Computed Tomography Pulmonary Angiography Research Articles

The Utility of Modified Wells Score as a Pre-Test Risk Stratification Tool in Suspected Cases of Pulmonary Embolism Undergoing Computed Tomography Pulmonary Angiogram

Objectives: To assess the validity of pre-test evaluation of patients with Modified Wells score at our facility to improve diagnostic yield of CT pulmonary angiography for pulmonary embolism. Study Design: Prospective longitudinal study. Place and Duration: Department of Computed Tomography, Armed Forces Institute of Radiology & Imaging, Pak Emirates Military Hospital, Rawalpindi Pakistan, from Feb to Jul 2019. Methodology: After approval from the Ethical Review Board, 60 patients with clinical suspicion of pulmonary embolism were selected. Modified Wells score was calculated for each patient, and a senior consultant reviewed their CT Pulmonary Angiograms. Results: Sixty patients were assessed from February to July 2019. Pulmonary embolism was diagnosed in 21(35%) patients; 22(37%) patients were unremarkable, while alternate diagnosis was provided for 17(28%) patients. Modified Wells score was 66.6% sensitive and 74.3% specific for diagnosis of pulmonary embolism, with a positive predictive value of 58.3% and a negative predictive value of 80.5%. Conclusion: Modified Wells score carries a good negative predictive value for ruling out pulmonary embolism. Hospitals must devise their departmental protocol utilizing risk stratification tools, lab tests and other safer imaging alternatives, where applicable.

Automated detection, segmentation and measurement of major vessels and the trachea in CT pulmonary angiography

Mediastinal structure measurements are important for the radiologist’s review of computed tomography pulmonary angiography (CTPA) examinations. In the reporting process, radiologists make measurements of diameters, volumes, and organ densities for image quality assessment and risk stratification. However, manual measurement of these features is time consuming. Here, we sought to develop a time-saving automated algorithm that can accurately detect, segment and measure mediastinal structures in routine clinical CTPA examinations. In this study, 700 CTPA examinations collected and annotated. Of these, a training set of 180 examinations were used to develop a fully automated deterministic algorithm. On the test set of 520 examinations, two radiologists validated the detection and segmentation performance quantitatively, and ground truth was annotated to validate the measurement performance. External validation was performed in 47 CTPAs from two independent datasets. The system had 86–100% detection and segmentation accuracy in the different tasks. The automatic measurements correlated well to those of the radiologist (Pearson’s r 0.68–0.99). Taken together, the fully automated algorithm accurately detected, segmented, and measured mediastinal structures in routine CTPA examinations having an adequate representation of common artifacts and medical conditions.

Diagnostic Yield, Radiation Exposure, and the Role of Clinical Decision Rules to Limit Computed Tomographic Pulmonary Angiography-Associated Complications.

Computed tomographic pulmonary angiography (CT-PA) is associated with significant cost, contrast, and radiation exposure. Clinical decision rules (CDRs) reduce the need for diagnostic imaging; however, their utility in the medical intensive care unit (MICU) remains unknown. We explored the diagnostic yield and complications associated with CT-PA (radiation exposure and contrast-induced acute kidney injury [AKI]) while investigating the efficacy of CDRs to reduce unnecessary testing. All CT-PAs performed in an academic MICU for 4 years were retrospectively reviewed. The Wells and revised Geneva scores (CDRs) and radiation dose per CT-PA were calculated, and the incidence of post-CT-PA AKI was recorded. A total of 439 studies were analyzed; the diagnostic yield was 11% (48 PEs). Positive CT-PAs were associated with a higher Wells score (5.8 versus 3.2, P < 0.001), but similar revised Geneva scores (6.4 versus 6.0, P = 0.32). A Wells score of ≥4 had a positive likelihood ratio of 2.1 with a negative predictive value of 98.2. More than half (88.9%) of patients with a Wells score of ≤4 developed an AKI, with 55.6% of those having recovery of renal function. There is overutilization of CT-PA in the MICU. The Wells score retains its negative predictive value in critically ill adult patients and may aid to limit radiation exposure and contrast-induced AKI in MICU.

A Retrospective Study on the Incidence of Pulmonary Embolism in Immobilized Spinal Cord Injury Patients.

Introduction The development of pulmonary embolism (PE)remains the main cause of morbidity and mortality in patients withspinal cord injuries.Multiple factors have been identified to predict the presence of PE in spinal cord injury patients, however, local data is scarce. The objective of this study is to evaluate the incidence of PE among spinal cord injury patients and analysis of predictive risk factors. Material and methods This is a retrospective study among spinal cord injury patients admitted to a tertiary hospital in Malaysia between January 2018 and December 2019. All spinal cord injury patients with symptoms suggestive of venous thromboembolism (VTE) such as tachycardia and shortness of breath were included in this study. Demographic data such as age, gender, types of VTE prophylaxis (mechanical or chemical), and radiological findings of computerized tomography pulmonary angiogram (CTPA) were analyzed. Results A total of 373 patients were included in the study. 301 of them have undergone spinal surgery. There were 251 males (75.75%) and 124 females (33.24%). The mean age of the subjects was 48.63 ± 17.45 years. The mean length of hospital stay is 42.74 ± 41.51 days. 151 (40.48%) patients received DVT prophylaxis while 222 (59.52%) patients did not. The incidence of PE among spinal cord injury patients with VTE symptoms is 25 patients (6.70%). Of that, 15 patients received medical treatment only, seven received mechanical and pharmacological prophylaxis and three did not receive any prophylaxis. In the variable analyses, none of the variables (age and types of prophylaxis) could significantly predict the occurrence of VTE. Conclusion PE is a frequent complication in immobilized spinal cord injury patients despite receiving thromboprophylaxis treatment. No strong predictors for PE were identified in the present study. However, patients who received any type of thromboprophylaxis was found to be statistically significant when compared to patients who did not receive any thromboprophylaxis in term of the presence of PE.

Cardiovascular metrics on CT pulmonary angiography in patients with pulmonary hypertension — re-evaluation under the updated guidelines of pulmonary hypertension

PurposeTo re-assess cardiovascular metrics on computed tomography pulmonary angiography (CTPA) in predicting pulmonary hypertension (PH) under the 2022 ESC/ERS guidelines.Materials and methodsThis observational study retrospectively included 272 patients (female 143, mean age = 54.9 ± 12.5 years old) with suspected PH. 218 patients were grouped to evaluate cardiovascular metrics on CTPA and develop a binary logistic regression model. The other 54 patients were grouped into the validation group to assess the performance of the prediction model under the updated criteria. Based on mean pulmonary artery pressure (mPAP), patients were divided into three groups: group A consisted of patients with mPAP ≤ 20 mmHg, group B included patients with 20 mmHg < mPAP < 25 mmHg, and group C comprised patients with mPAP ≥ 25 mmHg. Cardiovascular metrics among the three groups were compared, and receiver operating characteristic curves (ROCs) were used to evaluate the performance of cardiovascular metrics in predicting mPAP > 20 mmHg.ResultsThe main pulmonary arterial diameter (MPAd), MPAd/ascending aorta diameter ratio (MPAd/AAd ratio), and right ventricular free wall thickness (RVFWT) showed significant differences among the three groups (p < 0.05). The area under curve (AUC) of MPAd was larger than MPAd/AAd ratio and RVFWT. A MPAd cutoff value of 30.0 mm has a sensitivity of 83.1% and a specificity of 90.4%. The AUC of the binary logistic regression model (Z = − 12.98187 + 0.31053 MPAd + 1.04863 RVFWT) was 0.938 ± 0.018. In the validation group, the AUC, sensitivity, specificity, and accuracy of the prediction model were 0.878, 92.7%, 76.9%, and 88.9%, respectively.ConclusionUnder the updated criteria, MPAd with a threshold value of 30.0 mm has better sensitivity and specificity in predicting PH. The binary logistic regression model may improve the diagnostic accuracy.Critical relevance statementUnder the updated criteria, the main pulmonary arterial diameter with a threshold value of 30.0 mm has better sensitivity and specificity in predicting pulmonary hypertension. The binary logistic regression model may improve diagnostic accuracy.Key points• According to 2022 ESC/ERS guidelines, a MPAd cutoff value of 30.0 mm has better sensitivity and specificity in predicting mPAP > 20 mmHg• A binary logistic regression model (Z = − 12.98187 + 0.31053 MPAd + 1.04863 RVFWT) was developed and had a sensitivity, specificity, and accuracy of 92.7%, 76.9%, and 88.9% in predicting mPAP > 20 mmHg.• A binary logistic regression prediction model outperforms MPAd in predicting mPAP > 20 mmHg.Graphical

Incidental central pulmonary emboli in outpatients with deep vein thrombosis.

While it has been demonstrated that large asymptomatic pulmonary embolism (PE) can occur, many studies have assessed inpatients or patients with known risk factors for venous thromboembolism (such as malignancy). There are few data regarding incidental PE in outpatients with deep vein thrombosis (DVT) or assessing whether these emboli occur centrally or peripherally in the pulmonary arteries. To determine the rates of incidental central and peripheral PE in outpatients with proximal and distal DVT. This was a retrospective cross-sectional study of 120 patients attending our community imaging clinic between January 2015 and March 2020, with ultrasound-confirmed lower limb DVT, and subsequent computed tomography pulmonary angiogram (CTPA) within 24 h. Exclusion criteria were symptoms indicative of PE. Imaging reports were retrospectively reviewed to record the proximity of DVT and the location of any PE as either central (pulmonary trunk or main pulmonary arteries) or peripheral (lobar, segmental or subsegmental pulmonary arteries). Incidental PE occurred in 71 patients (59.2%), with a rate of 77.5% in proximal and 50.0% in distal DVT. Sixteen patients had central PE (13.3% of all patients with DVT; 22.5% of all patients with PE). Two patients (both with proximal DVT) had PE in the pulmonary trunk. Incidental PE occur frequently in outpatients with lower limb DVT, including the possibility of asymptomatic central emboli. Further studies with larger patient cohorts would be useful to assess the utility of baseline chest imaging in outpatients with DVT.

Subsegmental Pulmonary Embolism.

Subsegmental pulmonary embolism (SSPE) is increasingly diagnosed with the growing use and technological advancements of multidetector computed tomography pulmonary angiography. Its diagnosis is challenging, and some presumed SSPE may actually represent imaging artifacts. Indirect evidence and results from small observational studies suggest that SSPE may be more benign than more proximal pulmonary embolism, and may thus not always require treatment. Therefore, guidelines suggest to consider a management strategy without anticoagulation in selected patients with SSPE at low risk of recurrent venous thromboembolism (VTE), in whom proximal deep vein thrombosis is excluded. Recently, a large prospective study among low-risk patients with SSPE who were left untreated showed a higher VTE recurrence risk than initially deemed acceptable by the investigators, and thus was prematurely interrupted after recruitment of 97% of the target population. However, the risk-benefit ratio of anticoagulation for low-risk patients with SSPE remains unclear, and results from randomized trials are needed to answer the question about their optimal management.

The level of C-reactive protein in patients with acute pulmonary embolism

Background. Modern risk stratification in patients with pulmonary embolism (PE) does not consider all possible factors that may influence the course and prognosis of the disease. In view of this, it is promising to study inflammatory markers, especially C-reactive protein (CRP), for their prognostic value in PE. The purpose was to study the clinical and prognostic value of increased CRP levels in patients with PE during hospitalization. Material and methods. One hundred and three patients with diagnosis of acute PE confirmed by computed tomographic pulmonary angiography (CTPA) or by autopsy were examined, their СRP level was evaluated during hospitalization. In addition to a general clinical examination, all patients were screened for PE risk factors and risk of premature death according to the ESC 2019 guidelines. La-boratory tests included quantitative determination of CPK-MB, D-dimer and СRP in addition to basic ones. Electrocardiography, echocardiography, and CTPA were performed as part of the instrumental studies. Results. Depending on the СRP median (26 mg/l), patients were divided into two groups. Group 1 (СRP &lt; 26 mg/l) included 52 people, group 2 (СRP &gt; 26 mg/l) — 51 patients. It was found that in group 1, patients with low risk predominated (p = 0.044) and the average pulmonary embolism severity index was significantly lower (p = 0.011). In group 2, the average blood oxygen saturation (SpO2) during hospitalization was significantly lower (p = 0.004), the average parameters of the left ventricle, namely end-diastolic and end-systolic volume, were significantly greater (p = 0.045 and p = 0.043, respectively), and the ejection fraction was significantly lower (p = 0.002). Also, the average levels of leukocytes (p = 0.0003), neutrophils (p = 0.00001), urea (p = 0.008), CPK-MB (p = 0.028) and erythrocyte sedimentation rate (p = 0.012) were significantly higher in group 2. Correlation analysis revealed a reliable inverse connection between CRP levels and initial SpO2 and a direct relationship between CRP level and the size of the right atrium, CPK-MB level, and increased risk of early mortality from PE. Conclusions. Elevated CRP was found in 92 % of patients with acute PE and was associated with a significant decrease in SpO2 (p &lt; 0.004), increased left ventricular size (p &lt; 0.05), and decreased ejection fraction (p &lt; 0.003). Correlation analysis revealed a relationship between CRP level and the risk according to the pulmonary embolism severity index, SpO2, the level of CPK-MB and the size of the left atrium.

Risk stratification of poor short-term prognosis in patients with high-risk pulmonary embolism

Background. Pulmonary embolism (PE) is the third leading cause of cardiovascular mortality. High-risk PE patients have more severe course of disease and poor prognosis. The question of additional factors which may be associated with prognosis in such patients remains poorly understood. The aim is to identify factors associated with mortality in high-risk PE patients and to determine contribution of these factors to the risk stratification of death. Materials and methods. A retrospective analysis of 635 medical cards of patients with PE diagnosis hospitalized in Kharkiv City Clinical Hospital 8 from January 1, 2017 to January 1, 2023 was conducted. The diagnosis was verified by multispiral computed tomography pulmonary angiography (CTPA) and/or by autopsy. Patients were divided into groups: group І — high-risk (113 — 17.8 %); group II — low-risk (522 — 82.2 %). Group I was divided into subgroups: IA — 63 (55.8 %) individuals, who were discharged from the hospital with improved status, IB — 50 (44.2 %) people, who died in hospital. Clinical, anamnestic, laboratory and instrumental parameters were measured. Statistical analysis was performed. Results. Patients from group I were older, there were more women among them. In subgroup ІВ, there are more women, more cases of obesity and deep vein thrombosis. Patients from group IB had lower blood oxygen saturation (SpO2) and systolic blood pressure (SBP), lower left ventricular ejection fraction (LVEF) measured by echocardiography. Bilateral segmental pulmonary embolism according to CTPA was more common in IB group. According to uni- and multivariate analysis, independent factors affecting in-hospital mortality in patients with high-risk PE were determined, among them: female gender, presence of bifurcation thrombus, lower SBP, SpO2, LVEF. Using the additional regression analysis, the formula for individual prediction of the risk of in-hospital mortality in high-risk PE patients was created. According to the receiver operating characteristic analysis, the sensitivity of the formula is 64.0 %, specificity is 92.1 %. The SBAFS (S — SpO2, B — Bifurcation thrombus, A — Arterial hypotension, F — left ventricular ejection Fraction, S — female Sex) score was formed, where the score of 2 or more indicates additional high risk of in-hospital mortality in high-risk PE patients. Conclusion. The high-risk PE patients who died during hospitalization had more cases of obesity, hypertension, lower SpO2, SBP, and LVEF, and there were more women among them. According to the multivariate logistic regression analysis, additional independent factors associated with in-hospital mortality in high-risk PE patients are female sex, thrombus in bifurcation of the pulmonary trunk, SpO2, SBP, LVEF. The formula for risk stratification of in-hospital death in high-risk PE patients was created and validated. The SBAFS score was formed to indicate additional high risk of in-hospital mortality in high-risk PE patients.

Enhancing preoperative assessment in chronic thromboembolic pulmonary hypertension: A comprehensive analysis of interobserver agreement and proximity-based CT pulmonary angiography scoring

BackgroundSurgical risk in chronic thromboembolic pulmonary hypertension (CTEPH) depends on the proximity of thromboembolism on CT pulmonary angiography (CTPA). We assessed interobserver agreement for the quantification of thromboembolic lesions in CTEPH using a novel CTPA scoring index. MethodsForty CTEPH patients (mean age, 58 ± 16 years; 19 men) with preoperative CTPA who underwent pulmonary endarterectomy (PEA) (08/2020-09/2021) were retrospectively included. Three radiologists scored each CTPA for chronic thromboembolism (occlusions, eccentric thickening, webs) using a 32-vessel model of the pulmonary vasculature, with interobserver agreement evaluated using Fleiss’ kappa. CT level of disease was determined by the most proximal chronic thromboembolism: level 1 (main pulmonary artery), 2 (lobar), 3 (segmental) and 4 (subsegmental), and compared to surgical level at PEA. ResultsInterobserver agreement for CT level of disease was moderate overall (κ = 0.52). Agreement was substantial overall at the main/lobar level (κ, mean = 0.71) when excluding the left upper lobe (κ = 0.17). Though segmental and subsegmental agreement suffered (κ = 0.31), we found substantial agreement for occlusions (κ = 0.72) compared to eccentric thickening (κ = 0.45) and webs (κ = 0.14). Correlation between CT level and surgical level was strong overall (τb = 0.73) and in the right lung (τb = 0.68), but weak in the left lung (τb = 0.42) (p < 0.05). Radiologists often over- and underestimated the proximal extent of disease in right and left lung, respectively. ConclusionsCT level of disease demonstrated good agreement between radiologists and was highly predictive of the surgical level in CTEPH. Occlusions were the most reliable sign of chronic thromboembolism and are important in assessing the segmental vasculature.

Epicardial adipose tissue as aprognostic marker in acute pulmonary embolism.

Epicardial adipose tissue (EAT) has been established as aquantitative imaging biomarker associated with disease severity in coronary heart disease. Our aim was to use this prognostic marker derived from computed tomography pulmonary angiography (CTPA) for the prediction of mortality and prognosis in patients with acute pulmonary embolism. The clinical database was retrospectively screened for patients with acute pulmonary embolism between 2015 and 2021. Overall, 513 patients (216 female, 42.1%) were included in the analysis. The study end-point was 30-day mortality. Epicardial adipose tissue was measured on the diagnostic CTPA in asemiquantitative manner. The volume and density of EAT were measured for every patient. Overall, 60patients (10.4%) died within the 30-day observation period. The mean EAT volume was 128.3 ± 65.0 cm3 in survivors and 154.6 ± 84.5 cm3 in nonsurvivors (p = 0.02). The density of EAT was -79.4 ± 8.3 HU in survivors and -76.0 ± 8.4 HU in nonsurvivors (p = 0.86), and EAT density was associated with 30-day mortality (odds ratio [OR] = 1.07; 95% confidence interval [CI]: 1.03; 1.1, p < 0.001) but did not remain statistically significant in multivariable analysis. No association was identified between EAT volume and 30-day mortality (OR = 1.0; 95% CI: 1.0; 1.0, p = 0.48). There might be an association between EAT density and mortality in patients with acute pulmonary embolism. Further studies are needed to elucidate the prognostic relevance of EAT parameters in patients with acute pulmonary embolism.

Seeking an optimal approach for Computer-aided Diagnosis of Pulmonary Embolism

Pulmonary Embolism (PE) represents a thrombus (“blood clot”), usually originating from a lower extremity vein, that travels to the blood vessels in the lung, causing vascular obstruction and in some patients death. This disorder is commonly diagnosed using Computed Tomography Pulmonary Angiography (CTPA). Deep learning holds great promise for the Computer-aided Diagnosis (CAD) of PE. However, numerous deep learning methods, such as Convolutional Neural Networks (CNN) and Transformer-based models, exist for a given task, causing great confusion regarding the development of CAD systems for PE. To address this confusion, we present a comprehensive analysis of competing deep learning methods applicable to PE diagnosis based on four datasets. First, we use the RSNA PE dataset, which includes (weak) slice-level and exam-level labels, for PE classification and diagnosis, respectively. At the slice level, we compare CNNs with the Vision Transformer (ViT) and the Swin Transformer. We also investigate the impact of self-supervised versus (fully) supervised ImageNet pre-training, and transfer learning over training models from scratch. Additionally, at the exam level, we compare sequence model learning with our proposed transformer-based architecture, Embedding-based ViT (E-ViT). For the second and third datasets, we utilize the CAD-PE Challenge Dataset and Ferdowsi University of Mashad’s PE Dataset, where we convert (strong) clot-level masks into slice-level annotations to evaluate the optimal CNN model for slice-level PE classification. Finally, we use our in-house PE-CAD dataset, which contains (strong) clot-level masks. Here, we investigate the impact of our vessel-oriented image representations and self-supervised pre-training on PE false positive reduction at the clot level across image dimensions (2D, 2.5D, and 3D). Our experiments show that (1) transfer learning boosts performance despite differences between photographic images and CTPA scans; (2) self-supervised pre-training can surpass (fully) supervised pre-training; (3) transformer-based models demonstrate comparable performance but slower convergence compared with CNNs for slice-level PE classification; (4) model trained on the RSNA PE dataset demonstrates promising performance when tested on unseen datasets for slice-level PE classification; (5) our E-ViT framework excels in handling variable numbers of slices and outperforms sequence model learning for exam-level diagnosis; and (6) vessel-oriented image representation and self-supervised pre-training both enhance performance for PE false positive reduction across image dimensions. Our optimal approach surpasses state-of-the-art results on the RSNA PE dataset, enhancing AUC by 0.62% (slice-level) and 2.22% (exam-level). On our in-house PE-CAD dataset, 3D vessel-oriented images improve performance from 80.07% to 91.35%, a remarkable 11% gain. Codes are available at GitHub.com/JLiangLab/CAD_PE.

A case of metastatic breast cancer complicated by pulmonary tumor thrombotic microangiopathy

Pulmonary tumor thrombotic microangiopathy is a malignancy-related complication with rapid progression and high mortality. To improve the understanding of the disease, early diagnosis and treatment are key to successful treatment. A 39-year-old patient with pulmonary hypertension transferred from another hospital was admitted to the First Affiliated Hospital of Guangzhou Medical University on September 26, 2021. The patient developed shortness of breath and progressive exacerbation over the past month. No pulmonary artery embolism was seen on computed tomography pulmonary angiography (CTPA) at the outside hospital where the breast cancer was diagnosed. Pulmonary tumor thrombotic microangiopathy was immediately considered on admission and oncological endocrine therapy was started. After treatment, the patient's dyspnoea improved, PET-CT showed significant tumor regression, and cardiac ultrasound showed a significant decrease in pulmonary artery pressure. The successful treatment experience of this case was summarized for reference.

Detection and severity quantification of pulmonary embolism with 3D CT data using an automated deep learning-based artificial solution

PurposeThe purpose of this study was to propose a deep learning-based approach to detect pulmonary embolism and quantify its severity using the Qanadli score and the right-to-left ventricle diameter (RV/LV) ratio on three-dimensional (3D) computed tomography pulmonary angiography (CTPA) examinations with limited annotations. Materials and methodsUsing a database of 3D CTPA examinations of 1268 patients with image-level annotations, and two other public datasets of CTPA examinations from 91 (CAD-PE) and 35 (FUME-PE) patients with pixel-level annotations, a pipeline consisting of: (i), detecting blood clots; (ii), performing PE-positive versus negative classification; (iii), estimating the Qanadli score; and (iv), predicting RV/LV diameter ratio was followed. The method was evaluated on a test set including 378 patients. The performance of PE classification and severity quantification was quantitatively assessed using an area under the curve (AUC) analysis for PE classification and a coefficient of determination (R²) for the Qanadli score and the RV/LV diameter ratio. ResultsQuantitative evaluation led to an overall AUC of 0.870 (95% confidence interval [CI]: 0.850–0.900) for PE classification task on the training set and an AUC of 0.852 (95% CI: 0.810–0.890) on the test set. Regression analysis yielded R² value of 0.717 (95% CI: 0.668–0.760) and of 0.723 (95% CI: 0.668–0.766) for the Qanadli score and the RV/LV diameter ratio estimation, respectively on the test set. ConclusionThis study shows the feasibility of utilizing AI-based assistance tools in detecting blood clots and estimating PE severity scores with 3D CTPA examinations. This is achieved by leveraging blood clots and cardiac segmentations. Further studies are needed to assess the effectiveness of these tools in clinical practice.

Isolated pulmonary artery involvement in Takayasu arteritis: case report and review of the literature

BackgroundTakayasu arteritis (TAK) is a chronic inflammatory arteritis. It most often affects non-specific large vessel progressively, such as the aorta and its branches. The diagnosis in TAK is typically delayed. Isolated pulmonary artery involvement in Takayasu arteritis is uncommon. Owing to its rarity, the diagnosis is challenging and requires an integrated approach comprising clinical and imaging findings. In order to facilitate early diagnosis of TAK for clinicians, wider use of non-invasive imaging is impacting this.Case presentationIn this report, we present the imaging characteristics of a rare isolated pulmonary artery involvement in Takayasu arteritis. Pulmonary computed tomography angiography revealed only limited thick walls of both proximal pulmonary artery and stenosis of pulmonary artery lumen, and the other large blood vessels were not involved. The patient undergone pulmonary endarterectomy and pulmonary angioplasty. Then, approximately one month afterward, she passed away due to heart failure.ConclusionsImaging examination is the main basis for diagnosing this disease. This impression might improve disease awareness among doctors and progress in diagnosis.

Incidence of pulmonary thromboembolism in patients who underwent CT pulmonary angiography in the first quarter of 2023 at CHC Rijeka

Cilj: Plućna tromboembolija (PTE) je hitno medicinsko stanje kod kojeg dolazi do okluzije jedne ili više grana plućnih arterija tromboembolusom. CT plućna angiografija (CTPA) je brza, neinvazivna radiološka metoda kojom se postavlja dijagnoza PTE. Glavni cilj ovog rada je odrediti incidenciju pozitivnih PTE nalaza kod 524 pacijenta podvrgnutih CTPA te odrediti učestalost zahvaćenih grana plućnih arterija. Nadalje, jedan od ciljeva je i odrediti incidenciju usputnih nalaza (plućni nodusi, tumor pluća, pneumonija, infarkt pluća, pleuralni izljev). Ispitanici i metode: U ovom retrospektivnom istraživanju prikupljeni su podaci 524 bolesnika koji su zbog sumnje na PTE podvrgnuti CTPA pregledu na Kliničkom zavodu za dijagnostičku i intervencijsku radiologiju, na lokalitetu Sušak, u Kliničkom bolničkom centru (KBC) Rijeka, u razdoblju od siječnja do travnja 2023. godine. Svi potrebni podaci navedeni u nalazu radiologa prikupljeni su iz integriranog bolničkog informacijskog sustava (IBIS) i pripadajuće aplikacije (ISSA). Rezultati: Od ukupno 524 bolesnika podvrgnutih CTPA, PTE je dokazana kod 97 ili 18,51% pacijenata. U našem istraživanju bio je podjednak broj ispitanika muškog i ženskog spola (260 ili 49,62% naprema 264 ili 50,39%). Nešto je veća incidencija PTE bila kod muškaraca (52 ili 53,61%) nego kod žena (45 ili 46,39%). Najveći broj bolesnika (262 ili 50,00%) pripadao je dobnom razredu između 61. i 80. godine života. Usputni nalazi opisani na CTPA, prema redoslijedu učestalosti su bili: pneumonije (37,15%), pleuralni izljevi (34,86%), plućni nodusi (16,79%), tumori pluća (6,87%) i plućni infarkti (4,33%). Zaključak: CT plućna angiografija predstavlja zlatni standard za detekciju plućne tromboembolije. Pregledom dostupne literature, incidencija PTE je komparabilna s rezultatima našeg istraživanja. U našoj, kao i u većini recentnih studija, PTE je najčešće detektirana u segmentalnim i subsegmentalnim granama plućnih arterija, dok se incidencija usputnih nalaza detektiranih na CTPA ne podudara s rezultatima drugih istraživanja.

Algorithms of Electrocardiographic Changes for Quantitative and Localization Analysis of Thrombus Burden in Patients with Acute Pulmonary Thromboembolism.

Various electrocardiographic (ECG) abnormalities are associated with the severity of pulmonary thromboembolism (PTE). The utility of evaluating the clot burden of PTE based on ECG findings alone has yet to be thoroughly investigated in Chinese patients. The aim of this study was therefore to use ECG signs to establish novel models for quantitative and localization analysis of clot burden in patients with acute PTE. Acute PTE patients from three centers were enrolled between 2015 and 2019 in a retrospective cohort study (NCT03802929). We analyzed the 12-lead ECGs at admission and studied computed tomography pulmonary angiography (CTPA) features to obtain the Qanadli score of clot burden and location of thrombus. Novel risk prediction models were developed and validated using derivation and external validation cohorts, respectively. A total of 341 acute PTE patients were screened, of whom 246 (72.1%) were from Shanghai Tenth People's Hospital, 71 (20.8%) were from Shanghai Pulmonary Hospital and 24 (7.0%) were from Qidong People's Hospital. In the derivation cohort, predictors included in the final models were congestive heart failure, chronic obstructive pulmonary disease, hypertension, coronary heart disease, atrial fibrillation and ECG abnormalities. The CHARIS (COPD/CHF/CHD, HTN, Atrial arrhythmias/AF, RBBB/RAD, Inverted T wave and S1Q3T3/ Sinus tachycardia) I model was established for quantitatively assessing Qanadli score. It had moderate discrimination in both the derivation cohort (concordance index (c-index) of 0.720, 95% CI 0.655-0.780) and the validation cohort (c-index of 0.663, 95% CI 0.559-0.757). The CHARIS II model was used to predict the probability of trunk obstruction. It showed similar discrimination in the derivation cohort (c-index of 0.753, 95% CI 0.691-0.811) and in the validation cohort (c-index of 0.741, 95% CI 0.641-0.827). Calibration curves and Hosmer-Lemeshow test confirmed the accuracy of the risk prediction equations in the external validation dataset. Decision curve analysis showed the CHARIS I and CHARIS II algorithms had positive net benefits in both the derivation and validation cohorts. From quantitative and localization perspectives, the CHARIS algorithms can identify acute PTE patients with heavy thrombus burdens prior to imaging diagnosis. NCT03802929, https://www.clinicaltrials.gov/study/NCT03802929.

Automated detection and segmentation of pulmonary embolisms on computed tomography pulmonary angiography (CTPA) using deep learning but without manual outlining.

We present a novel computer algorithm to automatically detect and segment pulmonary embolisms (PEs) on computed tomography pulmonary angiography (CTPA). This algorithm is based on deep learning but does not require manual outlines of the PE regions. Given a CTPA scan, both intra- and extra-pulmonary arteries were firstly segmented. The arteries were then partitioned into several parts based on size (radius). Adaptive thresholding and constrained morphological operations were used to identify suspicious PE regions within each part. The confidence of a suspicious region to be PE was scored based on its contrast in the arteries. This approach was applied to the publicly available RSNA Pulmonary Embolism CT Dataset (RSNA-PE) to identify three-dimensional (3-D) PE negative and positive image patches, which were used to train a 3-D Recurrent Residual U-Net (R2-Unet) to automatically segment PE. The feasibility of this computer algorithm was validated on an independent test set consisting of 91 CTPA scans acquired from a different medical institute, where the PE regions were manually located and outlined by a thoracic radiologist (>18 years' experience). An R2-Unet model was also trained and validated on the manual outlines using a 5-fold cross-validation method. The CNN model trained on the high-confident PE regions showed a Dice coefficient of 0.676±0.168 and a false positive rate of 1.86 per CT scan, while the CNN model trained on the manual outlines demonstrated a Dice coefficient of 0.647±0.192 and a false positive rate of 4.20 per CT scan. The former model performed significantly better than the latter model (p<0.01). The promising performance of the developed PE detection and segmentation algorithm suggests the feasibility of training a deep learning network without dedicating significant efforts to manual annotations of the PE regions on CTPA scans.

Development and validation of a lung graph-based machine learning model to predict acute pulmonary thromboembolism on chest noncontrast computed tomography.

Computed tomography pulmonary angiography (CTPA) is a first-line noninvasive method to diagnose acute pulmonary thromboembolism (APE); however, whether chest noncontrast CT (NC-CT) could aid in the diagnosis of APE remains unknown. The aim of this study was to build and evaluate a holistic lung graph-based machine learning (HLG-ML) using NC-CT for the diagnosis of APE and to compare its performance with that of radiologists and the YEARS algorithm. This study enrolled 178 cases (77 males; age 63.9±16.7 years) who underwent NC-CT and CTPA in the same day from January 2019 to December 2020. Of these patients, 133 (75% of cases; 58 males; age 65.4±15.6 years) were placed into a training group and 45 (25% of cases; 19 males; age 59.6±19.2 years) into a testing group. The other 43 cases (18 males; age 62.8±20.0 years) were used to externally validate the model between January 2021 and March 2022. A HLG was developed with a pulmonary radiomics descriptor derived from NC-CT images. The approach extracted local radiomics features and encoded these local features into a radiomics descriptor as a characterization of global radiomics feature distribution. Subsequently, 8 ML models were trained and compared based on the radiomics descriptor. In the validation group, area under the curves (AUCs) of the HLG-ML model in the diagnosis of APE were compared with those of the 3 radiologists and the YEARS algorithm. Among the 8 ML models, gradient boosting decision tree demonstrated the best classification performance (AUC =0.772) on the training set. In the testing set, the AUC of gradient boosting decision trees was 0.857 [95% confidence intervals (CIs): 0.699-0.951]. In the validation set, the performance of gradient boosting decision tree (AUC =0.810; 95% CI: 0.669-0.952; Youden index =0.621) outperformed 3 radiologists (AUC =0.508, 95% CI: 0.335-0.681, Youden index =0.016; AUC =0.504, 95% CI: 0.354-0.654, Youden index =0.008; AUC =0.527, 95% CI: 0.363-0.691, Youden index =0.050) and the YEARS algorithm (AUC =0.618; 95% CI: 0.469-0.767; Youden index =0.237). Compared to all 3 radiologists and the YEARS algorithm, the proposed HLG-based gradient boosting decision tree model achieved a superior performance in the diagnosis of APE on the NC-CT and may thus serve as a valuable tool for physicians in the diagnosis of APE.

Prospective Evaluation of AI Triage of Pulmonary Emboli on CT Pulmonary Angiograms.

Background Artificial intelligence (AI) algorithms have shown high accuracy for detection of pulmonary embolism (PE) on CT pulmonary angiography (CTPA) studies in academic studies. Purpose To determine whether use of an AI triage system to detect PE on CTPA studies improves radiologist performance or examination and report turnaround times in a clinical setting. Materials and Methods This prospective single-center study included adult participants who underwent CTPA for suspected PE in a clinical practice setting. Consecutive CTPA studies were evaluated in two phases, first by radiologists alone (n = 31) (May 2021 to June 2021) and then by radiologists aided by a commercially available AI triage system (n = 37) (September 2021 to December 2021). Sixty-two percent of radiologists (26 of 42 radiologists) interpreted studies in both phases. The reference standard was determined by an independent re-review of studies by thoracic radiologists and was used to calculate performance metrics. Diagnostic accuracy and turnaround times were compared using Pearson χ2 and Wilcoxon rank sum tests. Results Phases 1 and 2 included 503 studies (participant mean age, 54.0 years ± 17.8 [SD]; 275 female, 228 male) and 1023 studies (participant mean age, 55.1 years ± 17.5; 583 female, 440 male), respectively. In phases 1 and 2, 14.5% (73 of 503) and 15.9% (163 of 1023) of CTPA studies were positive for PE (P = .47). Mean wait time for positive PE studies decreased from 21.5 minutes without AI to 11.3 minutes with AI (P < .001). The accuracy and miss rate, respectively, for radiologist detection of any PE on CTPA studies was 97.6% and 12.3% without AI and 98.6% and 6.1% with AI, which was not significantly different (P = .15 and P = .11, respectively). Conclusion The use of an AI triage system to detect any PE on CTPA studies improved wait times but did not improve radiologist accuracy, miss rate, or examination and report turnaround times. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Murphy and Tee in this issue.