Uncertainty quantification in multi-class image classification using chest X-ray images of COVID-19 and pneumonia.

This paper investigates uncertainty quantification (UQ) techniques in multi-class classification of chest X-ray images (COVID-19, Pneumonia, and Normal). We evaluate Bayesian Neural Networks (BNN) and the Deep Neural Network with UQ (DNN with UQ) techniques, including Monte Carlo dropout, Ensemble Bayesian Neural Network (EBNN), Ensemble Monte Carlo (EMC) dropout, across different evaluation metrics. Our analysis reveals that DNN with UQ, especially EBNN and EMC dropout, consistently outperform BNNs. For example, in Class 0 vs. All, EBNN achieved a UAcc of 92.6%, UAUC-ROC of 95.0%, and a Brier Score of 0.157, significantly surpassing BNN's performance. Similarly, EMC Dropout excelled in Class 1 vs. All with a UAcc of 83.5%, UAUC-ROC of 95.8%, and a Brier Score of 0.165. These advanced models demonstrated higher accuracy, better discriaminative capability, and more accurate probabilistic predictions. Our findings highlight the efficacy of DNN with UQ in enhancing model reliability and interpretability, making them highly suitable for critical healthcare applications like chest X-ray imageQ6 classification.

Accuracy of an artificial intelligence-enabled diagnostic assistance device in recognizing normal chest radiographs: a service evaluation.

Objectives: Artificial intelligence (AI) enabled devices may be able to optimize radiologists' productivity by identifying normal and abnormal chest X-rays (CXRs) for triaging. In this service evaluation, we investigated the accuracy of one such AI device (qXR). Methods: A randomly sampled subset of general practice and outpatient-referred frontal CXRs from a National Health Service Trust was collected retrospectively from examinations conducted during November 2022 to January 2023. Ground truth was established by consensus between 2 radiologists. The main objective was to estimate negative predictive value (NPV) of AI. Results: A total of 522 CXRs (458 [87.74%] normal CXRs) from 522 patients (median age, 64 years [IQR, 49-77]; 305 [58.43%] female) were analysed. AI predicted 348 CXRs as normal, of which 346 were truly normal (NPV: 99.43% [95% CI, 97.94-99.93]). The sensitivity, specificity, positive predictive value, and area under the ROC curve of AI were found to be 96.88% (95% CI, 89.16-99.62), 75.55% (95% CI, 71.34-79.42), 35.63% (95% CI, 28.53-43.23), and 91.92% (95% CI, 89.38-94.45), respectively. A sensitivity analysis was conducted to estimate NPV by varying assumptions of the prevalence of normal CXRs. The NPV ranged from 88.96% to 99.54% as prevalence increased. Conclusions: The AI device recognized normal CXRs with high NPV and has the potential to increase radiologists' productivity. Advances in knowledge: There is a need for more evidence on the utility of AI-enabled devices in identifying normal CXRs. This work adds to such limited evidence and enables researchers to plan studies to further evaluate the impact of such devices.

Utility of bone suppression imaging for the detection of pneumonia on chest radiographs.

INTRODUCTION: Chest X-rays (CXR) are routinely used to diagnose lung and heart conditions. AI based Bone suppression imaging (BSI) aims to enhance accuracy in identifying chest anomalies by eliminating bony structures such as the ribs, clavicles, and scapula from CXRs. The aim of this retrospective study was to assess the clinical value of BSI in detecting pneumonia. METHODS: Ninety-nine emergency patients with suspected pneumonia underwent erect postero-anterior CXRs. The BSI processing system was used to generate corresponding bone-suppressed images for the 99 radiographs. Each patient had undergone a computed tomography (CT) examination within 48 h, considered the standard of reference. Two blinded readers separately analyzed images, indicating confidence levels regarding signs of pneumonia for each lung separated in three fields, first with standard images, then with BSI. Sensitivity, specificity, predictive values, and readers' certitude were calculated, and inter-reader agreement was evaluated with the kappa statistic. RESULTS: Out of the 99 included cases, 39 cases of pneumonia were diagnosed (39.4%). Of the remaining 60 patients, 14 presented only pleural effusions (14.1%). BSI images led to a significant increase in false positives (+251%) and significantly affected one reader's diagnosis and certitude, decreasing accuracy (up to 17%) and specificity (up to 14%). Sensitivity increased by 66% with BSI. Inter-reader agreement ranged from weak to moderate (0.113-0.53) and did not improve with BSI. For both readers, BSI images were read with significantly lesser certitude than standard images. CONCLUSION: BSI did not add clinical value in pneumonia detection on CXR due to a significant increase in false positive results and a decrease one readers' certitude. IMPLICATION FOR PRACTICE: The study emphasizes the importance of proper clinical training before implementing new post-processing and artificial intelligence (AI) tools in clinical practice.

Chest X-Ray Comparison Between Drug-Resistant and Drug-Sensitive Pulmonary Tuberculosis in Children.

INTRODUCTION: Chest X-ray (CXR) remains one of the tools used in diagnosing tuberculosis (TB). However, few studies about such tools exist, specifically in children in Indonesia. We aim to investigate and compare the CXR findings of children with pulmonary drug-resistant TB (DR-TB) and drug-sensitive TB (DS-TB) that could help in the evaluation and management of TB cases in children. METHODS: Retrospective analysis with cross-sectional approach was conducted in children (<18 years old) diagnosed with pulmonary DR-TB and DS-TB from January 2018 to December 2021. Documented data were collected from the Paediatric Respirology Registry and Tuberculosis Information System at Dr. Hasan Sadikin General Hospital Bandung. Characteristics of children, CXR findings, and TB severity were assessed and compared using the chi-square and Fisher's exact tests with significance levels set at p value <0.05. RESULTS: Sixty-nine children (DR-TB 31 children vs. DS-TB 38 children) were assessed. Of the 31 children with DR-TB, 65% were classified as multidrug-resistant TB (MDR-TB), followed by rifampicin-resistant TB (RR-TB), pre-extensively drug-resistant TB (pre-XDR-TB), and extensively drug-resistant TB (XDR-TB). The most common CXR findings in DR-TB are consolidation (68%), fibrosis (42%), and cavity (29%), whereas in DS-TB, it is pleura effusion (37%). Severe TB accounts for 50% of DR-TB (p = 0.008). CONCLUSIONS: Consolidation, fibrosis, cavities, and findings of severe TB are most common in DR-TB. Pleural effusion is the most common in DS-TB. These findings have the potential to be considered in further examination of children with pulmonary DR-TB and DS-TB; hence, more extensive studies are needed to confirm these results.

Explainable artificial intelligence in deep learning-based detection of aortic elongation on chest X-ray images.

Aims: Aortic elongation can result from age-related changes, congenital factors, aneurysms, or conditions affecting blood vessel elasticity. It is associated with cardiovascular diseases and severe complications like aortic aneurysms and dissection. We assess qualitatively and quantitatively explainable methods to understand the decisions of a deep learning model for detecting aortic elongation using chest X-ray (CXR) images. Methods and results: In this study, we evaluated the performance of deep learning models (DenseNet and EfficientNet) for detecting aortic elongation using transfer learning and fine-tuning techniques with CXR images as input. EfficientNet achieved higher accuracy (86.7% ± 2.1), precision (82.7% ± 2.7), specificity (89.4% ± 1.7), F1 score (82.5% ± 2.9), and area under the receiver operating characteristic (92.7% ± 0.6) but lower sensitivity (82.3% ± 3.2) compared with DenseNet. To gain insights into the decision-making process of these models, we employed gradient-weighted class activation mapping and local interpretable model-agnostic explanations explainability methods, which enabled us to identify the expected location of aortic elongation in CXR images. Additionally, we used the pixel-flipping method to quantitatively assess the model interpretations, providing valuable insights into model behaviour. Conclusion: Our study presents a comprehensive strategy for analysing CXR images by integrating aortic elongation detection models with explainable artificial intelligence techniques. By enhancing the interpretability and understanding of the models' decisions, this approach holds promise for aiding clinicians in timely and accurate diagnosis, potentially improving patient outcomes in clinical practice.

Impact of Long-Term Non-Communicable Diseases on SARS-COV-2 Hospitalized Patients Supported by Radiological Imaging in Southern Pakistan.

COVID-19 patients with already existing chronic medical conditions are more likely to develop severe complications and, ultimately, a higher risk of mortality. This study analyzes the impacts of pre-existing chronic illnesses such as diabetes (DM), hypertension, and cardiovascular diseases (CVDs) on COVID-19 cases by using radiological chest imaging. The data of laboratory-confirmed COVID-19-infected hospitalized patients were analyzed from March 2020 to December 2020. Chest X-ray images were included to further identify the differences in X-ray patterns of patients with co-morbid conditions and without any co-morbidity. The Pearson chi-square test checks the significance of the association between co-morbidities and mortality. The magnitude and dimension of the association were calibrated by the odds ratio (OR) at a 95% confidence interval (95% CI) over the patients' status (mortality and discharged cases). A univariate binary logistic regression model was applied to examine the impact of co-morbidities on death cases independently. A multivariate binary logistic regression model was applied for the adjusted effects of possible confounders. For the sensitivity analysis of the model, receiver operating characteristic (ROC) was applied. Patients with different comorbidities, including diabetes (OR = 33.4, 95% CI: 20.31-54.78, p < 0.001), cardiovascular conditions (OR = 24.14, 95% CI: 10.18-57.73, p < 0.001), and hypertension (OR = 16.9, 95% CI: 10.20-27.33, p < 0.001), showed strong and significant associations. The opacities present in various zones of the lungs clearly show that COVID-19 patients with chronic illnesses such as diabetes, hypertension, cardiovascular disease, and obesity experience significantly worse outcomes, as evidenced by chest X-rays showing increased pneumonia and deterioration. Therefore, stringent precautions and a global public health campaign are crucial to reducing mortality in these high-risk groups.

Deep learning to predict long-term mortality from plain chest X-ray in patients referred for suspected coronary artery disease.

Background: The hypothesis that a deep learning (DL) model can produce long-term prognostic information from chest X-ray (CXR) has already been confirmed within cancer screening programs. We summarize our experience with DL prediction of long-term mortality, from plain CXR, in patients referred for angina and coronary angiography. Methods: Data of patients referred to an Italian academic hospital were analyzed retrospectively. We designed a deep convolutional neural network (DCNN) that, from CXR, could predict long-term mortality. External validation was performed on patients referred to a Dutch academic hospital. Results: A total of 6,031 were used for model training (71%; n=4,259) and fine-tuning/validation (10%; n=602). Internal validation was performed with the remaining patients (19%; n=1,170). Patients' stratification followed the DL-CXR risk score quartiles division. Median follow-up was 6.1 years [interquartile range (IQR), 3.3-8.7 years]. We observed an increment in estimated mortality with the increase of DL-CXR risk score (low-risk 5%, moderate 17%, high 29%, very high 46%; P<0.001). The DL-CXR risk score predicted median follow-up outcome with an area under the curve (AUC) of 0.793 [95% confidence interval (CI): 0.759-0.827, sensitivity 78%, specificity 68%]. Prediction was better than that achieved using coronary angiography findings (AUC: 0.569, 95% CI: 0.52-0.61, P<0.001) and age (AUC: 0.735, 95% CI: 0.69-0.77, P<0.004). At Cox regression, the DL-CXR risk score predicted follow-up mortality (P<0.005, hazard ratio: 3.30, 95% CI: 2.35-4.64). External validation confirmed the DL-CXR risk score performance (AUC: 0.71, 95% CI: 0.49-0.92; sensitivity 0.838; specificity 0.338). Conclusions: In patients referred for coronary angiogram because of angina, the DL-CXR risk score could be used to stratify mortality risk and predict long-term outcome better than age and coronary artery disease status.

A Skinfold Imitating a Pneumothorax.

We present the case of a male patient in his late 80s who presented with a fall with symptoms and signs of community-acquired pneumonia. Chest X-ray showed the suspicion of a left-sided pneumothorax. A CT of the chest subsequently ruled out the presence of a pneumothorax on the left side. The pseudo-pneumothorax on the chest X-ray was secondary to a skinfold. This case highlights how well a skinfold can mimic pneumothorax. Careful clinical and radiological examination with bedside lung ultrasound and/or CT of the chest can help differentiate true pneumothorax from pseudo-pneumothorax, provided the patient is hemodynamically stable. Our case highlights the importance of clinical examination, various imaging modalities, and confirmation of a diagnosis before proceeding to interventional procedures in the context of limited clinical suspicion of the differential.

Timing of ultra-portable ultrasound (UPUS) Examinations in detecting clinically concerning recurrent pneumothorax.

BACKGROUND: Recurrent pneumothorax (rPTX) is a common complication following thoracostomy tube (TT) removal in chest trauma patients. While chest X-ray (CXR) is most commonly used to detect a rPTX, bedside ultraportable ultrasound (UPUS) is a feasible, low cost, and radiation free alternative. No consensus exists with regards to the optimal timing of diagnostic imaging to assess for rPTX post-TT removal. Accordingly, we sought to identify an ideal UPUS timing to detect a rPTX METHODS: We conducted a single center prospective study of adult (≥18years) patients admitted with a chest trauma. UPUS examinations were performed using the Butterfly iQ+™ ultrasound. Three intercostal spaces (ICS) were evaluated (2nd through 4th). Post-TT UPUS examinations were performed at different timepoints following tube removal (1-6 h). A rPTX on UPUS was defined as the absence of lung-sliding in one or more intercostal spaces, and was considered a clinically concerning rPTX if lung-sliding was absent in ≥2 ICS. UPUS findings were compared to CXR. RESULTS: Ninety-two patients (97 hemi-thoraces) were included in the analysis. A total of 58 patients had a post-TT removal rPTX of which 11 were either clinically concerning or expanding. Comparing UPUS findings to CXR, the 3-hour post-TT removal ultrasound examinations were associated with the highest sensitivity. By hour 4, no rPTX showed expansion in size. Three patients required an intervention for a clinically concerning rPTX, all of whom were detected on UPUS 3-hour post-TT removal. CONCLUSION: Bedside UPUS performed at 3-hour post-TT removal has the highest sensitivity in detecting clinically concerning rPTX. Size of rPTX appears to stabilize by hour 4. In the absence of clinical symptoms, repeat imaging or observation of non-significant rPTX beyond 4 h may not provide added clinical benefit. LEVEL OF EVIDENCE: Level II, Diagnostic Tests or Criteria.

COVID-19 severity detection using chest X-ray segmentation and deep learning.

COVID-19 has resulted in a significant global impact on health, the economy, education, and daily life. The disease can range from mild to severe, with individuals over 65 or those with underlying medical conditions being more susceptible to severe illness. Early testing and isolation are vital due to the virus's variable incubation period. Chest radiographs (CXR) have gained importance as a diagnostic tool due to their efficiency and reduced radiation exposure compared to CT scans. However, the sensitivity of CXR in detecting COVID-19 may be lower. This paper introduces a deep learning framework for accurate COVID-19 classification and severity prediction using CXR images. U-Net is used for lung segmentation, achieving a precision of 0.9924. Classification is performed using a Convulation-capsule network, with high true positive rates of 86% for COVID-19, 93% for pneumonia, and 85% for normal cases. Severity assessment employs ResNet50, VGG-16, and DenseNet201, with DenseNet201 showing superior accuracy. Empirical results, validated with 95% confidence intervals, confirm the framework's reliability and robustness. This integration of advanced deep learning techniques with radiological imaging enhances early detection and severity assessment, improving patient management and resource allocation in clinical settings.

Generalizable diagnosis of chest radiographs through attention-guided decomposition of images utilizing self-consistency loss.

BACKGROUND: Chest X-ray (CXR) is one of the most commonly performed imaging tests worldwide. Due to its wide usage, there is a growing need for automated and generalizable methods to accurately diagnose these images. Traditional methods for chest X-ray analysis often struggle with generalization across diverse datasets due to variations in imaging protocols, patient demographics, and the presence of overlapping anatomical structures. Therefore, there is a significant demand for advanced diagnostic tools that can consistently identify abnormalities across different patient populations and imaging settings. We propose a method that can provide a generalizable diagnosis of chest X-ray. METHOD: Our method utilizes an attention-guided decomposer network (ADSC) to extract disease maps from chest X-ray images. The ADSC employs one encoder and multiple decoders, incorporating a novel self-consistency loss to ensure consistent functionality across its modules. The attention-guided encoder captures salient features of abnormalities, while three distinct decoders generate a normal synthesized image, a disease map, and a reconstructed input image, respectively. A discriminator differentiates the real and the synthesized normal chest X-rays, enhancing the quality of generated images. The disease map along with the original chest X-ray image are fed to a DenseNet-121 classifier modified for multi-class classification of the input X-ray. RESULTS: Experimental results on multiple publicly available datasets demonstrate the effectiveness of our approach. For multi-class classification, we achieve up to a 3% improvement in AUROC score for certain abnormalities compared to the existing methods. For binary classification (normal versus abnormal), our method surpasses existing approaches across various datasets. In terms of generalizability, we train our model on one dataset and tested it on multiple datasets. The standard deviation of AUROC scores for different test datasets is calculated to measure the variability of performance across datasets. Our model exhibits superior generalization across datasets from diverse sources. CONCLUSIONS: Our model shows promising results for the generalizable diagnosis of chest X-rays. The impacts of using the attention mechanism and the self-consistency loss in our method are evident from the results. In the future, we plan to incorporate Explainable AI techniques to provide explanations for model decisions. Additionally, we aim to design data augmentation techniques to reduce class imbalance in our model.

Three-Stage Framework for Accurate Pediatric Chest X-ray Diagnosis Using Self-Supervision and Transfer Learning on Small Datasets.

Pediatric respiratory disease diagnosis and subsequent treatment require accurate and interpretable analysis. A chest X-ray is the most cost-effective and rapid method for identifying and monitoring various thoracic diseases in children. Recent developments in self-supervised and transfer learning have shown their potential in medical imaging, including chest X-ray areas. In this article, we propose a three-stage framework with knowledge transfer from adult chest X-rays to aid the diagnosis and interpretation of pediatric thorax diseases. We conducted comprehensive experiments with different pre-training and fine-tuning strategies to develop transformer or convolutional neural network models and then evaluate them qualitatively and quantitatively. The ViT-Base/16 model, fine-tuned with the CheXpert dataset, a large chest X-ray dataset, emerged as the most effective, achieving a mean AUC of 0.761 (95% CI: 0.759-0.763) across six disease categories and demonstrating a high sensitivity (average 0.639) and specificity (average 0.683), which are indicative of its strong discriminative ability. The baseline models, ViT-Small/16 and ViT-Base/16, when directly trained on the Pediatric CXR dataset, only achieved mean AUC scores of 0.646 (95% CI: 0.641-0.651) and 0.654 (95% CI: 0.648-0.660), respectively. Qualitatively, our model excels in localizing diseased regions, outperforming models pre-trained on ImageNet and other fine-tuning approaches, thus providing superior explanations. The source code is available online and the data can be obtained from PhysioNet.

Increased lung field permeability in young patients.

When the chest radiograph of a young patient shows lung hyperlucency, it is important to obtain a detailed clinical history of any previous episodes of childhood infection. Previous chest radiographs should be reviewed to determine whether the condition is congenital or acquired, and thus assist in a diagnosis of SJMS.

Radiological Features of Extramedullary Hematopoiesis in a Young Male With Beta-Thalassemia: A Case Report.

The formation of the blood elements and their maturation is called hematopoiesis. In adults, this typically takes place in the bone marrow of vertebrae, ribs, and long bones. In contrast, during fetal development, the primary sites of hematopoiesis are the spleen, liver, and the yolk sac. This process of hematopoiesis, when it occurs in sites other than the bone marrow, is called the extramedullary hematopoiesis (EMH). Extramedullary hematopoiesis usually happens in patients with blood disorders like sickle cell disease and thalassemia, where there is failure of hematopoiesis in the primary sites. Here, we present a young male with beta-thalassemia who presented with shortness of breath and palpitations for one month. This manuscript discusses the imaging findings of the EMH in our patient.

A robust ensemble deep learning framework for accurate diagnoses of tuberculosis from chest radiographs.

Introduction: Tuberculosis (TB) stands as a paramount global health concern, contributing significantly to worldwide mortality rates. Effective containment of TB requires deployment of cost-efficient screening method with limited resources. To enhance the precision of resource allocation in the global fight against TB, this research proposed chest X-ray radiography (CXR) based machine learning screening algorithms with optimization, benchmarking and tuning for the best TB subclassification tasks for clinical application. Methods: This investigation delves into the development and evaluation of a robust ensemble deep learning framework, comprising 43 distinct models, tailored for the identification of active TB cases and the categorization of their clinical subtypes. The proposed framework is essentially an ensemble model with multiple feature extractors and one of three fusion strategies-voting, attention-based, or concatenation methods-in the fusion stage before a final classification. The comprised de-identified dataset contains records of 915 active TB patients alongside 1,276 healthy controls with subtype-specific information. Thus, the realizations of our framework are capable for diagnosis with subclass identification. The subclass tags include: secondary tuberculosis/tuberculous pleurisy; non-cavity/cavity; secondary tuberculosis only/secondary tuberculosis and tuberculous pleurisy; tuberculous pleurisy only/secondary tuberculosis and tuberculous pleurisy. Results: Based on the dataset and model selection and tuning, ensemble models show their capability with self-correction capability of subclass identification with rendering robust clinical predictions. The best double-CNN-extractor model with concatenation/attention fusion strategies may potentially be the successful model for subclass tasks in real application. With visualization techniques, in-depth analysis of the ensemble model's performance across different fusion strategies are verified. Discussion: The findings underscore the potential of such ensemble approaches in augmenting TB diagnostics with subclassification. Even with limited dataset, the self-correction within the ensemble models still guarantees the accuracies to some level for potential clinical decision-making processes in TB management. Ultimately, this study shows a direction for better TB screening in the future TB response strategy.

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

INTRODUCTION: In 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 METHODS: SRs 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/. RESULTS: After 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. CONCLUSIONS: The 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.

Automatic cardiothoracic ratio calculation based on lung fields abstracted from chest X-ray images without heart segmentation.

Introduction: The cardiothoracic ratio (CTR) based on postero-anterior chest X-rays (P-A CXR) images is one of the most commonly used cardiac measurement methods and an indicator for initially evaluating cardiac diseases. However, the hearts are not readily observable on P-A CXR images compared to the lung fields. Therefore, radiologists often manually determine the CTR's right and left heart border points of the adjacent left and right lung fields to the heart based on P-A CXR images. Meanwhile, manual CTR measurement based on the P-A CXR image requires experienced radiologists and is time-consuming and laborious. Methods: Based on the above, this article proposes a novel, fully automatic CTR calculation method based on lung fields abstracted from the P-A CXR images using convolutional neural networks (CNNs), overcoming the limitations to heart segmentation and avoiding errors in heart segmentation. First, the lung field mask images are abstracted from the P-A CXR images based on the pre-trained CNNs. Second, a novel localization method of the heart's right and left border points is proposed based on the two-dimensional projection morphology of the lung field mask images using graphics. Results: The results show that the mean distance errors at the x-axis direction of the CTR's four key points in the test sets T1 (21 × 512 × 512 static P-A CXR images) and T2 (13 × 512 × 512 dynamic P-A CXR images) based on various pre-trained CNNs are 4.1161 and 3.2116 pixels, respectively. In addition, the mean CTR errors on the test sets T1 and T2 based on four proposed models are 0.0208 and 0.0180, respectively. Discussion: Our proposed model achieves the equivalent performance of CTR calculation as the previous CardioNet model, overcomes heart segmentation, and takes less time. Therefore, our proposed method is practical and feasible and may become an effective tool for initially evaluating cardiac diseases.

M-VAAL: Multimodal Variational Adversarial Active Learning for Downstream Medical Image Analysis Tasks.

Acquiring properly annotated data is expensive in the medical field as it requires experts, time-consuming protocols, and rigorous validation. Active learning attempts to minimize the need for large annotated samples by actively sampling the most informative examples for annotation. These examples contribute significantly to improving the performance of supervised machine learning models, and thus, active learning can play an essential role in selecting the most appropriate information in deep learning-based diagnosis, clinical assessments, and treatment planning. Although some existing works have proposed methods for sampling the best examples for annotation in medical image analysis, they are not task-agnostic and do not use multimodal auxiliary information in the sampler, which has the potential to increase robustness. Therefore, in this work, we propose a Multimodal Variational Adversarial Active Learning (M-VAAL) method that uses auxiliary information from additional modalities to enhance the active sampling. We applied our method to two datasets: i) brain tumor segmentation and multi-label classification using the BraTS2018 dataset, and ii) chest X-ray image classification using the COVID-QU-Ex dataset. Our results show a promising direction toward data-efficient learning under limited annotations.

A Deep-Learning-Enabled Electrocardiogram and Chest X-Ray for Detecting Pulmonary Arterial Hypertension.

The diagnosis and treatment of pulmonary hypertension have changed dramatically through the re-defined diagnostic criteria and advanced drug development in the past decade. The application of Artificial Intelligence for the detection of elevated pulmonary arterial pressure (ePAP) was reported recently. Artificial Intelligence (AI) has demonstrated the capability to identify ePAP and its association with hospitalization due to heart failure when analyzing chest X-rays (CXR). An AI model based on electrocardiograms (ECG) has shown promise in not only detecting ePAP but also in predicting future risks related to cardiovascular mortality. We aimed to develop an AI model integrating ECG and CXR to detect ePAP and evaluate their performance. We developed a deep-learning model (DLM) using paired ECG and CXR to detect ePAP (systolic pulmonary artery pressure > 50 mmHg in transthoracic echocardiography). This model was further validated in a community hospital. Additionally, our DLM was evaluated for its ability to predict future occurrences of left ventricular dysfunction (LVD, ejection fraction < 35%) and cardiovascular mortality. The AUCs for detecting ePAP were as follows: 0.8261 with ECG (sensitivity 76.6%, specificity 74.5%), 0.8525 with CXR (sensitivity 82.8%, specificity 72.7%), and 0.8644 with a combination of both (sensitivity 78.6%, specificity 79.2%) in the internal dataset. In the external validation dataset, the AUCs for ePAP detection were 0.8348 with ECG, 0.8605 with CXR, and 0.8734 with the combination. Furthermore, using the combination of ECGs and CXR, the negative predictive value (NPV) was 98% in the internal dataset and 98.1% in the external dataset. Patients with ePAP detected by the DLM using combination had a higher risk of new-onset LVD with a hazard ratio (HR) of 4.51 (95% CI: 3.54-5.76) in the internal dataset and cardiovascular mortality with a HR of 6.08 (95% CI: 4.66-7.95). Similar results were seen in the external validation dataset. The DLM, integrating ECG and CXR, effectively detected ePAP with a strong NPV and forecasted future risks of developing LVD and cardiovascular mortality. This model has the potential to expedite the early identification of pulmonary hypertension in patients, prompting further evaluation through echocardiography and, when necessary, right heart catheterization (RHC), potentially resulting in enhanced cardiovascular outcomes.