Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 7 de 7
Filter
1.
4th International Conference on Innovative Computing (ICIC) ; : 806-812, 2021.
Article in English | Web of Science | ID: covidwho-1985470

ABSTRACT

The early diagnosis and treatment of COVID-19 has been a challenge all over the world. It is challenging to manufacture many testing kits and even then, their accuracy rate is very low. Studies carried out recently show that chest x-ray images are of great help in the diagnosis of COVID-19. In this study, we have developed a COVID-19 detection model that by observing the chest x-ray images of the patient, detects that either the patient is affected by COVID-19 or not. The model is developed using a custom Convolutional Neural Network (CNN) that differentiates between COVID-19 and healthy x-ray images so that the patient can be diagnosed and quarantined on time to prevent the spread of the pandemic. We used two different datasets which are publicly available for the training and validation of this model. Upon completion, the proposed model yields an accuracy of almost 98%. Upon further training, our model will be able to be used as a COVID-19 detection tool in hospitals worldwide and will play a vital role in early detection and timely containment of the pandemic.

2.
Studies in Big Data ; 109:79-113, 2022.
Article in English | Scopus | ID: covidwho-1941431

ABSTRACT

Recent Corona Virus Disease (COVID) outbreak, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV2), has been posing a big threat to global health since December 2019. In response, research community from all over the world has shifted all their efforts to contribute in this global war by providing crucial solutions. Various computer vision (CV) technologies along with other artificial intelligence (AI) subsets have significant potential to fight in frontline of this turbulent war. Normally radiologists and other clinicians are using reverse transcript polymerase chain reaction (RT-PCR) for diagnosing COVID-19, which requires strict examination environment and a set of resources. Further, this method is also prone to false negative errors. One of the potential solutions for effective and fast screening of doubtful cases is the intervention of computer vision-based support decision systems in healthcare. CT-scans, X-rays and ultrasound images are being widely used for detection, segmentation and classification of COVID-1. Computer vision is using these modalities and is providing the fast, optimal diagnosis at the early-stage controlling mortality rate. Computer vision-based surveillance technologies are also being used for monitoring physical distance, detecting people with or without face masks, screening infected persons, measuring their temperature, tracing body movements and detecting hand washing. In addition to these, it is also assisting in production of vaccine and contributing in administrative tasks and clinical management. This chapter presents an extensive study of some computer vision-based technologies for detection, diagnosis, prediction and prevention of COVID. Our main goal here is to draw a bigger picture and provide the role of computer vision in fight against COVID-19 pandemic. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

3.
5th International Conference of Women in Data Science at Prince Sultan University, WiDS-PSU 2022 ; : 91-94, 2022.
Article in English | Scopus | ID: covidwho-1874356

ABSTRACT

Social determinants of health have a major correlation to the health of a population during a pandemic. This was seen during the covid-19 pandemic when minority communities who were economically and culturally isolated were recorded with a higher rate of infection. This is compared to the average white population of the UK, where this study has taken place. Covid vaccines, primary Pfizer and AstraZeneca, have decreased infection by limiting transmission and increasing herd immunity. However, many minorities (including Muslims and middle eastern people) in the UK fail to take the vaccine and gain immunity to the disease as fears against Covid-19 vaccines grow. This issue can be further extended to the Middle East as a whole, with many countries failing to achieve herd immunity, thus leaving their population vulnerable to death and their economies to stunted growth. This could lead to millions dead and billions loss in GDP for countries like Iran, Pakistan and Turkey, who rely heavily on manual labour as their primary source of income. One of the primary reasons why Middle Eastern and North African ethnicities in the UK and middle eastern countries are failing to take vaccines is due to the conspiracies and misinforming stories they have heard regarding the Covid-19 vaccines. This paper aims to highlight the main conspiracies surrounding Covid vaccines and suggest ways in which an effective healthcare system can manage and encourage the use of vaccines within a population. This could in turn help to save the lives of many individuals and a faster recovery in income and economy for the middle east. © 2022 IEEE.

4.
15th International Conference on Information Technology and Applications, ICITA 2021 ; 350:207-217, 2022.
Article in English | Scopus | ID: covidwho-1844322

ABSTRACT

COVID-19 has been affecting people around the globe. It is affecting almost every country currently, according to the World Health Organization (WHO). This virus is transmitted to another person if an asymptomatic person makes close contact with the everyday person. There is no cure for this virus, and the only solution is social distancing and avoids the people doing these activities. In this paper, we proposed a system for detecting and recognizing the activities that violate social distancing. These activities involve handshakes and hugging. We implement a system that is capable of detecting and identifying multiple parallel activities. Temporal features are extracted for similar activities in 16 frames. We use the two models for this purpose: Faster RCNN for the detection and ResNet-50 to recognize the activities. First, Faster RCNN detects the group of people and that region of interest ROI saved and passes to the ResNet-50 to recognize the activities. We also generated our dataset on the local environment in multiple parallel activities. This system achieves the accuracy of 95.03% for the detection of testing dataset and recognition of multiple parallel activities 92.88% accuracy accomplished. The system used different public datasets and generated some local datasets for handshake and hugging activities. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

5.
Metabolism: Clinical and Experimental ; 128, 2022.
Article in English | EMBASE | ID: covidwho-1734818

ABSTRACT

Background and Objective: Non-alcoholic fatty liver disease (NAFLD) is reported to be the only hepatic ailment increasing in its prevalence concurrently with both;obesity & Type 2 Diabetes Mellitus. Abdominal ultrasonography is done for NAFLD screening diagnosis which has a high monetary cost associated with it. • In the wake of a massive strain on global health resources due to COVID-19 pandemic, NAFLD is bound to be neglected and shelved. Machine learning is explored, here, to propose screening-diagnostic tools for NAFLD that can be easily deployed without the requirement of substantial resources and can provide instantaneous screening-diagnosis predictive results. Methods: The study takes in data from Huang BX et al. : 4053 subjects, 2436 men and 1617 women between 20 and 88 years of age, after excluding those patients that had a history of co-morbid conditions as well as those with a lack of hepatic ultrasonography data. The Graif’s criteria was adopted to diagnose Fatty liver disease on ultrasonography. Mljar, the current state-of-the-art automated ML zero-code machine learning web platform, was adopted with a ‘homogenous’ approach for the development of the models vis-à-vis the preprocessing & tuning protocols as well as system specifications so as to keep the model development bias to a minimum. The discriminative ability of the models were the primary outcome variables. The ‘Area under the receiver operating curve’ (AUROC) analysis was adopted to measure that ability. Results: All 8 of the algorithms, trained in accordance with the aforementioned Homogenous Development Framework, came out to have good discriminating ability to designate the dichotomous variable of interest. Random Forest came out to have the highest discriminating ability with a computation time of minutes 9 seconds. Out of the proposed models, K-Nearest Neighbor had the least AUC but a considerably less computation time of only 6 seconds. Conclusion: Our proposed models are the very first effort, to the best of our knowledge, to leverage the current state-of-the-art for autoML to develop machine learning models that are trained to have a good discriminating ability to predict NAFLD using only anthropometric measures. The proposed models neither require costly analysis so that variables, such as ultrasonographic signals, may be fed into them for training nor do they require considerably high computation time & resources to be deployed. A study comparing the presented models’ predicted diagnosis with an abdominal ultrasound diagnosis for NAFLD, the predictions subsequently assessed against hepatic biopsy, is proposed to be in order to explore the presented models’ potential to replace abdominal ultrasound as a cost-effective screening diagnostic modality for NAFLD. Keywords: Non-alcoholic fatty liver disease, Machine learning, prediction, abdominal ultrasonography, Diabetes mellitus Abbreviations: NAFLD: Non-alcoholic fatty liver disease;autoML: Automated machine learning Funding and Conflicts of Interest None

6.
International Journal for Research in Applied Science and Engineering Technology ; 8(5):1053-1064, 2020.
Article in English | GIM | ID: covidwho-831069

ABSTRACT

This paper proposes Machine Learning based Methodology to assist Health staff to perform bulk reporting of patients on Chest X-ray images into Normal or Pneumonia diseased clusters which will be of assistance to currently overburdened health workers and possibly detect potential covid19 infected patients as pneumonia is known symptom of covid19. Also this paper demonstrates creating high accuracy models trained on existing clustered data capable of accurately predicting pneumonia in patients.

SELECTION OF CITATIONS
SEARCH DETAIL