scholarly journals An Explainable approach to Deep Learning from CT-scans for Covid Identification

2021 ◽  
Author(s):  
Eduardo Soares ◽  
Plamen Angelov ◽  
Ziyang Zhang
Keyword(s):  
Deep Learning ◽  
World Health ◽  
Learning Approaches ◽  
Heat Map ◽  
Algorithm Performance ◽  
Decision Mechanism ◽  
The World ◽  
Public Health Officials ◽  
Decision Structure ◽  
Health Organization

The Covid-19 disease has spread widely over the whole world since the beginning of 2020. Following the epidemic which started in Wuhan, China on January 30, 2020 the World Health Organization (WHO) declared a global health emergency and a pandemic. Researchers of different disciplines work along with public health officials to understand the SARS-CoV-2 pathogenesis and jointly with the policymakers urgently develop strategies to control the spread of this new disease. Recent findings have observed specific image patterns from computed tomography (CT) for patients infected by SARS-CoV-2 which are distinct from the other pulmonary diseases. In this paper, we propose an explainable-by-design that has an integrated image segmentation mechanism based on SLIC that improves the algorithm performance and the interpretability of the resulting model. In order to evaluate the proposed approach, we used the SARS-CoV-2 CT scan dataset that we published recently and has been widely used in the literature. The proposed Super-xDNN could obtain statistically better results than traditional deep learning approaches as DenseNet-201 and Resnet-152. Furthermore, it also improved the explainability and interpretability of its decision mechanism when compared with the xDNN basis approach that uses the whole image as prototype. The segmentation mechanism of Super-xDNN favored a decision structure that is more close to the human logic. Moreover, it also allowed the provision of new insights as a heat-map which highlights the areas with highest similarities with Covid-19 prototypes, and an estimation of the area affected by the disease.

scholarly journals An Explainable approach to Deep Learning from CT-scans for Covid Identification

2021 ◽  
Author(s):  
Eduardo Soares ◽  
Plamen Angelov ◽  
Ziyang Zhang
Keyword(s):  
Deep Learning ◽  
World Health ◽  
Learning Approaches ◽  
Heat Map ◽  
Algorithm Performance ◽  
Decision Mechanism ◽  
The World ◽  
Public Health Officials ◽  
Decision Structure ◽  
Health Organization

The Covid-19 disease has spread widely over the whole world since the beginning of 2020. Following the epidemic which started in Wuhan, China on January 30, 2020 the World Health Organization (WHO) declared a global health emergency and a pandemic. Researchers of different disciplines work along with public health officials to understand the SARS-CoV-2 pathogenesis and jointly with the policymakers urgently develop strategies to control the spread of this new disease. Recent findings have observed specific image patterns from computed tomography (CT) for patients infected by SARS-CoV-2 which are distinct from the other pulmonary diseases. In this paper, we propose an explainable-by-design that has an integrated image segmentation mechanism based on SLIC that improves the algorithm performance and the interpretability of the resulting model. In order to evaluate the proposed approach, we used the SARS-CoV-2 CT scan dataset that we published recently and has been widely used in the literature. The proposed Super-xDNN could obtain statistically better results than traditional deep learning approaches as DenseNet-201 and Resnet-152. Furthermore, it also improved the explainability and interpretability of its decision mechanism when compared with the xDNN basis approach that uses the whole image as prototype. The segmentation mechanism of Super-xDNN favored a decision structure that is more close to the human logic. Moreover, it also allowed the provision of new insights as a heat-map which highlights the areas with highest similarities with Covid-19 prototypes, and an estimation of the area affected by the disease.

scholarly journals Between art and information: communicating world health, 1948–70

2018 ◽  
Vol 13 (1) ◽  
pp. 94-120 ◽  
Author(s):  
Alexander Medcalf
Keyword(s):  
New Media ◽  
Public Information ◽  
World Health ◽  
Central Position ◽  
Media Technologies ◽  
The World ◽  
Public Health Officials ◽  
Health Organization ◽  
Health Situation ◽  
The Impact

AbstractWith the advent of new media technologies and approaches in the twentieth century, public health officials became convinced that health needed mass media support. The World Health Organization believed that educating people, as well as informing them about the health situation around the world, could assist in the enduring fight against disease. Yet in an increasingly competitive media landscape, the agency recognized the need to persuade people and hold their attention through attractive presentation. Public information, the name given to the multiple strategies used to communicate with the public, was rarely straightforward and required the agency not only to monitor the impact of its own efforts but also to identify opportunities to further enhance its reputation, especially when this was in danger of damage or misappropriation. The WHO’s understanding of public information provides insights into the development of international information, communication, and education networks and practices after 1945, as well as the increasingly central position of these processes in generating support for and evincing the value of international organizations.

scholarly journals Deep learning for classification of pediatric chest radiographs by WHO’s standardized methodology

PLoS ONE ◽  
2021 ◽  
Vol 16 (6) ◽  
pp. e0253239
Author(s):  
Yiyun Chen ◽  
Craig S. Roberts ◽  
Wanmei Ou ◽  
Tanaz Petigara ◽  
Gregory V. Goldmacher ◽  
...  
Keyword(s):  
Deep Learning ◽  
World Health Organization ◽  
Characteristic Curve ◽  
Learning Model ◽  
Observer Agreement ◽  
World Health ◽  
Widespread Application ◽  
The World ◽  
Health Organization ◽  
Deep Learning Model

Background The World Health Organization (WHO)-defined radiological pneumonia is a preferred endpoint in pneumococcal vaccine efficacy and effectiveness studies in children. Automating the WHO methodology may support more widespread application of this endpoint. Methods We trained a deep learning model to classify pneumonia CXRs in children using the World Health Organization (WHO)’s standardized methodology. The model was pretrained on CheXpert, a dataset containing 224,316 adult CXRs, and fine-tuned on PERCH, a pediatric dataset containing 4,172 CXRs. The model was then tested on two pediatric CXR datasets released by WHO. We also compared the model’s performance to that of radiologists and pediatricians. Results The average area under the receiver operating characteristic curve (AUC) for primary endpoint pneumonia (PEP) across 10-fold validation of PERCH images was 0.928; average AUC after testing on WHO images was 0.977. The model’s classification performance was better on test images with high inter-observer agreement; however, the model still outperformed human assessments in AUC and precision-recall spaces on low agreement images. Conclusion A deep learning model can classify pneumonia CXR images in children at a performance comparable to human readers. Our method lays a strong foundation for the potential inclusion of computer-aided readings of pediatric CXRs in vaccine trials and epidemiology studies.

scholarly journals Designing for deep learning in the context of digital and social media

Comunicar ◽  
2019 ◽  
Vol 27 (58) ◽  
pp. 9-18 ◽  
Author(s):  
James-Paul Gee ◽  
Moisés Esteban-Guitart
Keyword(s):  
Social Networks ◽  
Social Media ◽  
Deep Learning ◽  
World Health ◽  
Human Beings ◽  
Redes Sociales ◽  
Living Things ◽  
The World ◽  
Mind And Body ◽  
Health Organization

There is today a great deal of controversy over digital and social media. Even leaders in the tech industry are beginning to decry the time young people spend on smartphones and social networks. Recently, the World Health Organization proposed adding “gaming disorder” to its official list of diseases, defining it as a pattern of gaming behavior so severe that it takes “precedence over other life interests”. At the same time, many others have celebrated the positive properties of video games, social media, and social networks. This paper argues that a deeper understanding of human beings is needed to design for deep learning. For the purposes of this study “design for deep learning” means helping people matter and find meaning in ways that make them and others healthy in mind and body, while improving the state of the world for all living things, with due respect for truth, sensation, happiness, imagination, individuality, diversity, and the future. In particular, fifteen features related to human nature are suggested based on recent scientific developments to answer the question: What is a human being? Consequently, proposals that are linked to learning and transformation, as well as social improvement, should fit with the ways in which humans, as specific sorts of biological and social creatures, learn best (or can learn at all) and can change for the better. En la actualidad existe una nutrida controversia en relación a los medios de comunicación sociales y digitales que ha llevado, incluso, a censurar la utilización de las redes sociales y los móviles por parte de líderes en la industria tecnológica. En este sentido, la Organización Mundial para la Salud ha propuesto añadir el «desorden del juego» a su listado de enfermedades, definiéndolo como un modelo de comportamiento de juego tan severo que se impone como «preferencia sobre otros intereses». Al mismo tiempo, distintos académicos han enfatizado los aspectos positivos derivados de las redes sociales y los videojuegos. En este artículo se argumenta que es necesaria una mejor comprensión del ser humano para poder implementar lo que aquí se define como diseño para el aprendizaje profundo. El «diseño para el aprendizaje profundo» está encaminado al reconocimiento de las personas y el desarrollo de sentidos saludables, individual y colectivamente, así como la mejora, en general, del estado del mundo para todos los seres vivos, según principios de verdad, felicidad, imaginación, individualidad, diversidad y futuro. En particular, se sugieren quince características basadas en desarrollos científicos que responden a la pregunta: ¿Qué es un ser humano? Consecuentemente, propuestas vinculadas al aprendizaje y la transformación y mejora social deben ser coherentes con dichas características que permiten definir cómo las personas, en tanto que organismos biológicos y sociales, aprenden o pueden aprender óptimamente, así como cambiar para mejorar.

scholarly journals Autonomic Face Mask Detection with Deep Learning: an IoT Application

2021 ◽  
Author(s):  
◽  
V. H. Benitez-Baltazar
Keyword(s):  
Deep Learning ◽  
Detection System ◽  
Face Mask ◽  
World Health ◽  
Image Tracking ◽  
Critical Areas ◽  
Machine Learning Model ◽  
The World ◽  
The Face ◽  
Health Organization

A new and deadly virus known as SARS-CoV-2, which is responsible for the coronavirus disease (COVID-19), is spreading rapidly around the world causing more than 3 million deaths. Hence, there is an urgent need to find new and innovative ways to reduce the likelihood of infection. One of the most common ways of catching the virus is by being in contact with droplets delivered by a sick person. The risk can be reduced by wearing a face mask as suggested by the World Health Organization (WHO), especially in closed environments such as classrooms, hospitals, and supermarkets. However, people hesitate to use a face mask leading to an increase in the risk of spreading the disease, moreover when the face mask is used, sometimes it is worn in the wrong way. In this work, an autonomic face mask detection system with deep learning and powered by the image tracking technique used for the augmented reality development is proposed as a mechanism to request the correct use of face masks to grant access to people to critical areas. To achieve this, a machine learning model based on Convolutional Neural Networks was built on top of an IoT framework to enforce the correct use of the face mask in required areas as it is requested by law in some regions.

scholarly journals Internet Gaming Disorder: Prevention Adolescent Programs for Indonesia

2021 ◽  
Vol 9 (1) ◽  
pp. 50
Author(s):  
Kristina Dwi Novitasari Arnani
Keyword(s):  
Internet Gaming Disorder ◽  
World Health ◽  
Gaming Disorder ◽  
Internet Gaming ◽  
The World ◽  
Public Health Officials ◽  
Internet Games ◽  
Health Organization ◽  
Use Of Internet ◽  
Disorder Prevention

Background: Emerging Internet technologies are now creeping into the game arena. Increased incidence of gaming addiction is felt in the world, and no doubt in Indonesia could have an impact as well, especially in an adolescent. In Makassar, found the incidence of internet games disorders by 30% in high school children. Therefore, internet games eventually became an important issue in the world of health to the WHO (World Health Organization) and making it the responsibility of the world. The state has a duty and responsibility in preventing health problems caused by the development of internet gaming in Indonesia. Internet Gaming Disorder is a mental problem that should be considered in adolescents, and even no single governing restrictions on the use of internet gaming and prevention programs for adolescents in Indonesia. Purpose: The purpose of this study is to explore the problem of Internet Gaming disorder by describing programs that have been implemented by countries outside Indonesia in terms of health promotion for adolescents. Methods: This study was a literature review of several journals, thesis, as well as patient data reports Internet Gaming disorder in Indonesia and the world. Result: The result is a necessary regulation involving adolescents, parents, schools, governments, and public health officials to regulate Internet gaming restrictions to prevent Internet Gaming Disorder as has been done in China, Hong Kong, Iran, and Switzerland which can be adopted in Indonesia. Conclusion: The problem of Internet gaming disorder being ordered must be a concern of government and cross-sectoral to prevent the development of this problem in Indonesia as a protective way for adolescents.

scholarly journals Deep Learning Classification for Diabetic Foot Thermograms

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1762 ◽  
Author(s):  
Israel Cruz-Vega ◽  
Daniel Hernandez-Contreras ◽  
Hayde Peregrina-Barreto ◽  
Jose de Jesus Rangel-Magdaleno ◽  
Juan Manuel Ramirez-Cortes
Keyword(s):  
Deep Learning ◽  
Diabetic Foot ◽  
World Health ◽  
Diabetic Patients ◽  
Complex Data ◽  
Thermal Change ◽  
The World ◽  
Plantar Ulcers ◽  
Health Organization

According to the World Health Organization (WHO), Diabetes Mellitus (DM) is one of the most prevalent diseases in the world. It is also associated with a high mortality index. Diabetic foot is one of its main complications, and it comprises the development of plantar ulcers that could result in an amputation. Several works report that thermography is useful to detect changes in the plantar temperature, which could give rise to a higher risk of ulceration. However, the plantar temperature distribution does not follow a particular pattern in diabetic patients, thereby making it difficult to measure the changes. Thus, there is an interest in improving the success of the analysis and classification methods that help to detect abnormal changes in the plantar temperature. All this leads to the use of computer-aided systems, such as those involved in artificial intelligence (AI), which operate with highly complex data structures. This paper compares machine learning-based techniques with Deep Learning (DL) structures. We tested common structures in the mode of transfer learning, including AlexNet and GoogleNet. Moreover, we designed a new DL-structure, which is trained from scratch and is able to reach higher values in terms of accuracy and other quality measures. The main goal of this work is to analyze the use of AI and DL for the classification of diabetic foot thermograms, highlighting their advantages and limitations. To the best of our knowledge, this is the first proposal of DL networks applied to the classification of diabetic foot thermograms. The experiments are conducted over thermograms of DM and control groups. After that, a multi-level classification is performed based on a previously reported thermal change index. The high accuracy obtained shows the usefulness of AI and DL as auxiliary tools to aid during the medical diagnosis.

scholarly journals A Comparative Analysis of Variant Deep Learning Models for COVID-19 Protective Face Mask Detection

2021 ◽  
Vol 12 (6) ◽  
pp. 2841-2848
Author(s):  
Rohan Katari Et al.
Keyword(s):  
Deep Learning ◽  
Comparative Analysis ◽  
Disease Transmission ◽  
Detection System ◽  
Face Mask ◽  
Public Spaces ◽  
World Health ◽  
Learning Models ◽  
The World ◽  
Health Organization

The world is in the midst of a paramount pandemic owing to the rapid dissemination of coronavirus disease (COVID-19) brought about by the spread of the virus ‘SARS-CoV-2’. It is mainly transmitted among persons through airborne diffusion of droplets containing the virus produced by an infected person sneezing or coughing without covering their face. The World Health Organization (WHO) has issued numerous guidelines which state that the spread of this disease can be limited by people shielding their faces with protective face masks when in public or in crowded areas. As a precautionary measure, many nations have implemented obligations for face mask usage in public spaces. But manual monitoring of huge crowds in public spaces for face masks is laborious. Hence, this requires the development of an automated face mask detection system using deep learning models and related technologies. The detection system should be viable and deployable in real-time, predicting the result accurately so as to be used by monitoring bodies to ensure that the face mask guidelines are followed by the public thereby preventing the disease transmission. In this paper we aim to perform a comparative analysis of various sophisticated image classifiers based on deep learning, in terms of vital metrics of performance to identify the effective deep learning based model for face mask detection.

scholarly journals Automatically identify COVID-19 and Pneumonia patients using Deep learning in Chest X-rays image

2021 ◽  
Vol 12 (3) ◽  
pp. 011-019
Author(s):  
Haris Uddin Sharif ◽  
Shaamim Udding Ahmed
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Infected Patient ◽  
World Health ◽  
X Rays ◽  
Popular Approach ◽  
X Ray ◽  
The World ◽  
Chest X Ray ◽  
Health Organization

At the end of 2019, a new kind of coronavirus (SARS-CoV-2) suffered worldwide and has become the pandemic coronavirus (COVID-19). The outbreak of this virus let to crisis around the world and kills millions of people globally. On March 2020, WHO (World Health Organization) declared it as pandemic disease. The first symptom of this virus is identical to flue and it destroys the human respiratory system. For the identification of this disease, the first key step is the screening of infected patients. The easiest and most popular approach for screening of the COVID-19 patients is chest X-ray images. In this study, our aim to automatically identify the COVID-19 and Pneumonia patients by the X-ray image of infected patient. To identify COVID19 and Pneumonia disease, the convolution Neural Network was training on publicly available dataset on GitHub and Kaggle. The model showed the 98% and 96% training accuracy for three and four classes respectively. The accuracy scores showed the robustness of both model and efficiently deployment for identification of COVID-19 patients.

scholarly journals Automated Diagnosis of Chest X-Ray for Early Detection of COVID-19 Disease

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Ebrahim Mohammed Senan ◽  
Ali Alzahrani ◽  
Mohammed Y. Alzahrani ◽  
Nizar Alsharif ◽  
Theyazn H. H. Aldhyani
Keyword(s):  
Deep Learning ◽  
Data Augmentation ◽  
Area Under The Curve ◽  
World Health ◽  
X Ray ◽  
Medical Sector ◽  
The World ◽  
Health Organization ◽  
Many Sources ◽  
Sensitivity Specificity

In March 2020, the World Health Organization announced the COVID-19 pandemic, its dangers, and its rapid spread throughout the world. In March 2021, the second wave of the pandemic began with a new strain of COVID-19, which was more dangerous for some countries, including India, recording 400,000 new cases daily and more than 4,000 deaths per day. This pandemic has overloaded the medical sector, especially radiology. Deep-learning techniques have been used to reduce the burden on hospitals and assist physicians for accurate diagnoses. In our study, two models of deep learning, ResNet-50 and AlexNet, were introduced to diagnose X-ray datasets collected from many sources. Each network diagnosed a multiclass (four classes) and a two-class dataset. The images were processed to remove noise, and a data augmentation technique was applied to the minority classes to create a balance between the classes. The features extracted by convolutional neural network (CNN) models were combined with traditional Gray-level Cooccurrence Matrix (GLCM) and Local Binary Pattern (LBP) algorithms in a 1-D vector of each image, which produced more representative features for each disease. Network parameters were tuned for optimum performance. The ResNet-50 network reached accuracy, sensitivity, specificity, and Area Under the Curve (AUC) of 95%, 94.5%, 98%, and 97.10%, respectively, with the multiclasses (COVID-19, viral pneumonia, lung opacity, and normal), while it reached accuracy, sensitivity, specificity, and AUC of 99%, 98%, 98%, and 97.51%, respectively, with the binary classes (COVID-19 and normal).

Sign in / Sign up

Export Citation Format

Share Document