-art facial expression recognition. The robots’ joint configurations are optimized for various expressions by backpropagating the loss between the predicted expression and intended expression through the classification network and the generator network. To improve the transfer between human training images and images of different robots, we propose to use extracted features in the classifier as well as in the generator network. Unlike most studies on facial expression generation, ExGenNets can produce multiple configurations for each facial expression and be transferred between robots. Experimental evaluations on two robots with highly human-like faces, Alfie (Furhat Robot) and the android robot Elenoide, show that ExGenNet can successfully generate sets of joint configurations for predefined facial expressions on both robots. This ability of ExGenNet to generate realistic facial expressions was further validated in a pilot study where the majority of human subjects could accurately recognize most of the generated facial expressions on both the robots.

LINE AS AN EXPRESSIVE MEANS IN DESIGNOF THE VISUAL ENVIRONMENT

The purpose. The peculiarities of the use of linear images in the visual environment are investigated. Their influence on human perception and relevance of their use in design and art is determined. Methodology. General scientific research methods are used: analytical, typological, comparative analysis are used to determine the lines and their application in advertising and design. Results.Linear images differ against the background of modern media space. The use of linear, schematic, simplified images and their assimilation appeals to the historical development of human perception, which allows such images to be perceived on a subconscious level, without requiring concentration and distraction from other activities. Trends in the development of linear images in human perception are analyzed. The use of lines in various types of fine arts is emphasized and revealed. The scientific novelty. To consider the lines as a tool for visual emphasis of facilities and distribution lines in various forms of art. Their historical development and significance for human activity are established. Practical significance. The study allows to expand knowledge about the use of lines in modern and elite areas of art. The obtained results can be used in further studies of linear visual solutions.

The moral branding of Fairtrade: Opportunities and pitfalls of visual representations in the Fairtrade system – empirical insights on the perspectives of German consumers

The Fairtrade system can be considered as one of the most successful real-world experiments of alternative economies. However, Fairtrade more than other alternative economic approaches relies on mass market mechanisms and the sale of its products via conventional distribution channels and retail outlets. To gradually transform the ‘unfair’ mechanisms of world trade and to achieve its social and environmental goals, Fairtrade requires a constant growth in sales. This also means that the marketing of Fairtrade goods in the consumer countries is subject to established capitalist mechanisms of advertising and demand creation. Although Fairtrade’s brand building aims at differentiation through alternative values of cooperation, trust, and fairness, it also works within the constraints of simplified and abridged advertising messages. More often than not, contrasts and stereotypes are over-emphasized – a phenomenon that critics call the ‘exploitation of difference’. In this way, new forms of ‘distancing’ and ‘othering’ are constantly built up and reinforced. Individuals, livelihoods, production practices, and entire landscapes in the Global South become commodified and are used for brand development and sales promotion. This paper examines these issues based on interviews and a questionnaire survey among German consumers. Our empirical insights indicate that the visual language used for Fairtrade marketing has to be targeted to critical consumers, who are increasingly skeptical of overly moralizing and simplified images with exaggerated contrasts between the ‘different worlds’ of producers in the South and consumers in the North. However, it should also be noted that Fairtrade Germany is increasingly aware of challenges in its visual communication and is intensively reflecting on its visual language.

Spatial Models of Sugars and Their Compounds

The images of saccharide and polysaccharide molecules in spaces of various dimensions are considered. A method has been developed for obtaining simplified three-dimensional images of sugar molecules and their chains based on their images in spaces of higher dimensions. It was found that three-dimensional images of furanose and pyranose molecules fundamentally differ from each other to form convex and, accordingly, non-convex bodies. This leads to fundamental differences in the structure of polysaccharides from these molecules.

USE OF DYNAMIC BLOCKS FOR CREATION OF ELECTRONIC IMAGE LIBRARIES OF TYPICAL FIXING ELEMENTS BY AUTO CAD RESOURCES

The use of electronic libraries of parameterized images of objects made in the form of dynamic blocks in the practice of design work refers to resource-saving technologies that are actively used in modern production. The article considers an example of creating parameterized simplified images of fasteners using dynamic blocks of the computer-aided design system AutoCAD. Dynamic blocks can be used to create electronic image libraries of technological, design, electrical and other elements. The algorithm considered in the article to create dynamic blocks of simplified images of fasteners is introduced into the educational process at the Kyiv Polytechnic Institute named after Igor Sikorsky in the course "Engineering and Computer Graphics" and can be used in design practice.
 The testing of this algorithm in the training process yielded a positive result. When using parameterized drawings, students understand more deeply the impact of each parameter on the design of the object.

Diagnostic Imaging and Mechanical Objectivity in Medicine

Before the advent of automatism in image-making practices, scientists, anatomists, and physicians artistically depicted simplified images for scientific atlas making. This technique conferred subjectivity to a supposedly objective scientific process, sparking confrontations between anatomists regarding accuracy that heralded a new concept in the late 19th century - mechanical objectivity - that would revolutionize scientific knowledge and the field of medicine OBJECTIVES: The purpose of this health history research study is to trace the evolution of mechanical objectivity from empirical studies of early anatomists in the 19th century to the advent of x-ray technology, digitization of imaging, and disruptive technological innovations such as artificial intelligence, while simultaneously unveiling the challenges of mitigating human bias, despite advancements in medical imaging practices. This narrative literature review was conducted using the Scopus® database under the guidance of both medical historians and practicing physicians to ensure its applicability and historical accuracy CONCLUSION: Despite a century-long quest for optimizing mechanical objectivity in diagnostic imaging to more accurately and efficiently interpret medical images, human bias remains an important factor. This historical review describes the development of medical imaging technologies over the last century with emphasis on the role played by human bias and subjectivity in a rapidly expanding field of medical imaging technology including artificial intelligence.

Cooking with refugees and migrants. Performing authenticity and traditionality for Warsaw’s culinary tourists

During the migration crisis of 2015, a commonly shared belief about the integrating role of food resulted in the emergence of several culinary initiatives directed at refugees and migrants in Warsaw. I show that these culinary initiatives form a space for the creation of ostensibly opposing processes. On the one hand, they empower refugees and migrants by embracing the culinary cultures of their home countries; on the other, they facilitate the creation of simplified and folkloristic images of them. During culinary workshops, the role of migrants and refugees is to recreate traditional dishes, using “authentic” recipes. At the same time, they are restricted by the organizers’ ethical foodways and the demands of Warsaw’s culinary tourists, such as vegetarianism, to which migrants and refugees skilfully adapt. These processes result from a neoliberal logic, whereby refugees’ and migrants’ experiences and their ethnicity become commodities in the NGO market.This article draws on ethnographic research conducted between 2017 and 2018 in Warsaw. I look at the biographies of six women refugees cooperating with selected initiatives. I analyse their strategies of recreating traditionality in the dishes they cook in order to authenticate their migration stories. I also examine their experiences and practices in the context of “food capital” that emerges as a result of the exchange of cultural capital between migrants and residents, and “refugee capital”, defined as the ability to use refugee status for personal development and integration. Combining “food capital” with “refugee capital” turns out to be an excellent recipe for success for refugees’ migration projects.

Two New Marine Free-Living Nematodes from Jeju Island Together with a Review of the Genus Gammanema Cobb 1920 (Nematoda, Chromadorida, Selachinematidae)

In the context of exploration of meiofauna in a sandy intertidal zone of Jeju Island (South Korea), over 70 nematode species are identified, some which have been proven to be new for science. Two new free-living marine nematode species of the family Selachinematidae (Chromadorida, Selachinematidae, Choniolaiminae) are described from the intertidal sandy sediments of Jeju Island (South Korea). Gammanema okhlopkovi sp. n. is closest to Gammanema anthostoma (Okhlopkov, 2002) and differs by having longer cephalic setae (8.5–19 μm in G. okhlopkovi versus 6–7.5 μm in G. anthostoma) and by the presence of precloacal supplementary organs. The genus diagnosis of Gammanema is updated. The genus includes fourteen valid species, while three species are considered species inquirendae due to incomplete diagnoses and illustrations impeding their correct recognition. An annotated list of valid and invalid Gammanema species is provided. A pictorial key for valid Gammanema species is constructed, which consists of two components: (1) simplified images of heads, and (2) a table summarizing most of the significant measured and numeric characters between species. Latronema obscuramphis sp. n. differs from its related species Latronema aberrans (Allgén 1934), Latronema annulatum (Gerlach, 1953), and Latronema spinosum (Andrássy, 1973) by body size, number of supplementary organs, tail shape, length of spicules, and cuticle ornamentation.

Imaginarios, mitos y actitudes frente a la bisexualidad en estudiantes de Trabajo Social

En un sistema patriarcal, machista y heterocentrista, a través de un proceso de estigmatización, se hace uso de mitos (relatos tradicionales que contienen pautas de comportamiento y realidades alteradas) y estereotipos (falsas concepciones producto de imágenes mentales simplificadas), para devaluar a las personas cuya orientación sexual rompe con los modelos tradicionalmente establecidos. Tomando en consideración al alumnado de Trabajo Social de Santiago de Compostela (Galicia, España) se propone una investigación a través de un método mixto, con el objetivo de analizar el nivel aversión hacia la bisexualidad y el calado que tienen los mitos existentes sobre la bisexualidad. Los resultados muestran un nivel positivo de actitudes hacia la bisexualidad, así como un imaginario basado en el principio de igualdad y un nivel de interiorización de los distintos mitos sobre la bisexualidad medio-bajo. Por otro lado, se observa que las actitudes y los mitos están directamente relacionados.

Cognitive Study of Products’ User Interfaces for Use by Elderly People

This study investigated elderly users’ understanding of images and their perception of products. Various product function images served as testing samples, and participants were required to rank the various images. The results of this study revealed that the elderly participants easily recognized concrete images or images that were familiar. The elderly participants had various views concerning simplified images. Associative images that could not be associated with objects from daily life were not efficiently recognized. In the present study, we only explored the effect of shape and elements used on elderly people’s understanding of images. The results of this study can serve as a reference for designers when constructing a product’s user interface to be used by elderly people.

Towards a generalised development of synthetic CT images and assessment of their dosimetric accuracy

Background: The interest in generating “synthetic computed tomography (CT) images” from magnetic resonance (MR) images has been increasing over the past years due to advances in MR guidance for radiotherapy. A variety of methods for synthetic CT creation have been developed, from simple bulk density assignment to complex machine learning algorithms.Material and methods: In this study, we present a general method to determine simplistic synthetic CTs and evaluate them according to their dosimetric accuracy. It separates the requirements on the MR image and the associated calculation effort to generate a synthetic CT. To evaluate the significance of the dosimetric accuracy under realistic conditions, clinically common uncertainties including position shifts and Hounsfield lookup table (HLUT) errors were simulated. To illustrate our approach, we first translated CT images from a test set of six pelvic cancer patients to relative electron density (ED) via a clinical HLUT. For each patient, seven simplified ED images (simED) were generated at different levels of complexity, ranging from one to four tissue classes. Then, dose distributions optimised on the reference ED image and the simEDs were compared to each other in terms of gamma pass rates (2 mm/2% criteria) and dose volume metrics.Results: For our test set, best results were obtained for simEDs with four tissue classes representing fat, soft tissue, air, and bone. For this simED, gamma pass rates of 99.95% (range: 99.72–100%) were achieved. The decrease in accuracy from ED simplification was smaller in this case than the influence of the uncertainty scenarios on the reference image, both for gamma pass rates and dose volume metrics.Conclusions: The presented workflow helps to determine the required complexity of synthetic CTs with respect to their dosimetric accuracy. The investigated cases showed potential simplifications, based on which the synthetic CT generation could be faster and more reproducible.