Simple Chest X-ray Research Articles

Methodological evaluation of systematic reviews based on the use of artificial intelligence systems in chest radiography

IntroductionIn recent years, systems that use artificial intelligence (AI) in medical imaging have been developed, such as the interpretation of chest X-ray to rule out pathology. This has produced an increase in systematic reviews (SR) published on this topic. This article aims to evaluate the methodological quality of SRs that use AI for the diagnosis of thoracic pathology by simple chest X-ray. Material and methodsSRs evaluating the use of AI systems for the automatic reading of chest X-ray were selected. Searches were conducted (from inception to May 2022): PubMed, EMBASE, and the Cochrane Database of Systematic Reviews. Two investigators selected the reviews. From each SR, general, methodological and transparency characteristics were extracted. The PRISMA statement for diagnostic tests (PRISMA-DTA) and AMSTAR-2 were used. A narrative synthesis of the evidence was performed. Protocol registry: Open Science Framework: https://osf.io/4b6u2/. ResultsAfter applying the inclusion and exclusion criteria, 7 SRs were selected (mean of 36 included studies per review). All the included SRs evaluated “deep learning” systems in which chest X-ray was used for the diagnosis of infectious diseases. Only 2 (29%) SRs indicated the existence of a review protocol. None of the SRs specified the design of the included studies or provided a list of excluded studies with their justification. Six (86%) SRs mentioned the use of PRISMA or one of its extensions. The risk of bias assessment was performed in 4 (57%) SRs. One (14%) SR included studies with some validation of AI techniques. Five (71%) SRs presented results in favour of the diagnostic capacity of the intervention. All SRs were rated "critically low" following AMSTAR-2 criteria. ConclusionsThe methodological quality of SRs that use AI systems in chest radiography can be improved. The lack of compliance in some items of the tools used means that the SRs published in this field must be interpreted with caution.

Evaluación metodológica de las revisiones sistemáticas basadas en la utilización de sistemas de inteligencia artificial en radiografía de tórax

IntroductionIn recent years, systems that use artificial intelligence (AI) in medical imaging have been developed, such as the interpretation of chest X-ray to rule out pathology. This has produced an increase in systematic reviews (SR) published on this topic. This article aims to evaluate the methodological quality of SRs that use AI for the diagnosis of thoracic pathology by simple chest X-ray. Material and methodsSRs evaluating the use of AI systems for the automatic reading of chest X-ray were selected. Searches were conducted (from inception to May 2022): PubMed, EMBASE, and the Cochrane Database of Systematic Reviews. Two investigators selected the reviews. From each SR, general, methodological and transparency characteristics were extracted. The PRISMA statement for diagnostic tests (PRISMA-DTA) and AMSTAR-2 were used. A narrative synthesis of the evidence was performed. Protocol registry: Open Science Framework: https://osf.io/4b6u2/. ResultsAfter applying the inclusion and exclusion criteria, 7 SRs were selected (mean of 36 included studies per review). All the included SRs evaluated “deep learning” systems in which chest X-ray was used for the diagnosis of infectious diseases. Only 2 (29%) SRs indicated the existence of a review protocol. None of the SRs specified the design of the included studies or provided a list of excluded studies with their justification. Six (86%) SRs mentioned the use of PRISMA or one of its extensions. The risk of bias assessment was performed in 4 (57%) SRs. One (14%) SR included studies with some validation of AI techniques. Five (71%) SRs presented results in favour of the diagnostic capacity of the intervention. All SRs were rated “critically low” following AMSTAR-2 criteria. ConclusionsThe methodological quality of SRs that use AI systems in chest radiography can be improved. The lack of compliance in some items of the tools used means that the SRs published in this field must be interpreted with caution.

Abstract 11189: An Integrated Framework With Machine Learning and Radiomics Score for the Prediction of Coronary Artery Calcium Score From Chest Radiographs

Introduction: Recently introduced radiomics technology enables the detection of hidden information from commonly disregarded medical images, including CXR. Hypothesis: The purpose of this study was to evaluate the feasibility of a novel technique involving radiomics features combined with machine learning to identify moderate-to-severe coronary artery calcium (CAC) using simple chest X-ray radiography (CXR). Methods: We included 559 patients (women, 44.9%; mean age, 62.4 ± 9.4 years) from two independent clinical studies who underwent a calcium scan and CXR within 6 months. The total cohort was allocated to the training and validation cohorts in a 7:3 ratio, with all clinical characteristics well-matched, including the CAC score. Radiomics features were extracted from manually delineated cardiac contours, and a radiomics score formulation for the prediction of a CAC score ≥100 was generated using the machine learning method in the training cohort. To evaluate the incremental performance of the radiomics score, a basic clinical model including age, sex, and body mass index (model 1) and a radiomics score added model (model 2) were utilized. Results: The radiomics score was the most prominent predictive factor for CAC score ≥100 (odds ratio [OR] = 2.33; 95% confidence interval [CI] = 1.62-3.44; p < 0.001). In the training cohort, model 2 demonstrated significant incremental validity in predicting CAC scores ≥100 compared to model 1 (area under the curve [AUC]; 0.73 vs. 0.69, p = 0.022). The performance of model 2 was also similar in both the training and validation cohorts (AUC 0.73, 95% confidence interval [CI] 0.68 - 0.78 vs. AUC 0.72, 95% CI 0.64 - 0.80). Conclusions: We developed a machine learning-based radiomics scoring model that could be utilized as a potential imaging marker for predicting CAC scores from CXR. This novel method may be widely applicable to clinical practice and can improve the pre-test probability of coronary artery disease.

When less is more in thoracic surgery

For more than 25 years, a lobectomy rather than a sublobar resection has been regarded the best surgical treatment of early-stage non-small-cell lung cancer (NSCLC) in patients fit for surgery. This recommendation followed the interpretation of the only clinical trial so far (LCSG821), which randomised between the two surgeries and found superior recurrence-free survival.1 LCSG821 has since been criticised because it was performed when the diagnosis of lung cancer was based on simple chest x-rays performed for symptoms.

Chest x-ray scoring as a predictor of COVID-19 disease; correlation with comorbidities and in-hospital mortality.

To devise a novel, simple chest x-ray (CXR) scoring system which would help in prognosticating the disease severity and ability to predict comorbidities and in-hospital mortality. We included a total of 343 consecutive hospitalised patients with COVID-19 in this study. The chest x-rays of these patients were scored retrospectively by three radiologists independently. We divided CXR in to six zones (right upper, mid & lower and left, upper mid & lower zones). We scored each zone as- 0, 1 or 2 as follows- if that zone was clear (0) Ground glass opacity (1) or Consolidation (2). A total of score from 0 to 12 could be obtained. A CXR score cut off ≥3 independently predicted mortality. Along with a relatively higher NPV ≥80%, it reinforced the importance of CXR score is a screening tool to triage patients according to risk of mortality. We propose that Pennine score is a simple tool which can be adapted by various countries, experiencing a large surge in number of patients, to decide which patient would need a tertiary Hospital referral/admission as opposed to patients that can be managed locally or at basic/primary care hospitals.

Employing Texture Features of Chest X-Ray Images and Machine Learning in COVID-19 Detection and Classification

The novel coronavirus (nCoV-19) was first detected in December 2019. It had spread worldwide and was declared coronavirus disease (COVID-19) pandemic by March 2020. Patients presented with a wide range of symptoms affecting multiple organ systems predominantly the lungs. Severe cases required intensive care unit (ICU) admissions while there were asymptomatic cases as well. Although early detection of the COVID-19 virus by Real-time reverse transcription-polymerase chain reaction (RT-PCR) is effective, it is not efficient; as there can be false negatives, it is time consuming and expensive. To increase the accuracy of in-vivo detection, radiological image-based methods like a simple chest X-ray (CXR) can be utilized. This reduces the false negatives as compared to solely using the RT-PCR technique. This paper employs various image processing techniques besides extracted texture features from the radiological images and feeds them to different artificial intelligence (AI) scenarios to distinguish between normal, pneumonia, and COVID-19 cases. The best scenario is then adopted to build an automated system that can segment the chest region from the acquired image, enhance the segmented region then extract the texture features, and finally, classify it into one of the three classes. The best overall accuracy achieved is 93.1% by exploiting Ensemble classifier. Utilizing radiological data to conform to a machine learning format reduces the detection time and increase the chances of survival.

A simple x-ray scoring system for the diagnosis of chronic pulmonary aspergillosis.

Chronic pulmonary aspergillosis (CPA) is a severe form of post-tuberculosis lung disease (PTBLD). Considering the high burden of TB in India, it can be concluded that the prevalence of CPA is also high. Chest x-ray though most feasible, interpretation is subjective. Therefore, decision on evaluation for CPA cannot be based on x-ray alone. Present study evaluated an x-ray score as a marker for extent of lung damage in patients with PTBLD presenting with haemoptysis and its utility to predict Aspergillus serum IgG levels. We used a modified scoring system developed by Anna Ralph et al X-ray score cut-offs of >71 and 40, with or without history of massive haemoptysis, were compared with serum IgG levels. With a chest x-ray score cut-off of 71, specificity was 88%. With an x-ray score of >71 combined with history of massive haemoptysis, 86% cases were found to be IgG positive. The specificity of this combination was 96%. This study concluded that a simple chest x-ray scoring system in addition to the symptom of massive haemoptysis helped in the decision on further evaluation of the subject for CPA, especially in resource constrained settings.

ANALYZING CHEST X-RAY TO DIFFERENTIATE SYMPTOMS OF COVID-19 THROUGH ML APPROACH

The motive behind the project is to build a machine learning model for detection of Covid-19. Using this model, it is possible to classify images of chest x-rays into normal patients, pneumatic patients, and covid-19 positive patients. This CNN based model will help drastically to save time constraints among the patients. Instead of relying on limited RT-PCR kits, just a simple chest x-ray can help us determine health of the patient. Not only we get immediate results, but we can also practice social distancing norms more effectively.

A tale of ptosis, pharmacological tests, and Pancoast tumour

Horner syndrome (HS) classically presents with ipsilateral blepharoptosis, pupillary miosis, and facial anhidrosis and caused by a lesion along the oculosympathetic pathway from the hypothalamus to eye. The diagnosis of HS in a patient presenting with partial ptosis may be easily missed in the Asian patient. This is mainly due to the dark irides, making detection of anisocoria on direct visualization difficult. Index of suspicion must be high, especially in the absence of any extraocular motility or lid abnormalities. We present a case where a healthy asymptomatic patient presented with partial ptosis and diagnosis of HS was eventually confirmed through pharmacological tests. Non-targeted imaging with a simple chest X-ray revealed an apical lung lesion which eventually turned out to be malignant. Although it is a typical textbook description, this case highlights the importance of careful history and examination in an otherwise healthy patient presenting with mild ptosis.

Chest radiography for simplified evaluation of central venous catheter tip positioning for safe and accurate haemodynamic monitoring: a retrospective observational study

ObjectivesThe tip-to-carina (TC) distance on a simple chest X-ray (CXR) has proven value in the determination of correct central venous catheter (CVC) positioning. However, previous studies have mostly focused on...

DELAYED DIAGNOSIS OF EXTRAPULMONARY TUBERCULOSIS PRESENTING AS FEVER OF UNKNOWN ORIGIN IN BIHAR

Background: Tuberculosis (TB), especially extrapulmonary tuberculosis (EPTB), is an important cause of fever of unknown origin (FUO) in Bihar. Little information is known about patients with EPTB with clinical features presenting as FUO and about the factor of delaying the diagnosis. Material and Methods: We retrospectively analyzed EPTB patients at DMCH, Laheriasarai, Bihar, who were referred with FUO fromMithilanchal area around like; Darbhanga, Madhubani, Samsatipur and other places. The subjects were assigned to groups of early diagnosis and delayed diagnosis within3 days of an initial comprehensive evaluation from the referral. Clinical and laboratory variables were compared between the groups. Results: A total of 95 patients with febrile EPTB were included. Localizing symptoms and/or signs suggestive of anatomy were identified in 62.1% of the patients. Concurrent lung involvement by TB was presented by 49.5% (47/95) of the patients, and only 23.4% of them showed typical findings of pulmonary TB on simple chest X-ray. Most of the patients showed abnormal lesions on cross-sectional CT (98.9%) and MRI (100%). The clinical variables and blood test results of patients were not significantly different between the two groups. The less typical imaging finding of EPTB on CT (38.5% vs. 79.0%) and MRI (37.5% vs. 79.0%) in the delayed diagnosis group was a risk factor for delayed diagnosis. Conclusion: Febrile EPTB referred as FUO showed nonspecific clinical manifestations. The active application of cross-sectional imaging tests according to clinical clues or randomly in the absence of local manifestations, combined with invasive diagnostic approaches even for atypical presentations may lead to an earlier diagnosis of febrile EPTB.

Unveiling COVID-19 from CHEST X-Ray with Deep Learning: A Hurdles Race with Small Data.

The possibility to use widespread and simple chest X-ray (CXR) imaging for early screening of COVID-19 patients is attracting much interest from both the clinical and the AI community. In this study we provide insights and also raise warnings on what is reasonable to expect by applying deep learning to COVID classification of CXR images. We provide a methodological guide and critical reading of an extensive set of statistical results that can be obtained using currently available datasets. In particular, we take the challenge posed by current small size COVID data and show how significant can be the bias introduced by transfer-learning using larger public non-COVID CXR datasets. We also contribute by providing results on a medium size COVID CXR dataset, just collected by one of the major emergency hospitals in Northern Italy during the peak of the COVID pandemic. These novel data allow us to contribute to validate the generalization capacity of preliminary results circulating in the scientific community. Our conclusions shed some light into the possibility to effectively discriminate COVID using CXR.

A rare chest X-ray sign in the diagnosis of an intracardiac bullet: case report

BackgroundGunshot injuries to the heart are often seen in cardiac surgical practice. Victims are usually investigated with various types of imaging modalities, including simple chest X-ray (CXR) which is a primary imaging modality that can easily locate a bullet and is useful in monitoring before surgical intervention as bullets may migrate.Case presentationA 32-year-old man was admitted as an emergency in a primary hospital after being injured by a gunshot to the chest. The bullet entry was over the back of the left shoulder with no exit point seen. The chest X-ray showed left pleural effusion and a retained intrathoracic bullet. A chest drain was inserted and drained 2200 ml of blood. Computed tomography (CT) scan showed an intracardiac bullet associated with left lower lobe contusion and left-sided hemothorax. He was urgently transferred to theater for exploration via left thoracotomy. The bullet could not be found so the inlet point in the left ventricle was sutured. After stabilizing the patient, he was referred to a specialized cardiac center for further management.The patient arrived at our center 2 days after the injury, fully conscious and hemodynamically stable. On arrival, chest X-rays were obtained to rule out possible migration of the bullet and revealed that the retained bullet was still within the cardiac silhouette. The X-ray appearance of the bullet showed a characteristically double contour, strongly suggesting that the bullet had lodged in the heart muscle and was moving with each heartbeat.The patient was transferred to the theater for median sternotomy. The aorta, superior vena cava, and inferior vena cava were cannulated, and cardiopulmonary bypass was initiated. The aorta was cross-clamped and the cardioplegia was given. Palpation of the still heart readily identified the bullet within the interventricular septum. The left ventricle was opened 1 cm from the left anterior descending artery just on top of the bullet. The bullet was successfully retrieved. The bullet was transfixing the septum causing a small ventricular septal defect which was closed using a Teflon patch and the ventricle was repaired.ConclusionThe double contour appearance of the bullet indicates that the bullet is moving and strongly suggests an intramyocardial position.

Clinical Course and Outcomes of Patients with Severe Acute Respiratory Syndrome Coronavirus 2 Infection: a Preliminary Report of the First 28 Patients from the Korean Cohort Study on COVID-19.

BackgroundSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-infected pneumonia emerged in Wuhan, China in December 2019. In this retrospective multicenter study, we investigated the clinical course and outcomes of novel coronavirus disease 2019 (COVID-19) from early cases in Republic of Korea.MethodsAll of the cases confirmed by real time polymerase chain reaction were enrolled from the 1st to the 28th patient nationwide. Clinical data were collected and analyzed for changes in clinical severity including laboratory, radiological, and virologic dynamics during the progression of illness.ResultsThe median age was 40 years (range, 20–73 years) and 15 (53.6%) patients were male. The most common symptoms were cough (28.6%) and sore throat (28.6%), followed by fever (25.0%). Diarrhea was not common (10.7%). Two patients had no symptoms. Initial chest X-ray (CXR) showed infiltration in 46.4% of the patients, but computed tomography scan confirmed pneumonia in 88.9% (16/18) of the patients. Six patients (21.4%) required supplemental oxygen therapy, but no one needed mechanical ventilation. Lymphopenia was more common in severe cases. Higher level of C-reactive protein and worsening of chest radiographic score was observed during the 5–7 day period after symptom onset. Viral shedding was high from day 1 of illness, especially from the upper respiratory tract (URT).ConclusionThe prodromal symptoms of COVID-19 were mild and most patients did not have limitations of daily activity. Viral shedding from URT was high from the prodromal phase. Radiological pneumonia was common from the early days of illness, but it was frequently not evident in simple CXR. These findings could be plausible explanations for the easy and rapid spread of SARS-CoV-2 in the community.

Colecistite calculosa em situs inversus totalis: abordagem laparoscópica

Situs inversus é uma anomalia congênita rara caracterizada pela transposição dos órgãos abdominais, vísceras e vasculatura, em relação ao sentido sagital, oferecendo uma imagem chamada de “imagem no espelho”. É classificado como total quando associado à dextrocardia, enquanto o situs inversus parcial afeta apenas um ou mais órgãos. Analisar um caso de anomalia congênita rara de situs inversus totalis submetido à colecistectomia com colangiografia por abordagem laparoscópica, para tratamento de colecistite calculosa. Paciente de sexo feminino, 42 anos, apresentava episódios de dor abdominal em cólica, localizada em epigástrio e hipocôndrio esquerdo, sendo diagnosticada a colecistite calculosa em situs inversus abdominal. Com a confirmação do diagnóstico de situs inversus totalis por radiografia simples de tórax, realizou-se a colecistectomia com colangiografia videolaparoscópica. A experiência do médico no manejo instrumental e do bloco operatório da imagem espelhada do situs inversus, proporciona uma colecistectomia laparoscópica segura.

Effect of pleurodesis with negative pressure suction plus bleomycin on malignant pleural effusions; a clinical trial

Introduction: About 50% of patients with metastatic cancers suffer from malignant pleural effusion. The goals of treatment for these patients should be to relieve pain, restore normal function, reduce or eliminate hospitalization, and make efficient use of medical care resources. Objectives: This study aimed to evaluate the effect of negative pressure suction on the success rate of pleurodesis with bleomycin on malignant pleural effusion. Patients and Methods: This study was conducted as before-after interventional study, without randomization. For all patients, 1 mg/kg bleomycin diluted with 50 mL distilled water, was injected into the chest tube. After changing the position of the patient, the clamp was opened and the tube was connected to the central suction with two chest bottles. The negative pressure suction was 20 cmH2O. Before intervention and four weeks after the intervention, a simple decubitus lateral chest X-ray was taken and the amount of malignant pleural effusion was assessed and compared with the amount of pleural effusion before the insertion of the chest tube. If the pleural effusion level was greater than 10 mm from the outer part of the chest, pleurodesis was considered unsuccessful Results: The mean age of patients was 60.44 ± 10.48 years (32-79 years); of which 12 patients (48%) were male. The most common cancer was lung cancer (48%). The success rate of treatment with bleomycin accompanied by negative pressure suction was 80% without a significant relationship with age, gender and type of cancer. Additionally, the results showed the improvement of the respiratory status and the pain level after the pleurodesis. The only side effect after bleomycin injection was fever in 84% of patients. Conclusion: Bleomycin treatment accompanied by negative pressure suction can improve malignant pleural effusion and this method is recommended for these patients.

Delayed diagnosis of extrapulmonary tuberculosis presenting as fever of unknown origin in an intermediate-burden country

BackgroundTuberculosis (TB), especially extrapulmonary tuberculosis (EPTB), is an important cause of fever of unknown origin (FUO) in TB-burdened areas. Little information is known about patients with EPTB with clinical features presenting as FUO and about the factor of delaying the diagnosis.MethodsWe retrospectively analyzed EPTB patients who were referred with FUO at 3 university-affiliated hospitals over 8 years (2010–2017). The subjects were assigned to groups of early diagnosis and delayed diagnosis within 3 days of an initial comprehensive evaluation from the referral. Clinical and laboratory variables were compared between the groups.ResultsA total of 95 patients with febrile EPTB were included. Localizing symptoms and/or signs suggestive of anatomy were identified in 62.1% of the patients. Concurrent lung involvement by TB was presented by 49.5% (47/95) of the patients, and only 23.4% of them showed typical findings of pulmonary TB on simple chest X-ray. Most of the patients showed abnormal lesions on cross-sectional CT (98.9%) and MRI (100%). The clinical variables and blood test results of patients were not significantly different between the two groups. The less typical imaging finding of EPTB on CT (38.5% vs. 79.0%) and MRI (37.5% vs. 79.0%) in the delayed diagnosis group was a risk factor for delayed diagnosis.ConclusionFebrile EPTB referred as FUO showed nonspecific clinical manifestations. The active application of cross-sectional imaging tests according to clinical clues or randomly in the absence of local manifestations, combined with invasive diagnostic approaches even for atypical presentations may lead to an earlier diagnosis of febrile EPTB.

Protocolo de diagnóstico y tratamiento de la valvulopatía aórtica

Aortic stenosis. Diagnostic evaluationClinical evaluation, baseline ECG, and chest x-ray (Rx) are mandatory in the initial diagnostic evaluation. Transthoracic echocardiography (TTE) is the main tool for diagnosis and follow-up. Other diagnostic techniques are also useful. Therapeutic managementPatients with severe symptomatic aortic stenosis should be involved, especially with aortic valve replacement surgery, and if the surgical risk is high, transcatheter aortic valve implantation (TAVI) may be an adequate alternative. Asymptomatic individuals will only be involved according to several factors predictive of onset of symptoms and adverse events. Aortic insufficiency. Diagnostic assessmentclinical evaluation, baseline ECG, and simple chest x-ray are also mandatory in the initial diagnostic evaluation. The ETT is the main diagnostic and monitoring tool. Therapeutic managementAcute severe aortic insufficiency (AI) should undergo urgent surgery. In chronic severe AI, when symptoms and/or systolic dysfunction of the left ventricle appear, aortic valve surgery must be performed; The asymptomatic patients will be operated according to several factors regarding left ventricular systolic size and function. Sometimes aortic root and ascending aorta surgery must be performed irrespective of the severity of the AI.

Unrecognised diaphragmatic hernia in a refugee child: an incidental diagnosis

A 4-year-old boy from Syria was evaluated at the emergency department because of an upper airway viral illness. His physical examination showed a significant intensity reduction of all heart sounds...

Ruptured pulmonary hydatid cyst: a case report.

Ruptured pulmonary hydatid cyst (PHC) is an important clinical problem in endemic areas to echinococcal infection. Herein we present a rare case of ruptured PHC in an adolescent boy that was misdiagnosed as pulmonary tuberculosis in local health center. When sputum specimen was stained by acid-fast staining for detection of Mycobacterium tuberculosis, hooklets of Echinococcus granulosus were observed. A simple chest X-ray showed a multilobulated mass in the lower part of the left lung. Computed tomography scan verified existence of thick walled caviar lesion with irregular air-fluid level. The diagnosis was confirmed at the time of surgery. Misdiagnoses of PHC may even lead to irreparable damages. Therefore, accurate diagnosis is necessary to prevent severe complications.