scholarly journals COVID-XIX-Net: Deep learning empirical comparison between X-ray imaging and POCUS for COVID-19 detection

2021 ◽  
Vol 9 (4A) ◽  
Author(s):  
Marwa Kandil ◽  
◽  
Ali Kelkawi ◽  
Imtiaz Ahmad ◽  
Mohammed Al-Failakawi ◽  
...  

The novel COVID-19 virus has been spreading vigorously through the world, starting a pandemic that was never experienced before in our modern era. It is an infectious disease caused by severe acute respiratory syndrome and carries symptoms such as cough, fever, and shortness of breath. In March 2020, the World Health Organization recognized COVID-19 as a pandemic, with more than 130 million cases in over 200 countries and over 2.8 million deaths since its discovery. With a limited number of test kits available worldwide and the rapid spread of the disease on a daily basis, alternative means of detection are needed. The use of X-ray imaging, CT scans, and lung point-of-care ultrasound (POCUS) facilitated early diagnosis of COVID-19 cases. In this study, we leverage InceptionV3 and ResBlocks in building a deep convolutional neural network model, COVID-XIX-Net, to aid in the detection of COVID-19 positive cases through the detection of pneumonic patterns in chest X-ray images and ultrasound scans. COVID-XIX-Net is a multiclass classification model that classifies images into one of 3 classes: healthy, bacterial pneumonia, and COVID-19-induced pneumonic lungs. The proposed model architecture aims at accurately diagnosing COVID-19 cases while maintaining a low number of parameters. COVID-XIX-Net is tested on a balanced X-ray dataset composed of 1,011 images, and an imbalanced ultrasound dataset composed of 1,103 images. After training, cross-validation, and testing, COVID-XIX-Net achieved an accuracy of 99.9% with precision and recall of 0.99 for the X-ray dataset, and an accuracy, precision, and recall of 89.9%, 0.97, and 0.90, respectively, for the case of ultrasound dataset. Results are compared against recent literature showing promising results and great potential with only 24.6M parameters. This work can be further developed and trained to assist medical practitioners in diagnosing COVID-19 cases.

2020 ◽  
Author(s):  
Solomzi Makoliso ◽  
Bertrand Klaiber ◽  
Romain Sahli ◽  
Jean Roger Moulion Tapouh ◽  
Samuel Nko'o Amvene ◽  
...  

Technologies that have been designed for use in high-income countries often fail to deliver their full potential when transposed to Low and Middle-Income Contexts (LMICs). The health sector is a case in point, as medical devices, whether donated or purchased, are generally short lived in those contexts. The mismatch between needs and available solutions originates from the inadequacy of both the technology and the business models. Essential medical technologies such as oxygen concentrators, neonatal incubators, anesthesia machines or diagnostic X-ray systems are classic examples. The case of diagnostic X-ray imaging is particularly striking: 125 years after its invention, up to two thirds of the world population still does not have access to radiology services, according to the World Health Organisation. This is despite the fact that X-ray radiology is one of the cornerstone of healthcare and a crucial instrument for diagnosing a variety of health issues ranging from trauma to tuberculosis and other lung diseases.We are presenting an integrated methodological approach, to develop innovative solutions adapted to the context of LMICs. The approach relies on three crucial pillars: cooperation, interdisciplinarity and entrepreneurship with a long-term sustainability perspective. We propose a set of four complementary tools that increase the chances of successfully developing and deploying the technologies at scale. The tools, while very practical, allow striking a balance between economic viability, environmental and social impact. We illustrate the use of these tools with the case of diagnostic X-ray imaging. We propose that using the approach and tools presented here could allow to rethink other complex technologies that have the potential to address social challenges, in the perspective of making them suitable for LMICs. We also believe that this approach to developing solutions addressing the needs of poorer communities, may lead to better products in industrialized contexts as well.


2021 ◽  
Vol 8 (Supplement_1) ◽  
pp. S94-S95
Author(s):  
Ingrid Y Camelo ◽  
Rachel Pieciak ◽  
Ilse castro-aragon ◽  
Bindu Setty ◽  
Lauren Etter ◽  
...  

Abstract Background Childhood pneumonia is one of the leading causes of death in low-income countries. The diagnosis of pediatric pneumonia is a critical epidemiological duty for treatment effectiveness and vaccine surveillance. Previous studies have demonstrated an important lack in correlation between CXR findings and the clinical WHO case definition of severe pneumonia. Lung Point of Care Ultrasound (POCUS) has demonstrated in multiple studies to be more sensitive and specific for diagnosing pneumonia in the pediatric population. With no exposure to radiation, extensive availability in limited-resource settings, and easy interpretation, this modality can be a breakpoint in making a more accurate correlation between pneumonia clinical findings and diagnostic imaging. Methods 50 children from 1-59 months meeting the WHO case definition of severe pneumonia were enrolled at the Emergency Department at University Teaching Hospital (UTH) in Lusaka, Zambia. Children underwent lung POCUS and CXR. Correlation between symptoms and all abnormalities (consolidation, effusion, and interstitial patterns) seen in both imaging modalities were analyzed by calculating the proportion of children with abnormalities on CXR and ultrasound. Each participant was assigned a score based on findings. 0 = normal, 1 = consolidation only, 2 = Consolidation and non-consolidation (interstitial and/or effusion) and 3 = non-consolidation (interstitial and/or effusion) only. Results 44 (90%) of children had abnormalities on CXR and 46 (94%) on POCUS. Five children (10%) had normal findings on CXR vs 3 (6%) on Lung POCUS. 4 (8%) had consolidation only on CXR vs 0 (0%) on POCUS. 19 (39%) had consolidation and non-consolidation (interstitial and/or effusion) on CXR vs. 20 (41%) on POCUS. 21 (43%) had non-consolidation (interstitial and/or effusion) only on CXR vs. 26 (53%) on POCUS. Figure 1. Scores Asigned Based on Imaging Findings for CXR and Lung POCUS Figure 2. Chest X Ray Anterior Posterior (AP) view showing Bilareral Interstitial Pattern Figure 3. Lung POCUS (Point of Care Ultrasound) findings of bilateral Consolidation and non-consolidation pattern and bilateral interstitial pattern (only finding on CXR) Conclusion More children with clinical pneumonia had normal findings on CXR than on POCUS. POCUS was a better imaging technique to show consolidation and non-consolidation patterns than CXR. The higher proportion of children diagnosed with consolidation and non-consolidation patterns on POCUS suggest that CXR might not be the ideal gold standard to diagnose pneumonia in children. Disclosures All Authors: No reported disclosures


Author(s):  
Ossama Maadarani ◽  
Zouheir Bitar ◽  
Tamer Zaalouk ◽  
Mohammad Mohsen ◽  
Ragab Elshabasy

Coronavirus disease 2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the World Health Organization (WHO) declared it a pandemic on 11 March 2020. Point-of-care ultrasound (POCUS) is a real-time bedside tool used by physicians to guide rapid, focused and accurate evaluation in order to identify or rule out various pathologies. We describe the case of an elderly man who had fallen at home 3 days previously and was hypoxic at presentation to the emergency department (ED). POCUS in the ED helped to identify a combination of lung and vascular involvement that indicated COVID-19 infection, which was confirmed by a laboratory test.


2021 ◽  
Vol 13 (1) ◽  
Author(s):  
Matthew G. Hanson ◽  
Barry Chan

Abstract Background Symptomatic pericardial effusion (PCE) presents with non-specific features and are often missed on the initial physical exam, chest X-ray (CXR), and electrocardiogram (ECG). In extreme cases, misdiagnosis can evolve into decompensated cardiac tamponade, a life-threatening obstructive shock. The purpose of this study is to evaluate the impact of point-of-care ultrasound (POCUS) on the diagnosis and therapeutic intervention of clinically significant PCE. Methods In a retrospective chart review, we looked at all patients between 2002 and 2018 at a major Canadian academic hospital who had a pericardiocentesis for clinically significant PCE. We extracted the rate of presenting complaints, physical exam findings, X-ray findings, ECG findings, time-to-diagnosis, and time-to-pericardiocentesis and how these were impacted by POCUS. Results The most common presenting symptom was dyspnea (64%) and the average systolic blood pressure (SBP) was 120 mmHg. 86% of people presenting had an effusion > 1 cm, and 89% were circumferential on departmental echocardiogram (ECHO) with 64% having evidence of right atrial systolic collapse and 58% with early diastolic right ventricular collapse. The average time-to-diagnosis with POCUS was 5.9 h compared to > 12 h with other imaging including departmental ECHO. Those who had the PCE identified by POCUS had an average time-to-pericardiocentesis of 28.1 h compared to > 48 h with other diagnostic modalities. Conclusion POCUS expedites the diagnosis of symptomatic PCE given its non-specific clinical findings which, in turn, may accelerate the time-to-intervention.


2021 ◽  
Vol 5 (5) ◽  
pp. 1-17
Author(s):  
Azuonwu O

Background: Coronavirus disease-2019 was first discovered in a highly populated city of China in late 2019, and has since spread to most countries of the world, causing several morbidities and mortalities. In the bid to contain the disease and curtail its spread, different countries have instituted several policies, and while these policies may work for some countries, it may not work for others. Nevertheless, the disease has affected over 19 million people globally, killing as many as 700,000. Yet, thousands of persons are still being infected on a daily basis. Aim: To x-ray and evaluate the suitability of home-based treatment/care of COVID-19 patients in Nigeria. Methodology: Peer-reviewed articles revealing information about COVID-19 and its effects globally were sourced from different electronic databases (including WHO, Pub Med, Science Direct, NCDC, etc.), and appraised to extract valuable data and information from them for the purpose of analysis and synthesis of developing robust body of knowledge. Findings: The results obtained from our search include some details about COVID-19 infection, the disease epidemiology, diagnosis, management and guidelines for home-based treatment of COVID-19 patients. Also, discussed in this study are some loopholes in the Nigerian health system and leadership that makes it difficult for foreign policies or strategies (on COVID-19 containment) to be implemented in Nigeria. Conclusion/Recommendation: Differences in lifestyles and cultures among different countries of the world means that there is no one-size-fits-all solution to the problems created by the emergence of the COVID-19 pandemic. Hence, each country is advised to determine which policies best suit the lifestyles and cultures peculiar to her inhabitants. There is literally no room for copy and paste syndrome


Author(s):  
Oksana Rybachok

According to the World Health Organisation, deafness is one of the most widely spread sensory disorders in the world affecting about 360 million people worldwide. The causes of deafness can be very diverse, from genetic diseases, the impact of injury-risk factor and infectious agents to the administration of ototoxic drugs. Moreover, otolaryngologists believe that about half of deafness and hearing loss cases could have been prevented. Though otolaryngology was separated as an independent medical science in the mid-18th century, the decision to celebrate the Otolaryngologist Day on September 29 as a professional holiday for medical practitioners in this speciality was made not so long ago. This date at the end of September was chosen on purpose: the influx of patients to medical practitioners in this speciality is observed closer to the mid-autumn, after the first cold snap.


2004 ◽  
Vol 14 (2) ◽  
pp. 145-153 ◽  
Author(s):  
Caroline McGraw ◽  
Vari Drennan

The issue of not taking medicines as prescribed by medical practitioners has a history as long as the medical profession itself. The World Health Organization recently described the problem of patients diagnosed with chronic illnesses not taking their medication as prescribed as ‘a worldwide problem of striking magnitude’. Not taking medicines as prescribed has consequences not only for the individual in terms of therapeutic failure, but also for the wider society. For the individual, failure to take medication as prescribed may result in ill health, poorer quality of life, and reduced life expectancy. For the wider society, consequences include avoidable health care expenditure and the development of drug resistance.


CJEM ◽  
2015 ◽  
Vol 18 (5) ◽  
pp. 391-394
Author(s):  
Michael Romano ◽  
Tomislav Jelic ◽  
Jordan Chenkin

AbstractThere is evidence to suggest that point-of-care ultrasound assessment of the lungs has a higher sensitivity and specificity than chest radiography for the diagnosis of pneumonia. It is unknown if the same is true for pneumonia complications. We present and discuss the case of a 61-year-old woman who presented to the emergency department with confusion, decreased level of consciousness, and signs of sepsis. A chest x-ray revealed a right sided infiltrate. An ultrasound of the patient’s lungs was performed, and revealed a complex loculated fluid collection consistent with an empyema. A chest CT confirmed the diagnosis, and immediate percutaneous drainage was performed.


1999 ◽  
Vol 34 (2) ◽  
pp. 305-316 ◽  
Author(s):  
E.H. Bakraji ◽  
J. Karajo

Abstract Total reflection X-ray fluorescence spectrometry and chemical preconcentration have been applied for multi-elemental analysis of Damascus drinking water. Water was taken directly from taps of several city sectors and analyzed for the following trace elements: Ti, V, Cr, Fe, Co, Ni, Cu, Zn, Se and Pb. The detection limits were found to be in the range of 0.1 to 0.4 µg/L. The mean levels of trace elements in the Damascus drinking water were below the World Health Organization drinking water quality guidelines.


Sign in / Sign up

Export Citation Format

Share Document