Confidence-Aware Severity Assessment of Lung Disease from Chest X-Rays Using Deep Neural Network on a Multi-Reader Dataset.

In this study, we present a method based on Monte Carlo Dropout (MCD) as Bayesian neural network (BNN) approximation for confidence-aware severity classification of lung diseases in COVID-19 patients using chest X-rays (CXRs). Trained and tested on 1208 CXRs from Hospital 1 in the USA, the model categorizes severity into four levels (i.e., normal, mild, moderate, and severe) based on lung consolidation and opacity. Severity labels, determined by the median consensus of five radiologists, serve as the reference standard. The model's performance is internally validated against evaluations from an additional radiologist and two residents that were excluded from the median. The performance of the model is further evaluated on additional internal and external datasets comprising 2200 CXRs from the same hospital and 1300 CXRs from Hospital 2 in South Korea. The model achieves an average area under the curve (AUC) of 0.94 ± 0.01 across all classes in the primary dataset, surpassing human readers in each severity class and achieves a higher Kendall correlation coefficient (KCC) of 0.80 ± 0.03. The performance of the model is consistent across varied datasets, highlighting its generalization. A key aspect of the model is its predictive uncertainty (PU), which is inversely related to the level of agreement among radiologists, particularly in mild and moderate cases. The study concludes that the model outperforms human readers in severity assessment and maintains consistent accuracy across diverse datasets. Its ability to provide confidence measures in predictions is pivotal for potential clinical use, underscoring the BNN's role in enhancing diagnostic precision in lung disease analysis through CXR.

Clinical Validation of a Deep Learning Algorithm for Automated Coronary Artery Disease Detection and Classification Using a Heterogeneous Multivendor Coronary Computed Tomography Angiography Data Set.

PURPOSE: We sought to clinically validate a fully automated deep learning (DL) algorithm for coronary artery disease (CAD) detection and classification in a heterogeneous multivendor cardiac computed tomography angiography data set. MATERIALS AND METHODS: In this single-centre retrospective study, we included patients who underwent cardiac computed tomography angiography scans between 2010 and 2020 with scanners from 4 vendors (Siemens Healthineers, Philips, General Electrics, and Canon). Coronary Artery Disease-Reporting and Data System (CAD-RADS) classification was performed by a DL algorithm and by an expert reader (reader 1, R1), the gold standard. Variability analysis was performed with a second reader (reader 2, R2) and the radiologic reports on a subset of cases. Statistical analysis was performed stratifying patients according to the presence of CAD (CAD-RADS >0) and obstructive CAD (CAD-RADS ≥3). RESULTS: Two hundred ninety-six patients (average age: 53.66 ± 13.65, 169 males) were enrolled. For the detection of CAD only, the DL algorithm showed sensitivity, specificity, accuracy, and area under the curve of 95.3%, 79.7%, 87.5%, and 87.5%, respectively. For the detection of obstructive CAD, the DL algorithm showed sensitivity, specificity, accuracy, and area under the curve of 89.4%, 92.8%, 92.2%, and 91.1%, respectively. The variability analysis for the detection of obstructive CAD showed an accuracy of 92.5% comparing the DL algorithm with R1, and 96.2% comparing R1 with R2 and radiology reports. The time of analysis was lower using the DL algorithm compared with R1 (P < 0.001). CONCLUSIONS: The DL algorithm demonstrated robust performance and excellent agreement with the expert readers' analysis for the evaluation of CAD, which also corresponded with significantly reduced image analysis time.

Tumorous tissue characterization using integrated 18F-FDG PET/dual-energy CT in lung cancer: Combining iodine enhancement and glycolytic activity.

Positron emission tomography/computed tomography (PET/CT) with 18F-fluorodeoxyglucose (18F-FDG) has become the method of choice for tumor staging in lung cancer patients with improved diagnostic accuracy for the evaluation of lymph node involvement and distant metastasis. Due to its spectral capabilities, dual-energy CT (DECT) employs a material decomposition algorithm enabling precise quantification of iodine concentrations in distinct tissues. This technique enhances the characterization of tumor blood supply and has demonstrated promising results for the assessment of therapy response in patients with lung cancer. Several studies have demonstrated that DECT provides additional value to the PET-based evaluation of glycolytic activity, especially for the evaluation of therapy response and follow-up of patients with lung cancer. The combination of PET and DECT in a single scanner system enables the simultaneous assessment of glycolytic activity and iodine enhancement, offering further insight to the characterization of tumorous tissues. Recently a new approach of a novel integrated PET/DECT was investigated in a pilot study on patients with non-small cell lung cancer (NSCLC). The study showed a moderate correlation between PET-based standard uptake values (SUV) and DECT-based iodine densities in the evaluation of lung tumorous tissue but with limited assessment of lymph nodes. The following review on tumorous tissue characterization using PET and DECT imaging describes the strengths and limitations of this novel technique.

Chest radiography findings of COVID-19 pneumonia: a specific pattern for a confident differential diagnosis.

BACKGROUND: Chest radiography (CR) patterns for the diagnosis of COVID-19 have been established. However, they were not ideated comparing CR features with those of other pulmonary diseases. PURPOSE: To create the most accurate COVID-19 pneumonia pattern comparing CR findings of COVID-19 and non-COVID-19 pulmonary diseases and to test the model against the British Society of Thoracic Imaging (BSTI) criteria. MATERIAL AND METHODS: CR of COVID-19 and non-COVID-19 pulmonary diseases, admitted to the emergency department, were evaluated. Assessed features were interstitial opacities, ground glass opacities, and/or consolidations and the predominant lung alteration. We also assessed uni-/bilaterality, location (upper/middle/lower), and distribution (peripheral/perihilar), as well as pleural effusion and perihilar vessels blurring. A binary logistic regression was adopted to obtain the most accurate CR COVID-19 pattern, and sensitivity and specificity were computed. The newly defined pattern was compared to BSTI criteria. RESULTS: CR of 274 patients were evaluated (146 COVID-19, 128 non-COVID-19). The most accurate COVID-19 pneumonia pattern consisted of four features: bilateral alterations (Expß=2.8, P=0.002), peripheral distribution of the predominant (Expß=2.3, P=0.013), no pleural effusion (Expß=0.4, P=0.009), and perihilar vessels' contour not blurred (Expß=0.3, P=0.002). The pattern showed 49% sensitivity, 81% specificity, and 64% accuracy, while BSTI criteria showed 51%, 77%, and 63%, respectively. CONCLUSION: Bilaterality, peripheral distribution of the predominant lung alteration, no pleural effusion, and perihilar vessels contour not blurred determine the most accurate COVID-19 pneumonia pattern. Lower field involvement, proposed by BSTI criteria, was not a distinctive finding. The BSTI criteria has lower specificity.

Comparison of Triple-Rule-Out Prospectively ECG-triggered Systolic and Diastolic Acquisition Protocol in Patients With Acute Chest Pain.

PURPOSE: The purpose of this study was to compare image quality and coronary interpretability of triple-rule-out systolic and diastolic protocols in patients with acute chest pain. MATERIALS AND METHODS: From March 2016 to October 2017 the authors prospectively enrolled patients with undifferentiated acute chest pain, who were at low to intermediate cardiovascular risk. Those with heart rate >75 bpm underwent a systolic prospectively triggered acquisition (systolic triggering [ST]), and in those with ≤75 bpm, end-diastolic triggering (DT) was instead performed. Examinations were evaluated for coronary artery disease, aortic dissection, and pulmonary embolism. Image quality was assessed using a Likert scale. Coronary arteries interpretability was evaluated both on a per-vessel and a per segment basis. The occurrence of major adverse cardiovascular events was investigated. RESULTS: The final study population was 189 patients. Fifty-two patients (27.5%) underwent systolic acquisition and 137 (72.5%) underwent diastolic acquisition. No significant differences in overall image quality were observed between DT and ST groups (median score 5 [interquartile ranges 4 to 5] vs. 4 [interquartile ranges 4 to 5], P =0.074). Although both DT and ST protocols showed low percentages of noninterpretable coronary arteries on a per-vessel (1.5% and 6.7%, respectively) and per-segment analysis (1% and 4.7%, respectively), these percentages resulted significantly higher for ST groups ( P <0.001). Obstructive coronary stenosis was observed in 18 patients. Only one case of pulmonary embolism was diagnosed and no cases of aortic dissection were found in our population. No death or major adverse cardiovascular events were observed during follow-up among the 2 groups. CONCLUSIONS: Results showed that triple-rule-out computed tomography angiography is a reliable technique in patients with acute chest pain and that an ST acquisition protocol could be considered an alternative acquisition protocol in patients with higher heart rate, reaching a good image quality.

Prevention of Coronary Artery Disease-Related Heart Failure: The Role of Computed Tomography Scan.

During the past decade, coronary computed tomography angiography has emerged as the primary modality to noninvasively detect and rule out coronary artery disease. Therefore, this technique could play an important role in identifying patients at high risk of heart failure, considering the high prevalence of coronary artery disease in these patients. The latest technologies have also increased diagnostic accuracy, helping to close the gap with the other functional imaging modalities.