Automatic Quantification of Abnormal Lung Parenchymal Attenuation on Chest Computed Tomography Images Using Densitometry and Texture-based Analysis.

PURPOSE: To compare texture-based analysis using convolutional neural networks (CNNs) against lung densitometry in detecting chest computed tomography (CT) image abnormalities. MATERIAL AND METHODS: A U-NET was used for lung segmentation, and an ensemble of 7 CNN architectures was trained for the classification of low-attenuation areas (LAAs; emphysema, cysts), normal-attenuation areas (NAAs; normal parenchyma), and high-attenuation areas (HAAs; ground-glass opacities, crazy paving/linear opacity, consolidation). Lung densitometry also computes (LAAs, ≤-950 HU), NAAs (-949 to -700 HU), and HAAs (-699 to -250 HU). CNN-based and densitometry-based severity indices (CNN and Dens, respectively) were calculated as (LAA+HAA)/(LAA+NAA+HAA) in 812 CT scans from 176 normal subjects, 343 patients with emphysema, and 293 patients with interstitial lung disease (ILD). The correlation between CNN-derived and densitometry-derived indices was analyzed, alongside a comparison of severity indices among patient subgroups with emphysema and ILD, using the Spearman correlation and ANOVA with Bonferroni correction. RESULTS: CNN-derived and densitometry-derived severity indices (SIs) showed a strong correlation (ρ=0.90) and increased with disease severity. CNN-SIs differed from densitometry SIs, being lower for emphysema and higher for moderate to severe ILD cases. CNN estimations for normal attenuation areas were higher than those from densitometry across all groups, indicating a potential for more accurate characterization of lung abnormalities. CONCLUSIONS: CNN outputs align closely with densitometry in assessing lung abnormalities on CT scans, offering improved estimates of normal areas and better distinguishing similar abnormalities. However, this requires higher computing power.

Automated Segmentation of Lymph Nodes on Neck CT Scans Using Deep Learning.

Early and accurate detection of cervical lymph nodes is essential for the optimal management and staging of patients with head and neck malignancies. Pilot studies have demonstrated the potential for radiomic and artificial intelligence (AI) approaches in increasing diagnostic accuracy for the detection and classification of lymph nodes, but implementation of many of these approaches in real-world clinical settings would necessitate an automated lymph node segmentation pipeline as a first step. In this study, we aim to develop a non-invasive deep learning (DL) algorithm for detecting and automatically segmenting cervical lymph nodes in 25,119 CT slices from 221 normal neck contrast-enhanced CT scans from patients without head and neck cancer. We focused on the most challenging task of segmentation of small lymph nodes, evaluated multiple architectures, and employed U-Net and our adapted spatial context network to detect and segment small lymph nodes measuring 5-10 mm. The developed algorithm achieved a Dice score of 0.8084, indicating its effectiveness in detecting and segmenting cervical lymph nodes despite their small size. A segmentation framework successful in this task could represent an essential initial block for future algorithms aiming to evaluate small objects such as lymph nodes in different body parts, including small lymph nodes looking normal to the naked human eye but harboring early nodal metastases.

Computed tomography versus ultrasound for the diagnosis of acute cholecystitis: a systematic review and meta-analysis.

OBJECTIVES: Some patients undergo both computed tomography (CT) and ultrasound (US) sequentially as part of the same evaluation for acute cholecystitis (AC). Our goal was to perform a systematic review and meta-analysis comparing the diagnostic performance of US and CT in the diagnosis of AC. MATERIALS AND METHODS: Databases were searched for relevant published studies through November 2023. The primary objective was to compare the head-to-head performance of US and CT using surgical intervention or clinical follow-up as the reference standard. For the secondary analysis, all individual US and CT studies were analyzed. The pooled sensitivities, specificities, and areas under the curve (AUCs) were determined along with 95% confidence intervals (CIs). The prevalence of imaging findings was also evaluated. RESULTS: Sixty-four studies met the inclusion criteria. In the primary analysis of head-to-head studies (n = 5), CT had a pooled sensitivity of 83.9% (95% CI, 78.4-88.2%) versus 79.0% (95% CI, 68.8-86.6%) of US (p = 0.44). The pooled specificity of CT was 94% (95% CI, 82.0-98.0%) versus 93.6% (95% CI, 79.4-98.2%) of US (p = 0.85). The concordance of positive or negative test between both modalities was 82.3% (95% CI, 72.1-89.4%). US and CT led to a positive change in management in only 4 to 8% of cases, respectively, when ordered sequentially after the other test. CONCLUSION: The diagnostic performance of CT is comparable to US for the diagnosis of acute cholecystitis, with a high rate of concordance between the two modalities. CLINICAL RELEVANCE STATEMENT: A subsequent US after a positive or negative CT for suspected acute cholecystitis may be unnecessary in most cases. KEY POINTS: When there is clinical suspicion of acute cholecystitis, patients will often undergo both CT and US. CT has similar sensitivity and specificity compared to US for the diagnosis of acute cholecystitis. The concordance rate between CT and US for the diagnosis of acute cholecystitis is 82.3%.

Chronic Chest Computed Tomography Findings Following COVID-19 Pneumonia.

Respiratory symptoms are a frequent manifestation of patients with post-acute sequela of SARS-CoV-2 (PASC), also known as long-COVID. Many cohorts of predominantly hospitalized patients have shown that a significant subset may have persistent chest computed tomography findings for more than 12 months after the acute infection. Proper understanding of the evolving long-term imaging findings and terminology is crucial for accurate imaging interpretation and patient care. The goal of this article is to review the chronic chest computed tomography findings of patients with PASC and common pitfalls.

Diagnostic Performance of MRI for the Detection of Pulmonary Nodules: A Systematic Review and Meta-Analysis.

Purpose To perform a meta-analysis of the diagnostic performance of MRI for the detection of pulmonary nodules, with use of CT as the reference standard. Materials and Methods PubMed, Embase, Scopus, and other databases were systematically searched for studies published from January 2000 to March 2023 evaluating the performance of MRI for diagnosis of lung nodules measuring 4 mm or larger, with CT as reference. Studies including micronodules, nodules without size stratification, or those from which data for contingency tables could not be extracted were excluded. Primary outcomes were the per-lesion sensitivity of MRI and the rate of false-positive nodules per patient (FPP). Subgroup analysis by size and meta-regression with other covariates were performed. The study protocol was registered in the International Prospective Register of Systematic Reviews, or PROSPERO (no. CRD42023437509). Results Ten studies met inclusion criteria (1354 patients and 2062 CT-detected nodules). Overall, per-lesion sensitivity of MRI for nodules measuring 4 mm or larger was 87.7% (95% CI: 81.1, 92.2), while the FPP rate was 12.4% (95% CI: 7.0, 21.1). Subgroup analyses demonstrated that MRI sensitivity was 98.5% (95% CI: 90.4, 99.8) for nodules measuring at least 8-10 mm and 80.5% (95% CI: 71.5, 87.1) for nodules less than 8 mm. Conclusion MRI demonstrated a good overall performance for detection of pulmonary nodules measuring 4 mm or larger and almost equal performance to CT for nodules measuring at least 8-10 mm, with a low rate of FPP. Systematic review registry no. CRD42023437509 Keywords: Lung Nodule, Lung Cancer, Lung Cancer Screening, MRI, CT Supplemental material is available for this article. © RSNA, 2024.

Appropriateness and imaging outcomes of ultrasound, CT, and MR in the emergency department: a retrospective analysis from an urban academic center.

PURPOSE: To evaluate the appropriateness and outcomes of ultrasound (US), computed tomography (CT), and magnetic resonance (MR) orders in the ED. METHODS: We retrospectively reviewed consecutive US, CT, and MR orders for adult ED patients at a tertiary care urban academic center from January to March 2019. The American College of Radiology Appropriateness Criteria (ACRAC) guidelines were primarily used to classify imaging orders as "appropriate" or "inappropriate". Two radiologists in consensus judged specific clinical scenarios that were unavailable in the ACRAC. Final imaging reports were compared with the initial clinical suspicion for imaging and categorized into "normal", "compatible with initial diagnosis", "alternative diagnosis", or "inconclusive". The sample was powered to show a prevalence of inappropriate orders of 30% with a margin of error of 5%. RESULTS: The rate of inappropriate orders was 59.4% for US, 29.1% for CT, and 33.3% for MR. The most commonly imaged systems for each modality were neuro (130/330) and gastrointestinal (95/330) for CT, genitourinary (132/330) and gastrointestinal (121/330) for US, neuro (273/330) and gastrointestinal (37/330) for MR. Compared to inappropriately ordered tests, the final reports of appropriate orders were nearly three times more likely to demonstrate findings compatible with the initial diagnosis for all modalities: US (45.5 vs. 14.3%, p < 0.001), CT (46.6 vs. 14.6%, p < 0.001), and MR (56.3 vs. 21.8%, p < 0.001). Inappropriate orders were more likely to show no abnormalities compared to appropriate orders: US (65.8 vs. 38.8%, p < 0.001), CT (62.5 vs. 34.2%, p < 0.001), and MR (61.8 vs. 38.7%, p < 0.001). CONCLUSION: The prevalence of inappropriate imaging orders in the ED was 59.4% for US, 29.1% for CT, and 33.3% for MR. Appropriately ordered imaging was three times more likely to yield findings compatible with the initial diagnosis across all modalities.

Predicting major adverse cardiovascular events after orthotopic liver transplantation using a supervised machine learning model: A cohort study.

BACKGROUND: Liver transplant (LT) patients have become older and sicker. The rate of post-LT major adverse cardiovascular events (MACE) has increased, and this in turn raises 30-d post-LT mortality. Noninvasive cardiac stress testing loses accuracy when applied to pre-LT cirrhotic patients. AIM: To assess the feasibility and accuracy of a machine learning model used to predict post-LT MACE in a regional cohort. METHODS: This retrospective cohort study involved 575 LT patients from a Southern Brazilian academic center. We developed a predictive model for post-LT MACE (defined as a composite outcome of stroke, new-onset heart failure, severe arrhythmia, and myocardial infarction) using the extreme gradient boosting (XGBoost) machine learning model. We addressed missing data (below 20%) for relevant variables using the k-nearest neighbor imputation method, calculating the mean from the ten nearest neighbors for each case. The modeling dataset included 83 features, encompassing patient and laboratory data, cirrhosis complications, and pre-LT cardiac assessments. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC). We also employed Shapley additive explanations (SHAP) to interpret feature impacts. The dataset was split into training (75%) and testing (25%) sets. Calibration was evaluated using the Brier score. We followed Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis guidelines for reporting. Scikit-learn and SHAP in Python 3 were used for all analyses. The supplementary material includes code for model development and a user-friendly online MACE prediction calculator. RESULTS: Of the 537 included patients, 23 (4.46%) developed in-hospital MACE, with a mean age at transplantation of 52.9 years. The majority, 66.1%, were male. The XGBoost model achieved an impressive AUROC of 0.89 during the training stage. This model exhibited accuracy, precision, recall, and F1-score values of 0.84, 0.85, 0.80, and 0.79, respectively. Calibration, as assessed by the Brier score, indicated excellent model calibration with a score of 0.07. Furthermore, SHAP values highlighted the significance of certain variables in predicting postoperative MACE, with negative noninvasive cardiac stress testing, use of nonselective beta-blockers, direct bilirubin levels, blood type O, and dynamic alterations on myocardial perfusion scintigraphy being the most influential factors at the cohort-wide level. These results highlight the predictive capability of our XGBoost model in assessing the risk of post-LT MACE, making it a valuable tool for clinical practice. CONCLUSION: Our study successfully assessed the feasibility and accuracy of the XGBoost machine learning model in predicting post-LT MACE, using both cardiovascular and hepatic variables. The model demonstrated impressive performance, aligning with literature findings, and exhibited excellent calibration. Notably, our cautious approach to prevent overfitting and data leakage suggests the stability of results when applied to prospective data, reinforcing the model's value as a reliable tool for predicting post-LT MACE in clinical practice.

Lung cancer screening in Brazil: recommendations from the Brazilian Society of Thoracic Surgery, Brazilian Thoracic Association, and Brazilian College of Radiology and Diagnostic Imaging.

Although lung cancer (LC) is one of the most common and lethal tumors, only 15% of patients are diagnosed at an early stage. Smoking is still responsible for more than 85% of cases. Lung cancer screening (LCS) with low-dose CT (LDCT) reduces LC-related mortality by 20%, and that reduction reaches 38% when LCS by LDCT is combined with smoking cessation. In the last decade, a number of countries have adopted population-based LCS as a public health recommendation. Albeit still incipient, discussion on this topic in Brazil is becoming increasingly broad and necessary. With the aim of increasing knowledge and stimulating debate on LCS, the Brazilian Society of Thoracic Surgery, the Brazilian Thoracic Association, and the Brazilian College of Radiology and Diagnostic Imaging convened a panel of experts to prepare recommendations for LCS in Brazil. The recommendations presented here were based on a narrative review of the literature, with an emphasis on large population-based studies, systematic reviews, and the recommendations of international guidelines, and were developed after extensive discussion by the panel of experts. The following topics were reviewed: reasons for screening; general considerations about smoking; epidemiology of LC; eligibility criteria; incidental findings; granulomatous lesions; probabilistic models; minimum requirements for LDCT; volumetric acquisition; risks of screening; minimum structure and role of the multidisciplinary team; practice according to the Lung CT Screening Reporting and Data System; costs versus benefits of screening; and future perspectives for LCS.

Evaluating Acute Pulmonary Changes of Coronavirus 2019: Comparative Analysis of the Pertinent Modalities.

This review explores imaging's crucial role in acute Coronavirus Disease 2019 (COVID-19) assessment. High Resolution Computer Tomography is especially effective in detection of lung abnormalities. Chest radiography has limited utility in the initial stages of COVID-19 infection. Lung Ultrasound has emerged as a valuable, radiation-free tool in critical care, and Magnetic Resonance Imaging shows promise as a Computed Tomography alternative. Typical and atypical findings of COVID-19 by each of these modalities are discussed with emphasis on their prognostic value. Considerations for pediatric and immunocompromised cases are outlined. A comprehensive diagnostic approach is recommended, as radiological diagnosis remains challenging in the acute phase.

Evaluating ChatGPT's Accuracy in Providing Screening Mammography Recommendations among Older Women: Artificial Intelligence and Cancer Communication.

Abstract Objective: The U.S. Preventive Services Task Force (USPSTF) recommends biennial screening mammography through age 74. Guidelines vary as to whether or not they recommended mammography screening to women aged 75 and older. This study aims to determine the ability of ChatGPT to provide appropriate recommendations for breast cancer screening in patients aged 75 years and older. Methods: 12 questions and 4 clinical vignettes addressing fundamental concepts about breast cancer screening and prevention in patients aged 75 years and older were created and asked to ChatGPT three consecutive times to generate 3 sets of responses. The responses were graded by a multi-disciplinary panel of experts in the intersection of breast cancer screening and aging . The responses were graded as 'appropriate', 'inappropriate', or 'unreliable' based on the reviewer's clinical judgment, content of the response, and whether the content was consistent across the three responses . Appropriateness was determined through a majority consensus. Results: The responses generated by ChatGPT were appropriate for 11/17 questions (64%). Three questions were graded as inappropriate (18%) and 2 questions were graded as unreliable (12%). A consensus was not reached on one question (6%) and was graded as no consensus. Conclusions: While recognizing the limitations of ChatGPT, it has potential to provide accurate health care information and could be utilized by healthcare professionals to assist in providing recommendations for breast cancer screening in patients age 75 years and older. Physician oversight will be necessary, due to the possibility of ChatGPT to provide inappropriate and unreliable responses, and the importance of accuracy in medicine.

The International Association for the Study of Lung Cancer Early Lung Imaging Confederation Open-Source Deep Learning and Quantitative Measurement Initiative.

INTRODUCTION: With global adoption of computed tomography (CT) lung cancer screening, there is increasing interest to use artificial intelligence (AI) deep learning methods to improve the clinical management process. To enable AI research using an open-source, cloud-based, globally distributed, screening CT imaging data set and computational environment that are compliant with the most stringent international privacy regulations that also protect the intellectual properties of researchers, the International Association for the Study of Lung Cancer sponsored development of the Early Lung Imaging Confederation (ELIC) resource in 2018. The objective of this report is to describe the updated capabilities of ELIC and illustrate how this resource can be used for clinically relevant AI research. METHODS: In this second phase of the initiative, metadata and screening CT scans from two time points were collected from 100 screening participants in seven countries. An automated deep learning AI lung segmentation algorithm, automated quantitative emphysema metrics, and a quantitative lung nodule volume measurement algorithm were run on these scans. RESULTS: A total of 1394 CTs were collected from 697 participants. The LAV950 quantitative emphysema metric was found to be potentially useful in distinguishing lung cancer from benign cases using a combined slice thickness more than or equal to 2.5 mm. Lung nodule volume change measurements had better sensitivity and specificity for classifying malignant from benign lung nodules when applied to solid lung nodules from high-quality CT scans. CONCLUSIONS: These initial experiments revealed that ELIC can support deep learning AI and quantitative imaging analyses on diverse and globally distributed cloud-based data sets.