Search results for: interactive learning environments

Authors: Ravija Gunawardana, Banuka Athuraliya

Abstract:

Machine learning has emerged as a powerful tool for disease diagnosis and prediction. The use of machine learning algorithms has the potential to improve the accuracy of disease prediction, thereby enabling medical professionals to provide more effective and personalized treatments. This study focuses on developing a machine-learning model for disease prediction using symptoms and X-ray images. The importance of this study lies in its potential to assist medical professionals in accurately diagnosing diseases, thereby improving patient outcomes. Respiratory diseases are a significant cause of morbidity and mortality worldwide, and chest X-rays are commonly used in the diagnosis of these diseases. However, accurately interpreting X-ray images requires significant expertise and can be time-consuming, making it difficult to diagnose respiratory diseases in a timely manner. By incorporating machine learning algorithms, we can significantly enhance disease prediction accuracy, ultimately leading to better patient care. The study utilized the Mask R-CNN algorithm, which is a state-of-the-art method for object detection and segmentation in images, to process chest X-ray images. The model was trained and tested on a large dataset of patient information, which included both symptom data and X-ray images. The performance of the model was evaluated using a range of metrics, including accuracy, precision, recall, and F1-score. The results showed that the model achieved an accuracy rate of over 90%, indicating that it was able to accurately detect and segment regions of interest in the X-ray images. In addition to X-ray images, the study also incorporated symptoms as input data for disease prediction. The study used three different classifiers, namely Random Forest, K-Nearest Neighbor and Support Vector Machine, to predict diseases based on symptoms. These classifiers were trained and tested using the same dataset of patient information as the X-ray model. The results showed promising accuracy rates for predicting diseases using symptoms, with the ensemble learning techniques significantly improving the accuracy of disease prediction. The study's findings indicate that the use of machine learning algorithms can significantly enhance disease prediction accuracy, ultimately leading to better patient care. The model developed in this study has the potential to assist medical professionals in diagnosing respiratory diseases more accurately and efficiently. However, it is important to note that the accuracy of the model can be affected by several factors, including the quality of the X-ray images, the size of the dataset used for training, and the complexity of the disease being diagnosed. In conclusion, the study demonstrated the potential of machine learning algorithms for disease prediction using symptoms and X-ray images. The use of these algorithms can improve the accuracy of disease diagnosis, ultimately leading to better patient care. Further research is needed to validate the model's accuracy and effectiveness in a clinical setting and to expand its application to other diseases.

Keywords: K-nearest neighbor, mask R-CNN, random forest, support vector machine

Procedia PDF Downloads 154

Authors: Modreck Gomo

Abstract:

The evolution of groundwater chemistry and its quality is largely controlled by hydrogeochemical processes and their understanding is therefore important for groundwater quality assessments and protection of the water resources. A study was conducted in Bloemfontein town of South Africa to assess and compare the groundwater chemistry and quality characteristics in an alluvial aquifer and single-plane fractured-rock aquifers. 9 groundwater samples were collected from monitoring boreholes drilled into the two aquifer systems during a once-off sampling exercise. Samples were collected through low-flow purging technique and analysed for major ions and trace elements. In order to describe the hydrochemical facies and identify dominant hydrogeochemical processes, the groundwater chemistry data are interpreted using stiff diagrams and principal component analysis (PCA), as complimentary tools. The fitness of the groundwater quality for domestic and irrigation uses is also assessed. Results show that the alluvial aquifer is characterised by a Na-HCO₃ hydrochemical facie while fractured-rock aquifer has a Ca-HCO₃ facie. The groundwater in both aquifers originally evolved from the dissolution of calcite rocks that are common on land surface environments. However the groundwater in the alluvial aquifer further goes through another evolution as driven by cation exchange process in which Na in the sediments exchanges with Ca²⁺ in the Ca-HCO₃ hydrochemical type to result in the Na-HCO₃ hydrochemical type. Despite the difference in the hydrogeochemical processes between the alluvial aquifer and single-plane fractured-rock aquifer, this did not influence the groundwater quality. The groundwater in the two aquifers is very hard as influenced by the elevated magnesium and calcium ions that evolve from dissolution of carbonate minerals which typically occurs in surface environments. Based on total dissolved levels (600-900 mg/L), groundwater quality of the two aquifer systems is classified to be of fair quality. The negative potential impacts of the groundwater quality for domestic uses are highlighted.

Keywords: alluvial aquifer, fractured-rock aquifer, groundwater quality, hydrogeochemical processes

Procedia PDF Downloads 204

Authors: Felix Liedel

Abstract:

The rise of artificial intelligence has numerous and far-reaching consequences. In addition to the obvious consequences for entire professions, the increasing interaction with chatbots also has a wide range of social consequences and implications. We are already increasingly used to interacting with digital chatbots, be it in virtual consulting situations, creative development processes or even in building personal or intimate virtual relationships. A media-theoretical classification of these phenomena has so far been difficult, partly because the interactive element in the exchange with artificial intelligence has undeniable similarities to human-to-human communication but is not identical to it. The proposed study, therefore, aims to reformulate the concept of symbolic interaction in the tradition of George Herbert Mead as symbolic interactivity in communication with chatbots. In particular, Mead's socio-psychological considerations will be brought into dialog with the specific conditions of digital media, the special dispositive situation of chatbots and the characteristics of artificial intelligence. One example that illustrates this particular communication situation with chatbots is so-called consensus fiction: In face-to-face communication, we use symbols on the assumption that they will be interpreted in the same or a similar way by the other person. When briefing a chatbot, it quickly becomes clear that this is by no means the case: only the bot's response shows whether the initial request corresponds to the sender's actual intention. This makes it clear that chatbots do not just respond to requests. Rather, they function equally as projection surfaces for their communication partners but also as distillations of generalized social attitudes. The personalities of the chatbot avatars result, on the one hand, from the way we behave towards them and, on the other, from the content we have learned in advance. Similarly, we interpret the response behavior of the chatbots and make it the subject of our own actions with them. In conversation with the virtual chatbot, we enter into a dialog with ourselves but also with the content that the chatbot has previously learned. In our exchanges with chatbots, we, therefore, interpret socially influenced signs and behave towards them in an individual way according to the conditions that the medium deems acceptable. This leads to the emergence of situationally determined digital identities that are in exchange with the real self but are not identical to it: In conversation with digital chatbots, we bring our own impulses, which are brought into permanent negotiation with a generalized social attitude by the chatbot. This also leads to numerous media-ethical follow-up questions. The proposed approach is a continuation of my dissertation on moral decision-making in so-called interactive films. In this dissertation, I attempted to develop a concept of symbolic interactivity based on Mead. Current developments in artificial intelligence are now opening up new areas of application.

Keywords: artificial intelligence, chatbot, media theory, symbolic interactivity

Procedia PDF Downloads 52

Authors: Carla Teixeira, Diana Azevedo, Jonatas Bessa, Maria Guilam

Abstract:

The flipped classroom is a blended learning modality that combines face-to-face and virtual activities of self-learning, mediated by digital information and communication technologies, which reverses traditional teaching approaches and presents, as a presupposition, the previous study of contents by students. In the following face-to-face activities, the contents are discussed, producing active learning. This work aims to describe the systematization process of the use of flipped classrooms as a method to develop complementary national activities in PROFSAÚDE, a professional master's program in the area of public health, offered as a distance learning course, in the network, in Brazil. The complementary national activities were organized with the objective of strengthening and qualifying students´ learning process. The network gathers twenty-two public institutions of higher education in the country. Its national coordination conducted a survey to detect complementary educational needs, supposed to improve the formative process and align important content sums for the program nationally. The activities were organized both asynchronously, making study materials available in Google classrooms, and synchronously in a tele presential way, organized on virtual platforms to reach the largest number of students in the country. The asynchronous activities allowed each student to study at their own pace and the synchronous activities were intended for deepening and reflecting on the themes. The national team identified some professors' areas of expertise, who were contacted for the production of audiovisual content such as video classes and podcasts, guidance for supporting bibliographic materials and also to conduct synchronous activities together with the technical team. The contents posted in the virtual classroom were organized by modules and made available before the synchronous meeting; these modules, in turn, contain “pills of experience” that correspond to reports of teachers' experiences in relation to the different themes. In addition, activity was proposed, with questions aimed to expose doubts about the contents and a learning challenge, as a practical exercise. Synchronous activities are built with different invited teachers, based on the participants 'discussions, and are the forum where teachers can answer students' questions, providing feedback on the learning process. At the end of each complementary activity, an evaluation questionnaire is available. The responses analyses show that this institutional network experience, as pedagogical innovation, provides important tools to support teaching and research due to its potential in the participatory construction of learning, optimization of resources, the democratization of knowledge and sharing and strengthening of practical experiences on the network. One of its relevant aspects was the thematic diversity addressed through this method.

Keywords: active learning, flipped classroom, network education experience, pedagogic innovation

Procedia PDF Downloads 159

Authors: Tali Neta

Abstract:

Humans’ impact on the environment has been incredibly brutal, with enormous plastic- and other pollutants (e.g., cigarette buds, paper cups, tires) worldwide. On land, litter costs taxpayers a fortune. Most of the litter pollution comes from the land, yet it is one of the greatest hazards to marine environments. Due to spatial and temporal limitations, previous litter data covered very small areas. Currently, smartphones can be used to obtain information on various pollutants (through citizen science), and they can greatly assist in acknowledging and mitigating the environmental impact of litter. Litter app data, such as the Litterati, are available for study through a global map only; these data are not available for download, and it is not clear whether irrelevant hashtags have been eliminated. Instagram and Twitter open-source geospatial data are available for download; however, these are considered inaccurate, computationally challenging, and impossible to quantify. Therefore, the resulting data are of poor quality. Other downloadable geospatial data (e.g., Marine Debris Tracker8 and Clean Swell10) are focused on marine- rather than terrestrial litter. Therefore, accurate terrestrial geospatial documentation of litter distribution is needed to improve environmental awareness. The current research employed citizen science to examine litter distribution in Lethbridge, Alberta, Canada, using the OpenLitterMap (OLM) app. The OLM app is an application used to track litter worldwide, and it can mark litter locations through photo georeferencing, which can be presented through GIS-designed maps. The OLM app provides open-source data that can be downloaded. It also offers information on various litter types and “hot-spots” areas where litter accumulates. In this study, Lethbridge College students collected litter data with the OLM app. The students produced GIS Story Maps (interactive web GIS illustrations) and presented these to school children to improve awareness of litter's impact on environmental health. Preliminary results indicate that towards the Lethbridge Coulees’ (valleys) East edges, the amount of litter significantly increased due to shrubs’ presence, that acted as litter catches. As wind generally travels from west to east in Lethbridge, litter in West-Lethbridge often finds its way down in the east part of the coulees. The students’ documented various litter types, while the majority (75%) included plastic and paper food packaging. The students also found metal wires, broken glass, plastic bottles, golf balls, and tires. Presentations of the Story Maps to school children had a significant impact, as the children voluntarily collected litter during school recess, and they were looking into solutions to reduce litter. Further litter distribution documentation through Citizen Science is needed to improve public awareness. Additionally, future research will be focused on Drone imagery of highly concentrated litter areas. Finally, a time series analysis of litter distribution will help us determine whether public education through Citizen Science and Story Maps can assist in reducing litter and reaching a cleaner and healthier environment.

Keywords: citizen science, litter pollution, Open Litter Map, GIS Story Map

Procedia PDF Downloads 79

Authors: Agah Tuğrul Korucu, Handan Atun

Abstract:

Diversity and usefulness of information that used in education are have increased due to development of technology. Web technologies have made enormous contributions to the distance learning system especially. Mobile systems, one of the most widely used technology in distance education, made much easier to access web technologies. Not bounding by space and time, individuals have had the opportunity to access the information on web. In addition to this, the storage of educational information and resources and accessing these information and resources is crucial for both students and teachers. Because of this importance, development and dissemination of web technologies supply ease of access to information and resources are provided by web technologies. Dynamic web technologies introduced as new technologies that enable sharing and reuse of information, resource or applications via the Internet and bring websites into expandable platforms are commonly known as Web 2.0 technologies. Cloud systems are one of the dynamic web technologies that defined as a model provides approaching the demanded information independent from time and space in appropriate circumstances and developed by NIST. One of the most important advantages of cloud systems is meeting the requirements of users directly on the web regardless of hardware, software, and dealing with install. Hence, this study aims at using cloud services in education and investigating the services provided by the cloud computing. Survey method has been used as research method. In the findings of this research the fact that cloud systems are used such studies as resource sharing, collaborative work, assignment submission and feedback, developing project in the field of education, and also, it is revealed that cloud systems have plenty of significant advantages in terms of facilitating teaching activities and the interaction between teacher, student and environment.

Keywords: cloud systems, cloud systems in education, online learning environment, integration of information technologies, e-learning, distance learning

Procedia PDF Downloads 349

Authors: A. Georgiou, A. Galani, A. Karatza, G. E. Bampasidis

Abstract:

Augmented Reality (AR) technology has become integral in enhancing visitor experiences at Greece's UNESCO World Heritage Sites. This research meticulously investigates various facets of AR applications/games associated with these revered sites. The cultural heritage represents the identity of each nation in the world. Technology can breathe life into this identity. Through Augmented Reality (AR), individuals can travel back in time, visit places they cannot access in real life, discover the history of these places, and live unique experiences. The study examines the objectives and intended goals behind the development and deployment of each augmented reality application/game pertaining to the UNESCO World Heritage Sites in Greece. It thoroughly analyzes the scenarios presented within these AR games/applications, examining how historical narratives, interactive elements, and cultural context are incorporated to engage users. Furthermore, the research identifies and assesses the technological platforms utilized for the development and implementation of these AR experiences, encompassing mobile devices, AR headsets, or specific software frameworks. It classifies and examines the types of augmented reality employed within these applications/games, including marker-based, markerless, location-based, or immersive AR experiences. Evaluation of the benefits accrued by visitors engaging with these AR applications/games, such as enhanced learning experiences, improved cultural understanding, and heightened engagement with the heritage sites, forms a crucial aspect of this study. Additionally, the research scrutinizes potential drawbacks or limitations associated with the AR applications/games, considering technological barriers, user accessibility issues, or constraints affecting user experience. By thoroughly investigating these pivotal aspects, this research aims to provide a comprehensive overview and analysis of the landscape of augmented reality applications/games linked to the UNESCO World Heritage Sites in Greece. The findings seek to contribute nuanced insights into the effectiveness, challenges, and opportunities associated with leveraging AR technology for heritage site preservation, visitor engagement, and cultural enrichment.

Keywords: augmented reality, AR applications, UNESCO sites, cultural heritage, Greece, visitor engagement, historical narratives

Procedia PDF Downloads 64

Authors: Michael Karnes, Alper Yilmaz

Abstract:

This paper presents the application of the omni-modeler towards pedestrian redetection. The pedestrian redetection task creates several challenges when applying deep neural networks (DNN) due to the variety of pedestrian appearance with camera position, the variety of environmental conditions, and the specificity required to recognize one pedestrian from another. DNNs require significant training sets and are not easily adapted for changes in class appearances or changes in the set of classes held in its knowledge domain. Pedestrian redetection requires an algorithm that can actively manage its knowledge domain as individuals move in and out of the scene, as well as learn individual appearances from a few frames of a video. The Omni-Modeler is a dynamically learning few-shot visual recognition algorithm developed for tasks with limited training data availability. The Omni-Modeler adapts the knowledge domain of pre-trained deep neural networks to novel concepts with a calculated localized language encoder. The Omni-Modeler knowledge domain is generated by creating a dynamic dictionary of concept definitions, which are directly updatable as new information becomes available. Query images are identified through nearest neighbor comparison to the learned object definitions. The study presented in this paper evaluates its performance in re-identifying individuals as they move through a scene in both single-camera and multi-camera tracking applications. The results demonstrate that the Omni-Modeler shows potential for across-camera view pedestrian redetection and is highly effective for single-camera redetection with a 93% accuracy across 30 individuals using 64 example images for each individual.

Keywords: dynamic learning, few-shot learning, pedestrian redetection, visual recognition

Procedia PDF Downloads 76

Authors: Michael Lousis

Abstract:

The systematic identification of the most conspicuous and significant errors made by learners during three-years of testing of their progress in learning Arithmetic are presented in this article. How these errors have changed over three-years of school instruction of Arithmetic also is shown. The sample is comprised of two hundred (200) English students and one hundred and fifty (150) Greek students. These students were purposefully selected according to their participation in each testing session in the development of the three-year Kassel Project in England and Greece, in both domains simultaneously in Arithmetic and Algebra. The data sample includes six test-scripts corresponding to three testing sessions in both Arithmetic and Algebra respectively.

Keywords: arithmetic, errors, Kassel Project, progress of learning

Procedia PDF Downloads 264

Authors: Milad Abbasi

Abstract:

Carbon fiber-reinforced polymer (CFRP) composite structures are increasingly being utilized in the automotive industry due to their lightweight and specific energy absorption capabilities. Although it is impossible to predict composite mechanical properties directly using theoretical methods, various research has been conducted so far in the literature for accurate simulation of CFRP structures' energy-absorbing behavior. In this research, axial compression experiments were carried out on hand lay-up unidirectional CFRP composite tubes. The fabrication method allowed the authors to extract the material properties of the CFRPs using ASTM D3039, D3410, and D3518 standards. A neural network machine learning algorithm was then utilized to build a robust prediction model to forecast the axial compressive properties of CFRP tubes while reducing high-cost experimental efforts. The predicted results have been compared with the experimental outcomes in terms of load-carrying capacity and energy absorption capability. The results showed high accuracy and precision in the prediction of the energy-absorption capacity of the CFRP tubes. This research also demonstrates the effectiveness and challenges of machine learning techniques in the robust simulation of composites' energy-absorption behavior. Interestingly, the proposed method considerably condensed numerical and experimental efforts in the simulation and calibration of CFRP composite tubes subjected to compressive loading.

Keywords: CFRP composite tubes, energy absorption, crushing behavior, machine learning, neural network

Procedia PDF Downloads 153

Authors: Mary Jane B. Suarez

Abstract:

Public speaking anxiety (PSA), also known as speech anxiety, is innumerably persistent in any traditional communication classes, especially for students who learn English as a second language. The speech anxiety intensifies when communication skills assessments have taken their toll in an online or a remote mode of learning due to the perils of the COVID-19 virus. Both teachers and students have experienced vast ambiguity on how to realize a still effective way to teach and learn speaking skills amidst the pandemic. Communication skills assessments like public speaking, oral presentations, and student reporting have defined their new meaning using Google Meet, Zoom, and other online platforms. Though using such technologies has paved for more creative ways for students to acquire and develop communication skills, the effectiveness of using such assessment tools stands in question. This mixed method study aimed to determine the factors that affected the public speaking skills of students in a communication class, to probe on the assessment gaps in assessing speaking skills of students attending online classes vis-à-vis the implementation of remote and blended modalities of learning, and to recommend ways on how to address the public speaking anxieties of students in performing a speaking task online and to bridge the assessment gaps based on the outcome of the study in order to achieve a smooth segue from online to on-ground instructions maneuvering towards a much better post-pandemic academic milieu. Using a convergent parallel design, both quantitative and qualitative data were reconciled by probing on the public speaking anxiety of students and the potential assessment gaps encountered in an online English communication class under remote and blended learning. There were four phases in applying the convergent parallel design. The first phase was the data collection, where both quantitative and qualitative data were collected using document reviews and focus group discussions. The second phase was data analysis, where quantitative data was treated using statistical testing, particularly frequency, percentage, and mean by using Microsoft Excel application and IBM Statistical Package for Social Sciences (SPSS) version 19, and qualitative data was examined using thematic analysis. The third phase was the merging of data analysis results to amalgamate varying comparisons between desired learning competencies versus the actual learning competencies of students. Finally, the fourth phase was the interpretation of merged data that led to the findings that there was a significantly high percentage of students' public speaking anxiety whenever students would deliver speaking tasks online. There were also assessment gaps identified by comparing the desired learning competencies of the formative and alternative assessments implemented and the actual speaking performances of students that showed evidence that public speaking anxiety of students was not properly identified and processed.

Keywords: blended learning, communication skills assessment, public speaking anxiety, speech anxiety

Procedia PDF Downloads 102

Authors: Mamunur Rashid, Jialin Yan

Abstract:

The current years have witnessed a rapid evolution and integration of artificial intelligence (AI) in various fields, prominently influencing the education industry. Acknowledging this transformative wave, AI tools like ChatGPT and Grammarly have undeniably introduced perspectives and skills, enriching the educational experiences of higher education students. The prevalence of AI utilization in higher education also drives an increasing number of researchers' attention in various dimensions. Departments, offices, and professors in universities also designed and released a set of policies and guidelines on using AI effectively. In regard to this, the study targets exploring and analyzing graduate students' perspectives regarding AI guidelines set by teachers. A mixed-methods study will be mainly conducted in this study, employing in-depth interviews and focus groups to investigate and collect students' perspectives. Relevant materials, such as syllabi and course instructions, will also be analyzed through the documentary analysis to facilitate understanding of the study. Surveys will also be used for data collection and students' background statistics. The integration of both interviews and surveys will provide a comprehensive array of student perspectives across various academic disciplines. The study is anchored in the theoretical framework of self-determination theory (SDT), which emphasizes and explains the students' perspective under the AI guidelines through three core needs: autonomy, competence, and relatedness. This framework is instrumental in understanding how AI guidelines influence students' intrinsic motivation and sense of empowerment in their learning environments. Through qualitative analysis, the study reveals a sense of confusion and uncertainty among students regarding the appropriate application and ethical considerations of AI tools, indicating potential challenges in meeting their needs for competence and autonomy. The quantitative data further elucidates these findings, highlighting a significant communication gap between students and educators in the formulation and implementation of AI guidelines. The critical findings of this study mainly come from two aspects: First, the majority of graduate students are uncertain and confused about relevant AI guidelines given by teachers. Second, this study also demonstrates that the design and effectiveness of course materials, such as the syllabi and instructions, also need to adapt in regard to AI policies. It indicates that certain of the existing guidelines provided by teachers lack consideration of students' perspectives, leading to a misalignment with students' needs for autonomy, competence, and relatedness. More emphasize and efforts need to be dedicated to both teacher and student training on AI policies and ethical considerations. To conclude, in this study, graduate students' perspectives on teacher-provided AI guidelines are explored and reflected upon, calling for additional training and strategies to improve how these guidelines can be better disseminated for their effective integration and adoption. Although AI guidelines provided by teachers may be helpful and provide new insights for students, educational institutions should take a more anchoring role to foster a motivating, empowering, and student-centered learning environment. The study also provides some relevant recommendations, including guidance for students on the ethical use of AI and AI policy training for teachers in higher education.

Keywords: higher education policy, graduate students’ perspectives, higher education teacher, AI guidelines, AI in education

Procedia PDF Downloads 74

Authors: Laurence C. Hanes, Robert J. Stone

Abstract:

Video games have reached a point of huge commercial success as well as wide familiarity with audiences both young and old. Much attention and research have also been directed towards serious games and their potential learning affordances. It is little surprise that the field of virtual heritage has taken a keen interest in using serious games to present cultural heritage information to users, with applications ranging from museums and cultural heritage institutions, to academia and research, to schools and education. Many researchers have already documented their efforts to develop and distribute virtual heritage serious games. Although attempts have been made to create classifications of the different types of virtual heritage games (somewhat akin to the idea of game genres), no formal taxonomy has yet been produced to define the different types of cultural heritage and historical information that can be presented through these games at a content level, and how the information can be manifested within the game. This study proposes such a taxonomy. First the informational content is categorized as heritage or historical, then further divided into tangible, intangible, natural, and analytical. Next, the characteristics of the manifestation within the game are covered. The means of manifestation, level of demonstration, tone, and focus are all defined and explained. Finally, the potential learning outcomes of the content are discussed. A demonstration of the taxonomy is then given by describing the informational content and corresponding manifestations within several examples of virtual heritage serious games as well as commercial games. It is anticipated that this taxonomy will help designers of virtual heritage serious games to think about and clearly define the information they are presenting through their games, and how they are presenting it. Another result of the taxonomy is that it will enable us to frame cultural heritage and historical information presented in commercial games with a critical lens, especially where there may not be explicit learning objectives. Finally, the results will also enable us to identify shared informational content and learning objectives between any virtual heritage serious and/or commercial games.

Keywords: informational content, serious games, taxonomy, virtual heritage

Procedia PDF Downloads 367

Authors: Caroline Selai, Sushrut Jadhav

Abstract:

Background: University College London (UCL), the first secular university in England to admit students regardless of their religion and gender, has nearly 29,000 students of which approximately 30% are international students. The UCL Cultural Consultation Service (CCS) for staff and students is a unique service that provides assistance to staff and students experiencing challenges in their teaching, enabling, support work or studies which they believe may have a cultural component. The service provides one-to-one and group consultations, lectures, seminars, ‘grand rounds’, interactive workshops and bespoke interventions. Data: This paper presents a content analysis of CCS referrals over the last 36 months. We focus on the experience of international students, many of whom experience not only a challenge to their academic identity but also a profound challenge to their personal cultural identity. We also present 3 vignettes to illustrate how students interpret, accept, contest and resist changes in their cultural and academic identity. Discussion: This paper highlights (i) how students from collectivist cultures attempt to assimilate within an individualistic, highly competitive western university that is bound by its own institutional norms; (ii) problems in negotiating challenges at the interface of culture and gender (iii) the impact of culturally different hierarchies of power, discrimination and authority and (iv) the significance of earlier traumatic and kinship conflicts. Many international students’ social identities are shaped by their cultural and family scripts. A large number have been taught that their teachers are to be revered and their teachings unchallenged. This is at odds with quintessential goal of the western university to encourage healthy scepticism and hone students’ critical thinking skills. Conclusions: Pupil-teacher ‘cultural transference’ and shifts in cultural academic identities of students underscore critical aspects of developmental and learning challenges for students. Staff-student cultural conflict requires a broader, systemic analysis of students, staff and the wider organisation. Our findings challenge Eurocentric psychodynamic concepts such as the nature of parent-child relationship in Western Europe. We argue for a broader, more inclusive approach to develop both effective pedagogic skills in euro-american academic institutions and culturally- appropriate psychodynamic theory to underpin counselling international students.

Keywords: academic identity, cultural transference, cultural consultation in higher education, cultural formulation, cultural identity.

Procedia PDF Downloads 460

Authors: Mumtaz Mooncey, Alexander Jolly, Megan Fisher, Kerry Robinson, Robert Lloyd, Dave Fielding

Abstract:

Introduction: The integration of the WingFactors in-situ simulation programme has transformed the education landscape at the Whittington Health NHS Trust. To date, there have been a total of 90 simulations - 19 aimed at Paediatric trainees, including 2 Child and Adolescent Mental Health (CAMHS) scenarios. The opportunity for joint debriefs provided by clinical faculty and airline pilots, has created a new exciting avenue to explore human factors within psychiatry. Through the use of real clinical environments and primed actors; the benefits of high fidelity simulation, interdisciplinary and interprofessional learning has been highlighted. The use of in-situ simulation within Psychiatry is a newly emerging concept and its success here has been recognised by unanimously positive feedback from participants and acknowledgement through nomination for the Health Service Journal (HSJ) Award (Best Education Programme 2021). Methodology: The first CAMHS simulation featured a collapsed patient in the toilet with a ligature tied around her neck, accompanied by a distressed parent. This required participants to consider:; emergency physical management of the case, alongside helping to contain the mother and maintaining situational awareness when transferring the patient to an appropriate clinical area. The second simulation was based on a 17- year- old girl attempting to leave the ward after presenting with an overdose, posing potential risk to herself. The safe learning environment enabled participants to explore techniques to engage the young person and understand their concerns, and consider the involvement of other members of the multidisciplinary team. The scenarios were followed by an immediate ‘hot’ debrief, combining technical feedback with Human Factors feedback from uniformed airline pilots and clinicians. The importance of psychological safety was paramount, encouraging open and honest contributions from all participants. Key learning points were summarized into written documents and circulated. Findings: The in-situ simulations demonstrated the need for practical changes both in the Emergency Department and on the Paediatric ward. The presence of airline pilots provided a novel way to debrief on Human Factors. The following key themes were identified: -Team-briefing (‘Golden 5 minutes’) - Taking a few moments to establish experience, initial roles and strategies amongst the team can reduce the need for conversations in front of a distressed patient or anxious relative. -Use of checklists / guidelines - Principles associated with checklist usage (control of pace, rigor, team situational awareness), instead of reliance on accurate memory recall when under pressure. -Read-back - Immediate repetition of safety critical instructions (e.g. drug / dosage) to mitigate the risks associated with miscommunication. -Distraction management - Balancing the risk of losing a team member to manage a distressed relative, versus it impacting on the care of the young person. -Task allocation - The value of the implementation of ‘The 5A’s’ (Availability, Address, Allocate, Ask, Advise), for effective task allocation. Conclusion: 100% of participants have requested more simulation training. Involvement of airline pilots has led to a shift in hospital culture, bringing to the forefront the value of Human Factors focused training and multidisciplinary simulation. This has been of significant value in not only physical health, but also mental health simulation.

Keywords: human factors, in-situ simulation, inter-professional, multidisciplinary

Procedia PDF Downloads 108

Authors: Joseph Wolff, Jeffrey Eilbott

Abstract:

Autism Spectrum Disorders (ASD) are a spectrum of social disorders characterized by deficits in social communication, verbal ability, and interaction that can vary in severity. In recent years, researchers have used magnetic resonance imaging (MRI) to help detect how neural patterns in individuals with ASD differ from those of neurotypical (NT) controls for classification purposes. This study analyzed the classification of ASD within males and females using functional MRI data. Functional connectivity (FC) correlations among brain regions were used as feature inputs for machine learning algorithms. Analysis was performed on 558 cases from the Autism Brain Imaging Data Exchange (ABIDE) I dataset. When trained specifically on females, the algorithm underperformed in classifying the ASD subset of our testing population. Although the subject size was relatively smaller in the female group, the manual matching of both male and female training groups helps explain the algorithm’s bias, indicating the altered sex abnormalities in functional brain networks compared to typically developing peers. These results highlight the importance of taking sex into account when considering how generalizations of findings on males with ASD apply to females.

Keywords: autism spectrum disorder, machine learning, neuroimaging, sex differences

Procedia PDF Downloads 209

Authors: Antonello Durante, James Duff, Kevin Kelly

Abstract:

The visual needs of people in modern work based buildings are changing. Self-illuminated screens of computers, televisions, tablets and smart phones have changed the relationship between people and the lit environment. In the past, lighting design practice was primarily based on providing uniform horizontal illuminance on the working plane, but this has failed to ensure good quality lit environments. Lighting standards of today continue to be set based upon a 100 year old approach that at its core, considers the task illuminance of the utmost importance, with this task typically being located on a horizontal plane. An alternative method focused on appearance has been proposed, as opposed to the traditional performance based approach. Mean Room Surface Exitance (MRSE) and Target-Ambient Illuminance Ratio (TAIR) are two new metrics proposed to assess illumination adequacy in interiors. The hypothesis is that these factors will be superior to the existing metrics used, which are horizontal illuminance led. For the six past years, research has examined this, within the Dublin Institute of Technology, with a view to determining the suitability of this approach for application to general lighting practice. Since the start of this research, a number of key findings have been produced that centered on how occupants will react to various levels of MRSE. This paper provides a broad update on how this research has progressed. More specifically, this paper will: i) Demonstrate how MRSE can be measured using HDR images technology, ii) Illustrate how MRSE can be calculated using scripting and an open source lighting computation engine, iii) Describe experimental results that demonstrate how occupants have reacted to various levels of MRSE within experimental office environments.

Keywords: illumination hierarchy (IH), mean room surface exitance (MRSE), perceived adequacy of illumination (PAI), target-ambient illumination ratio (TAIR)

Procedia PDF Downloads 187

Authors: Alessandro Evangelista, Vito M. Manghisi, Michele Gattullo, Enricoandrea Laviola

Abstract:

In this work, we propose a custom visualization technique for Out-Of-View-Objects in Virtual and Augmented Reality applications using Head Mounted Displays. In the last two decades, Augmented Reality (AR) and Virtual Reality (VR) technologies experienced a remarkable growth of applications for navigation, interaction, and collaboration in different types of environments, real or virtual. Both environments can be potentially very complex, as they can include many virtual objects located in different places. Given the natural limitation of the human Field of View (about 210° horizontal and 150° vertical), humans cannot perceive objects outside this angular range. Moreover, despite recent technological advances in AR e VR Head-Mounted Displays (HMDs), these devices still suffer from a limited Field of View, especially regarding Optical See-Through displays, thus greatly amplifying the challenge of visualizing out-of-view objects. This problem is not negligible when the user needs to be aware of the number and the position of the out-of-view objects in the environment. For instance, during a maintenance operation on a construction site where virtual objects serve to improve the dangers' awareness. Providing such information can enhance the comprehension of the scene, enable fast navigation and focused search, and improve users' safety. In our research, we investigated how to represent out-of-view-objects in HMD User Interfaces (UI). Inspired by commercial video games such as Call of Duty Modern Warfare, we designed a customized Compass. By exploiting the Unity 3D graphics engine, we implemented our custom solution that can be used both in AR and VR environments. The Compass Bar consists of a graduated bar (in degrees) at the top center of the UI. The values of the bar range from -180 (far left) to +180 (far right), the zero is placed in front of the user. Two vertical lines on the bar show the amplitude of the user's field of view. Every virtual object within the scene is represented onto the compass bar as a specific color-coded proxy icon (a circular ring with a colored dot at its center). To provide the user with information about the distance, we implemented a specific algorithm that increases the size of the inner dot as the user approaches the virtual object (i.e., when the user reaches the object, the dot fills the ring). This visualization technique for out-of-view objects has some advantages. It allows users to be quickly aware of the number and the position of the virtual objects in the environment. For instance, if the compass bar displays the proxy icon at about +90, users will immediately know that the virtual object is to their right and so on. Furthermore, by having qualitative information about the distance, users can optimize their speed, thus gaining effectiveness in their work. Given the small size and position of the Compass Bar, our solution also helps lessening the occlusion problem thus increasing user acceptance and engagement. As soon as the lockdown measures will allow, we will carry out user-tests comparing this solution with other state-of-the-art existing ones such as 3D Radar, SidebARs and EyeSee360.

Keywords: augmented reality, situation awareness, virtual reality, visualization design

Procedia PDF Downloads 127

Authors: David F. Nettleton, Christian Wasiak, Jonas Dorissen, David Gillen, Alexandr Tretyak, Elodie Bugnicourt, Alejandro Rosales

Abstract:

In this work, preliminary results are given for the modeling and calibration of two inline processes, pultrusion, and laser ablation, using machine learning techniques. The end product of the processes is the core of a medical guidewire, manufactured to comply with a user specification of diameter and flexibility. An ensemble approach is followed which requires training several models. Two state of the art machine learning algorithms are benchmarked: Kernel Recursive Least Squares (KRLS) and Support Vector Regression (SVR). The final objective is to build a precise digital model of the pultrusion and laser ablation process in order to calibrate the resulting diameter and flexibility of a medical guidewire, which is the end product while taking into account the friction on the forming die. The result is an ensemble of models, whose output is within a strict required tolerance and which covers the required range of diameter and flexibility of the guidewire end product. The modeling and automatic calibration of complex in-line industrial processes is a key aspect of the Industry 4.0 movement for cyber-physical systems.

Keywords: calibration, data modeling, industrial processes, machine learning

Procedia PDF Downloads 299

Authors: Siti Nurma Hanim Hadie, Asma’ Hassan, Zul Izhar Ismail, Ahmad Fuad Abdul Rahim, Mohd. Zarawi Mat Nor, Hairul Nizam Ismail

Abstract:

Anatomy is a medical subject, which contributes to high cognitive load during learning. Despite its complexity, anatomy remains as the most important basic sciences subject with high clinical relevancy. Although anatomy knowledge is required for safe practice, many medical students graduated without having sufficient knowledge. In fact, anatomy knowledge among the medical graduates was reported to be declining and this had led to various medico-legal problems. Applying cognitive load theory (CLT) in anatomy teaching particularly lecture would be able to address this issue since anatomy information is often perceived as cognitively challenging material. CLT identifies three types of loads which are intrinsic, extraneous and germane loads, which combine to form the total cognitive load. CLT describe that learning can only occur when the total cognitive load does not exceed human working memory capacity. Hence, managing these three types of loads with the aim of optimizing the working memory capacity would be beneficial to the students in learning anatomy and retaining the knowledge for future application.

Keywords: cognitive load theory, intrinsic load, extraneous load, germane load

Procedia PDF Downloads 466

Authors: Mohamed Hamza, Alex Ohoussou

Abstract:

At JAWDA, our primary focus is on ensuring the satisfaction of our clients worldwide. We are committed to delivering new features on our SaaS platform as quickly as possible while maintaining high-quality standards. In this paper, we highlight two key aspects of testing that represent an evolution of current methods and a potential trend for the future, which have enabled us to uphold our commitment effectively. These aspects are: "One Sandbox per Pull Request" (dynamic test environments instead of static ones) and "QA for All.".

Keywords: QA for all, dynamic sandboxes, QAOPS, CICD, continuous testing, all testers, QA matters for all, 1 sandbox per PR, utilization rate, coverage rate

Procedia PDF Downloads 31

Authors: Samart Porncharoen, Nisa Pakvilai

Abstract:

The elevated levels of air pollutants in regional atmospheric environments is a significant problem that affects human health in Thailand, particularly in the Chao Phraya River Basin. Of concern are issues surrounding ambient air pollution such as particulate matter, gaseous pollutants and more specifically concerning air pollution along the river. Therefore, the spatio-temporal study of air pollution in this real environment can gain more accurate air quality data for making formalized environmental policy in river basins. In order to inform such a policy, a study was conducted over a period of January –December, 2015 to continually collect measurements of various pollutants in both urban and regional locations in the Chao Phraya River Basin. This study investigated the air pollutants in many diverse environments along the Chao Phraya River Basin, Thailand in 2015. Multivariate Analysis Techniques such as Principle Component Analysis (PCA) and Path analysis were utilised to classify air pollution in the surveyed location. Measurements were collected in both urban and rural areas to see if significant differences existed between the two locations in terms of air pollution levels. The meteorological parameters of various particulates were collected continually from a Thai pollution control department monitoring station over a period of January –December, 2015. Of interest to this study were the readings of SO2, CO, NOx, O3, and PM10. Results showed a daily arithmetic mean concentration of SO2, CO, NOx, O3, PM10 reading at 3±1 ppb, 0.5± 0.5 ppm, 30±21 ppb, 19±16 ppb, and 40±20 ug/m3 in urban locations (Bangkok). During the same time period, the readings for the same measurements in rural areas, Ayutthaya (were 1±0.5 ppb, 0.1± 0.05 ppm, 25±17 ppb, 30±21 ppb, and 35±10 ug/m3respectively. This show that Bangkok were located in highly polluted environments that are dominated source emitted from vehicles. Further, results were analysed to ascertain if significant seasonal variation existed in the measurements. It was found that levels of both gaseous pollutants and particle matter in dry season were higher than the wet season. More broadly, the results show that levels of pollutants were measured highest in locations along the Chao Phraya. River Basin known to have a large number of vehicles and biomass burning. This correlation suggests that the principle pollutants were from these anthropogenic sources. This study contributes to the body of knowledge surrounding ambient air pollution such as particulate matter, gaseous pollutants and more specifically concerning air pollution along the Chao Phraya River Basin. Further, this study is one of the first to utilise continuous mobile monitoring along a river in order to gain accurate measurements during a data collection period. Overall, the results of this study can be used for making formalized environmental policy in river basins in order to reduce the physical effects on human health.

Keywords: air pollution, Chao Phraya river basin, meteorology, seasonal variation, principal component analysis

Procedia PDF Downloads 285

Authors: M. K. Wijayarathna, S. M. Buddika Harshanath

Abstract:

Monitoring student engagement has become a crucial part of the educational process and a reliable indicator of the capacity to retain information. As online learning classrooms are now more common these days, students' attention levels have become increasingly important, making it more difficult to check each student's concentration level in an online classroom setting. To profile student attention to various gradients of engagement, a study is a plan to conduct using machine learning models. Using a convolutional neural network, the findings and confidence score of the high accuracy model are obtained. In this research, convolutional neural networks are using to help discover essential emotions that are critical in defining various levels of participation. Students' attention levels were shown to be influenced by emotions such as calm, enjoyment, surprise, and fear. An improved virtual learning system was created as a result of these data, which allowed teachers to focus their support and advise on those students who needed it. Student participation has formed as a crucial component of the learning technique and a consistent predictor of a student's capacity to retain material in the classroom. Convolutional neural networks have a plan to implement the platform. As a preliminary step, a video of the pupil would be taken. In the end, researchers used a convolutional neural network utilizing the Keras toolkit to take pictures of the recordings. Two convolutional neural network methods are planned to use to determine the pupils' attention level. Finally, those predicted student attention level results plan to display on the graphical user interface of the System.

Keywords: HTML5, JavaScript, Python flask framework, AI, graphical user

Procedia PDF Downloads 99

Authors: Sriraman Kothuri, Komal Teja Mattupalli

Abstract:

In this paper we proposed a novel algorithm for recognizing hulls in a hand written digits. This is an extension to the work on “Digit Recognition Using Freeman Chain code”. In order to find out the hulls in a user given digit it is necessary to follow three steps. Those are pre-processing, Boundary Extraction and at last apply the Hull Detection system in a way to attain the better results. The detection of Hull Regions is mainly intended to increase the machine learning capability in detection of characters or digits. This can also extend this in order to get the hull regions and their intensities in Black Holes in Space Exploration.

Keywords: chain code, machine learning, hull regions, hull recognition system, SASK algorithm

Procedia PDF Downloads 400

Authors: Sandra Arango-Caro, Kristine Callis-Duehl

Abstract:

The Education Research and Outreach Lab at the Donald Danforth Plant Science Center established the Plant and Agriculture Augmented and Virtual Reality Learning Laboratory (PAVRLL) to promote science education through professional development, school programs, internships, and outreach events. Professional development is offered to high school and college science and agriculture educators on the use and applications of zSpace and Oculus platforms. Educators learn to use, edit, or create lesson plans in the zSpace platform that are aligned with the Next Generation Science Standards. They also learn to use virtual reality experiences created by the PAVRLL available in Oculus (e.g. The Soybean Saga). Using a cost-free loan rotation system, educators can bring the AVR units to the classroom and offer AVR activities to their students. Each activity has user guides and activity protocols for both teachers and students. The PAVRLL also offers activities for 3D plant modeling. High school students work in teams of art-, science-, and technology-oriented students to design and create 3D models of plant species that are under research at the Danforth Center and present their projects at scientific events. Those 3D models are open access through the zSpace platform and are used by PAVRLL for professional development and the creation of VR activities. Both teachers and students acquire knowledge of plant and agriculture content and real-world problems, gain skills in AVR technology, 3D modeling, and science communication, and become more aware and interested in plant science. Students that participate in the PAVRLL activities complete pre- and post-surveys and reflection questions that evaluate interests in STEM and STEM careers, students’ perceptions of three design features of biology lab courses (collaboration, discovery/relevance, and iteration/productive failure), plant awareness, and engagement and learning in AVR environments. The PAVRLL was established in the fall of 2019, and since then, it has trained 15 educators, three of which will implement the AVR programs in the fall of 2021. Seven students have worked in the 3D plant modeling activity through a virtual internship. Due to the COVID-19 pandemic, the number of teachers trained, and classroom implementations have been very limited. It is expected that in the fall of 2021, students will come back to the schools in person, and by the spring of 2022, the PAVRLL activities will be fully implemented. This will allow the collection of enough data on student assessments that will provide insights on benefits and best practices for the use of AVR technologies in the classrooms. The PAVRLL uses cutting-edge educational technologies to promote science education and assess their benefits and will continue its expansion. Currently, the PAVRLL is applying for grants to create its own virtual labs where students can experience authentic research experiences using real Danforth research data based on programs the Education Lab already used in classrooms.

Keywords: assessment, augmented reality, education, plant science, virtual reality

Procedia PDF Downloads 172

Authors: Tanya Greathosue, Adrianna Taylor, Lori Darnel, Eileen Starr, Susie Ryder, Julie Clockston, Dawn Matera Bassett, Jess Retrum

Abstract:

The United States has an identified need to improve the social work mental and behavioral health workforce shortage with a focus on culturally diverse and responsive mental and behavioral health practitioners to adequately serve its rapidly growing multicultural communities. The U.S. is experiencing rapid demographic changes. Ensuring that mental and behavioral health services are effective and accessible for diverse communities is essential for improving overall health outcomes. In response to this need, we developed a training program focused on interdisciplinary collaboration, evidence-based practices, and culturally responsive services. The success of the training program, funded by the Health Resource Service Administration (HRSA) Behavioral Health Workforce Education and Training (BHWET), has provided the foundation for stage two of our programming. In addition to HRSA/BHWET, we are receiving funding from Colorado Access, a state workforce development initiative, and Kaiser Permanente, a healthcare provider network in the United States. We have moved beyond improved learner outcomes to increasing recruitment of historically excluded, disproportionately mistreated learners, mentorship of students to improve retention, and successful, culturally responsive, diverse workforce development. These authors will utilize a pretest-posttest comparison group design and trend analysis to evaluate the success of the training program. Comparison groups will be matched based on age, gender identification, race, income, as well as prior experience in the field, and time in the degree program. This article describes our culturally responsive training program. Our goals are to increase the recruitment and retention of historically excluded, disproportionately mistreated learners. We achieve this by integrating cultural humility and sensitivity training into educational curricula for our scholars who participate in cohort classroom and seminar learning. Additionally, we provide our community partners who serve as internship sites with ongoing continuing education on how to promote and develop inclusive and supportive work environments for our learners. This work will be of value to mental and behavioral health care practitioners who serve historically excluded and mistreated populations. Participants will learn about culturally informed best practices to increase recruitment and retention of culturally diverse learners. Additionally, participants will hear how to create a culturally responsive training program that encourages an inclusive community for their learners through cohort learning, mentoring, community networking, and critical accountability.

Keywords: culturally diverse mental health practitioners, recruitment, mentorship, workforce development, underserved clinics, professional development

Procedia PDF Downloads 23

Authors: Daban Mohammed Haji

Abstract:

This paper explores the application of constructive alignment in digital education at the University of Sulaimani, focusing specifically on the Language and Culture Center, Translation Department, and English Department. Constructive alignment, an outcome-based pedagogical framework developed by John Biggs, ensures that learning activities and assessments are directly aligned with the intended learning outcomes (ILOs). The study's findings reveal a significant gap in awareness and understanding of this pedagogical concept among lecturers. Many instructors are unfamiliar with constructive alignment, and those who have some knowledge of it face considerable challenges. These challenges include aligning learning activities and assessments with the ILOs and fostering higher-order cognitive skills as outlined in the SOLO taxonomy and revised Bloom’s taxonomy. To address this issue, the existing pedagogy center at the University of Sulaimani could play a pivotal role. This center has the potential to foster faculty development and promote the adoption of constructive alignment in online teaching. By leveraging the center's expertise and resources, a tailored program can be designed to enhance faculty understanding and application of this pedagogical framework.

Keywords: constructive alignment, student-centerdness, pedagogy, bologna process

Procedia PDF Downloads 33

Authors: Vahid Bairami Rad, Shapoor Norazar, Moslem Talebi Asl

Abstract:

The fast growing accessibility and capability of emerging technologies have fashioned enormous possibilities of designing, developing and implementing innovative teaching methods in the classroom. Using teaching methods and technology together have a fantastic results, because the global technological scenario has paved the way to new pedagogies in teaching-learning process. At the other side methods by focusing on students and the ways of learning in them, that can demonstrate logical ways of improving student achievement in English as a foreign language in Iran. The sample of study was 90 students of 10th grade of high school located in Ardebil. A pretest-posttest equivalent group designed to compare the achievement of groups. Students divided to 3 group, Control base, computer base, method and technology base. Pretest and post test contain 30 items each from English textbook were developed and administrated, then obtained data were analyzed. The results showed that there was an important difference. The 3rd group performance was better than other groups. On the basis of this result it was obviously counseled that teaching-learning capabilities.

Keywords: method, technology based environment, computer based environment, english as a foreign language, student achievement

Procedia PDF Downloads 474

Authors: Boadi Agyekum

Abstract:

In this study, a social location perspective is used to explore the challenges of creating opportunities that will foster lifelong education, inclusion, and equity for residents of rural communities in Ghana. The differentiated experiences of rural adults are under-researched and often unacknowledged in lifelong education literature and distance education policy. There is a need to examine carefully the structural inequalities that create disadvantages for residents of rural communities and women in pursuing distance education in designated cities in Ghana. The paper uses in-depth interviews to explore participants’ experiences of learning at a distance and to scrutinise the narratives of lifelong education. The paper reflects on the implications of the framework employed for educators and social justice in lifelong education. It further recommends the need to provide IT laboratories and fully online programs that would require stable and regular internet and access to ICT equipment for potential learning in rural communities. The social location approach presented a number of axes of diversity as comparatively more important than others; these included gender, age, education, work commitment, geography, and degree of social connectedness. This can inform lifelong education policy and programs to sustain quality education.

Keywords: equity, distance education, lifelong learning, social location, intersectionality, rural communities

Procedia PDF Downloads 99

Authors: Pranav Ajmera, Amit Kharat, Tanveer Gupte, Richa Pant, Viraj Kulkarni, Vinay Duddalwar, Purnachandra Lamghare

Abstract:

The cardiothoracic ratio (CTR) is the ratio of the diameter of the heart to the diameter of the thorax. An abnormal CTR, that is, a value greater than 0.55, is often an indicator of an underlying pathological condition. The accurate prediction of an abnormal CTR from chest X-rays (CXRs) aids in the early diagnosis of clinical conditions. We propose a deep learning-based model for automatic CTR calculation that can assist the radiologist with the diagnosis of cardiomegaly and optimize the radiology flow. The study population included 1012 posteroanterior (PA) CXRs from a single institution. The Attention U-Net deep learning (DL) architecture was used for the automatic calculation of CTR. A CTR of 0.55 was used as a cut-off to categorize the condition as cardiomegaly present or absent. An observer performance test was conducted to assess the radiologist's performance in diagnosing cardiomegaly with and without artificial intelligence (AI) assistance. The Attention U-Net model was highly specific in calculating the CTR. The model exhibited a sensitivity of 0.80 [95% CI: 0.75, 0.85], precision of 0.99 [95% CI: 0.98, 1], and a F1 score of 0.88 [95% CI: 0.85, 0.91]. During the analysis, we observed that 51 out of 1012 samples were misclassified by the model when compared to annotations made by the expert radiologist. We further observed that the sensitivity of the reviewing radiologist in identifying cardiomegaly increased from 40.50% to 88.4% when aided by the AI-generated CTR. Our segmentation-based AI model demonstrated high specificity and sensitivity for CTR calculation. The performance of the radiologist on the observer performance test improved significantly with AI assistance. A DL-based segmentation model for rapid quantification of CTR can therefore have significant potential to be used in clinical workflows.

Keywords: cardiomegaly, deep learning, chest radiograph, artificial intelligence, cardiothoracic ratio

Procedia PDF Downloads 98