BERTHop: An Effective Vision-and-Language Model for Chest X-ray Disease Diagnosis

The coronavirus pandemic (COVID-19) is disrupting the entire world; its rapid global spread threatens to affect millions of people. Accurate and timely diagnosis of COVID-19 is essential to control the spread and alleviate risk. Due to the promising results achieved by integrating machine learning (ML), particularly deep learning (DL), in automating the multiple disease diagnosis process. In the current study, a model based on deep learning was proposed for the automated diagnosis of COVID-19 using chest X-ray images (CXR) and clinical data of the patient. The aim of this study is to investigate the effects of integrating clinical patient data with the CXR for automated COVID-19 diagnosis. The proposed model used data collected from King Fahad University Hospital, Dammam, KSA, which consists of 270 patient records. The experiments were carried out first with clinical data, second with the CXR, and finally with clinical data and CXR. The fusion technique was used to combine the clinical features and features extracted from images. The study found that integrating clinical data with the CXR improves diagnostic accuracy. Using the clinical data and the CXR, the model achieved an accuracy of 0.970, a recall of 0.986, a precision of 0.978, and an F-score of 0.982. Further validation was performed by comparing the performance of the proposed system with the diagnosis of an expert. Additionally, the results have shown that the proposed system can be used as a tool that can help the doctors in COVID-19 diagnosis.

Download Full-text

Precise Prediction of COVID-19 in Chest X-Ray Images Using KE Sieve Algorithm

10.1101/2020.08.13.20174144 ◽

2020 ◽

Author(s):

S Sai Thejeshwar ◽

Chaitanya Chokkareddy ◽

K Eswaran

Keyword(s):

Ct Scan ◽

Network Architecture ◽

Intelligent Systems ◽

Imaging Techniques ◽

Health Care Systems ◽

Prediction Method ◽

Lung Infection ◽

Disease Diagnosis ◽

X Ray ◽

Chest X Ray

The novel coronavirus (COVID-19) pandemic is pressurizing the healthcare systems across the globe and few of them are on the verge of failing. The detection of this virus as early as possible will help in contaminating the spread of it as the virus is mutating itself as fast as possible and currently there are about 4,300 strains of the virus according to the reports. Clinical studies have shown that most of the COVID-19 patients suffer from a lung infection similar to influenza. So, it is possible to diagnose lung infection using imaging techniques. Although a chest computed tomography (CT) scan has been shown to be an effective imaging technique for lung-related disease diagnosis, chest X-ray is more widely available across the hospitals due to its considerably lower cost and faster imaging time than CT scan. The advancements in the area of machine learning and pattern recognition has resulted in intelligent systems that analyze CT Scans or X-ray images and classify between pneumonia and normal patients. This paper proposes KE Sieve Neural Network architecture, which helps in the rapid diagnosis of COVID-19 using chest X-ray images. This architecture is achieving an accuracy of 98.49%. This noninvasive prediction method can assist the doctors in this pandemic and reduce the stress on health care systems.

Download Full-text

A deep learning framework to detect Covid-19 disease via chest X-ray and CT scan images

International Journal of Electrical and Computer Engineering (IJECE) ◽

10.11591/ijece.v11i1.pp844-850 ◽

2021 ◽

Vol 11 (1) ◽

pp. 844

Author(s):

Mohammed Y. Kamil

Keyword(s):

Deep Learning ◽

High Performance ◽

Early Stage ◽

Diagnostic Technique ◽

Disease Diagnosis ◽

X Ray ◽

Learning Framework ◽

Chest X Ray ◽

Low Sensitivity ◽

Deep Learning Model

COVID-19 disease has rapidly spread all over the world at the beginning of this year. The hospitals' reports have told that low sensitivity of RT-PCR tests in the infection early stage. At which point, a rapid and accurate diagnostic technique, is needed to detect the Covid-19. CT has been demonstrated to be a successful tool in the diagnosis of disease. A deep learning framework can be developed to aid in evaluating CT exams to provide diagnosis, thus saving time for disease control. In this work, a deep learning model was modified to Covid-19 detection via features extraction from chest X-ray and CT images. Initially, many transfer-learning models have applied and comparison it, then a VGG-19 model was tuned to get the best results that can be adopted in the disease diagnosis. Diagnostic performance was assessed for all models used via the dataset that included 1000 images. The VGG-19 model achieved the highest accuracy of 99%, sensitivity of 97.4%, and specificity of 99.4%. The deep learning and image processing demonstrated high performance in early Covid-19 detection. It shows to be an auxiliary detection way for clinical doctors and thus contribute to the control of the pandemic.

Download Full-text

AB0170 RHEUMATOID ARTHRITIS ASSOCIATED LUNG DISEASE: EXPERIENCE IN A BIOLOGICAL THERAPY UNIT

Annals of the Rheumatic Diseases ◽

10.1136/annrheumdis-2021-eular.3178 ◽

2021 ◽

Vol 80 (Suppl 1) ◽

pp. 1111.2-1112

Author(s):

L. Vega ◽

I. Calvo ◽

O. Ibarguengoitia ◽

D. Montero ◽

C. García ◽

...

Keyword(s):

Rheumatoid Arthritis ◽

Lung Disease ◽

Interstitial Pneumonia ◽

Gold Standard ◽

Biological Therapy ◽

Pulmonary Involvement ◽

Disease Diagnosis ◽

Classification Criteria ◽

X Ray ◽

Chest X Ray

Background:Rheumatoid arthritis (RA) associated lung disease is a relatively frequent extra articular disease manifestation, with a prevalence between 5% and 30%. The rather wide range of estimated prevalence is a result of differences in study designs and studied populations, as well as lacking diagnostic and classification criteria for lung disease in patients with RA.Objectives:To evaluate the prevalence of RA associated lung disease in patients with biological therapy (BT), as well as its severity, treatment changes and possible associated factors.Methods:Review of clinical records of 257 patients with RA treated with BT (TNFi, non-TNFi) between January 2015 to December 2020 in a single center. Patients with preexisting lung disease for other causes (asthma, smoking) have been excluded. RA diagnosis was performed according to ACR 2010 classification criteria. Epidemiological variables, clinical characteristics, type of pulmonary involvement, evolution, type of BT, changes in treatment and concomitant treatment were collected. For the analysis frequencies and percentages are used in qualitative variables, and mean ± SD in the quantitative ones. Statistical analysis was performed with IBM SPSS v 23.Results:We registered 21 patients (85.7% women) mean aged 70.3±11.9 years. 52.4% were never smokers. RF was positive in 100% and 20 patients were anti-CCP positive. Erosive disease was present in 13 (61.9%) patients.At the time of lung disease diagnosis, 15 patients (66.7%) were receiving TNFi (Etanercept 7, Adalimumab 6, Infliximab 1, Golimumab 1), 2 were with non-TNFi (Rituximab) and 4 had never received BT previously. Symptoms (cough and/or dyspnea) were reported in 10 (47.6%) patients. The median time of treatment with BT until lung disease diagnosis was 33 [15.5-95.5] months. Conventional synthetic DMARDs (csDMARDs) were used in 85.7% of cases (methotrexate 72.2%, leflunomide 22.2%, other 5.6%). The inflammatory activity was mild (DAS28: 3.22±1.6). The median time until lung disease diagnosis was 104 [56.2-156] months.After the lung disease diagnosis, BT was only modified in 1 patient. In the 4 patients who had not previously received BT, non-TNFi was started (Rituximab 2, Abatacept 1, Tocilizumab 1). csDMARD was discontinued in 1 patient.Interstitial lung disease (ILD) was the most frequent pulmonary involvement (16 patients, 76.2%): 8 usual interstitial pneumonia (UIP), 6 non-specific interstitial pneumonia (NSIP), 1 organising pneumonia (OP) and 1 lymphocytic interstitial pneumonia (LIP). Other pulmonary manifestations observed in our patients were: nodular lung disease (2 patients) and small airways disease (bronchiectasis 2, obliterative bronchiolitis 1). Chest x-ray was normal in almost half of the patients (42.9%). Gold standard image diagnostic technique was high resolution CT.In respiratory function tests (PFTs) at diagnosis, only 4 patients (19%) had a FVC<80% and 4 (19%) a DLCO<60%. In the following 2 years, in 2 patients the FVC worsened > 10% and in 5 there was a worsening of the DLCO > 15%. In 3 (14.3%) patients PFTs were never performed and in 7 (43.7%) were not repeated after the diagnosis.We haven´t found association between different types of pulmonary involvement and the variables analysed.Conclusion:In our series, prevalence of RA associated lung disease is similar to that described in the literature. Lung involvement is asymptomatic and chest X-ray is normal in most RA patients. High resolution CT is the gold standard for diagnosis.ILD was the most frequent pulmonary involvement. Although in most patients the diagnosis of lung disease did not imply a BT change, it had an influence on the type of BT chosen for those who started treatment. Maintenance of csDMARD was not associated with a worsening of lung disease.Screening and treatment protocols for lung disease in patients with RA in clinical practice are needed.Disclosure of Interests:None declared

Download Full-text

Fusion of multi-scale bag of deep visual words features of chest X-ray images to detect COVID-19 infection

Scientific Reports ◽

10.1038/s41598-021-03287-8 ◽

2021 ◽

Vol 11 (1) ◽

Author(s):

Chiranjibi Sitaula ◽

Tej Bahadur Shahi ◽

Sunil Aryal ◽

Faezeh Marzbanrad

Keyword(s):

Classification Accuracy ◽

Semantic Information ◽

Disease Diagnosis ◽

Support Vector ◽

X Ray ◽

Visual Words ◽

Multi Scale ◽

Single Scale ◽

Chest X Ray ◽

Public Datasets

AbstractChest X-ray (CXR) images have been one of the important diagnosis tools used in the COVID-19 disease diagnosis. Deep learning (DL)-based methods have been used heavily to analyze these images. Compared to other DL-based methods, the bag of deep visual words-based method (BoDVW) proposed recently is shown to be a prominent representation of CXR images for their better discriminability. However, single-scale BoDVW features are insufficient to capture the detailed semantic information of the infected regions in the lungs as the resolution of such images varies in real application. In this paper, we propose a new multi-scale bag of deep visual words (MBoDVW) features, which exploits three different scales of the 4th pooling layer’s output feature map achieved from VGG-16 model. For MBoDVW-based features, we perform the Convolution with Max pooling operation over the 4th pooling layer using three different kernels: $$1 \times 1$$ 1 × 1 , $$2 \times 2$$ 2 × 2 , and $$3 \times 3$$ 3 × 3 . We evaluate our proposed features with the Support Vector Machine (SVM) classification algorithm on four CXR public datasets (CD1, CD2, CD3, and CD4) with over 5000 CXR images. Experimental results show that our method produces stable and prominent classification accuracy (84.37%, 88.88%, 90.29%, and 83.65% on CD1, CD2, CD3, and CD4, respectively).

Download Full-text

Underdiagnosis bias of artificial intelligence algorithms applied to chest radiographs in under-served patient populations

Nature Medicine ◽

10.1038/s41591-021-01595-0 ◽

2021 ◽

Cited By ~ 1

Author(s):

Laleh Seyyed-Kalantari ◽

Haoran Zhang ◽

Matthew B. A. McDermott ◽

Irene Y. Chen ◽

Marzyeh Ghassemi

Keyword(s):

Artificial Intelligence ◽

Medical Imaging ◽

Low Socioeconomic Status ◽

Disease Diagnosis ◽

Low Socioeconomic ◽

X Ray ◽

Female Patients ◽

Pathology Classification ◽

Black Patients ◽

Chest X Ray

AbstractArtificial intelligence (AI) systems have increasingly achieved expert-level performance in medical imaging applications. However, there is growing concern that such AI systems may reflect and amplify human bias, and reduce the quality of their performance in historically under-served populations such as female patients, Black patients, or patients of low socioeconomic status. Such biases are especially troubling in the context of underdiagnosis, whereby the AI algorithm would inaccurately label an individual with a disease as healthy, potentially delaying access to care. Here, we examine algorithmic underdiagnosis in chest X-ray pathology classification across three large chest X-ray datasets, as well as one multi-source dataset. We find that classifiers produced using state-of-the-art computer vision techniques consistently and selectively underdiagnosed under-served patient populations and that the underdiagnosis rate was higher for intersectional under-served subpopulations, for example, Hispanic female patients. Deployment of AI systems using medical imaging for disease diagnosis with such biases risks exacerbation of existing care biases and can potentially lead to unequal access to medical treatment, thereby raising ethical concerns for the use of these models in the clinic.

Download Full-text

AI-based Diagnosis of COVID-19 Patients Using X-ray Scans with Stochastic Ensemble of CNNs

10.36227/techrxiv.12464402.v1 ◽

2020 ◽

Cited By ~ 1

Author(s):

Ridhi Arora ◽

Vipul Bansal ◽

Himanshu Buckchash ◽

Rahul Kumar ◽

Vinodh J Sahayasheela ◽

...

Keyword(s):

Comprehensive Evaluation ◽

State Of The Art ◽

Empirical Evaluation ◽

Main Idea ◽

Feature Space ◽

Disease Diagnosis ◽

X Ray ◽

Main Challenge ◽

Latent Space ◽

Chest X Ray

<div>According to WHO, COVID-19 is an infectious disease and has a significant social and economic impact. The main challenge in ?fighting against this disease is its scale. Due to the imminent outbreak, the medical facilities are over exhausted and unable to accommodate the piling cases. A quick diagnosis system is required to address these challenges. To this end, a stochastic deep learning model is proposed. The main idea is to constrain the deep representations over a gaussian prior to reinforce the discriminability in feature space. The model can work on chest X-ray or CT-scan images. It provides</div><div>a fast diagnosis of COVID-19 and can scale seamlessly. This work presents a comprehensive evaluation of previously proposed approaches for X-ray based</div><div>disease diagnosis. Our approach works by learning a latent space over X-ray image distribution from the ensemble of state-of-the-art convolutional-nets,</div><div>and then linearly regressing the predictions from an ensemble of classifi?ers which take the latent vector as input. We experimented with publicly available datasets having three classes { COVID-19, normal, Pneumonia. Moreover, for robust evaluation, experiments were performed on a large chest X-ray dataset with fi?ve different very similar diseases. Extensive empirical evaluation shows</div><div>how the proposed approach advances the state-of-the-art.</div>

Download Full-text

Chest X-ray interpretation

British Journal of Cardiac Nursing ◽

10.12968/bjca.2019.0004 ◽

2019 ◽

Vol 14 (5) ◽

pp. 1-9

Author(s):

Susan Dawkes ◽

Michelle O'Reilly

Keyword(s):

Cardiovascular Disease ◽

Low Cost ◽

Disease Diagnosis ◽

Medical Intervention ◽

X Rays ◽

Early Medical Intervention ◽

X Ray ◽

Anatomy And Physiology ◽

Patient Will ◽

Chest X Ray

The chest X-ray is a common, low-cost investigation that is an important aid in cardiovascular disease diagnosis. Although newer, more sophisticated modalities of imaging are available, chest X-rays remain fundamental, first-line investigations used to determine patient care. Although a radiologist should report all X-rays, nurses and other health professionals frequently examine and interpret chest X-rays. A sound knowledge of normal anatomy and physiology is fundamental. The technical quality of the chest X-ray, specifically the projection of the film, its orientation, rotation of the patient and penetration of the X-rays is important to determine. To prevent abnormalities going undetected when examining chest X-rays, a comprehensive systematic approach to assessment and interpretation is essential. Common abnormalities seen on X-rays from patients with cardiovascular disease have specific differentiating features and these, combined with the physical examination of the patient, will allow for faster diagnosis and early medical intervention.

Download Full-text