A Deep Learning-Based Radiomic Classifier for Usual Interstitial Pneumonia.

BACKGROUND: Because chest CT scan has largely supplanted surgical lung biopsy for diagnosing most cases of interstitial lung disease (ILD), tools to standardize CT scan interpretation are urgently needed. RESEARCH QUESTION: Does a deep learning (DL)-based classifier for usual interstitial pneumonia (UIP) derived using CT scan features accurately discriminate radiologist-determined visual UIP? STUDY DESIGN AND METHODS: A retrospective cohort study was performed. Chest CT scans acquired in individuals with and without ILD were drawn from a variety of public and private data sources. Using radiologist-determined visual UIP as ground truth, a convolutional neural network was used to learn discrete CT scan features of UIP, with outputs used to predict the likelihood of UIP using a linear support vector machine. Test performance characteristics were assessed in an independent performance cohort and multicenter ILD clinical cohort. Transplant-free survival was compared between UIP classification approaches using the Kaplan-Meier estimator and Cox proportional hazards regression. RESULTS: A total of 2,907 chest CT scans were included in the training (n = 1,934), validation (n = 408), and performance (n = 565) data sets. The prevalence of radiologist-determined visual UIP was 12.4% and 37.1% in the performance and ILD clinical cohorts, respectively. The DL-based UIP classifier predicted visual UIP in the performance cohort with sensitivity and specificity of 93% and 86%, respectively, and in the multicenter ILD clinical cohort with 81% and 77%, respectively. DL-based and visual UIP classification similarly discriminated survival, and outcomes were consistent among cases with positive DL-based UIP classification irrespective of visual classification. INTERPRETATION: A DL-based classifier for UIP demonstrated good test performance across a wide range of UIP prevalence and similarly discriminated survival when compared with radiologist-determined UIP. This automated tool could efficiently screen for UIP in patients undergoing chest CT scan and identify a high-risk phenotype among those with known ILD.

Assessing the role of an artificial intelligence assessment tool for thoracic aorta diameter on routine chest CT.

OBJECTIVE: To assess the diagnostic accuracy and clinical impact of automated artificial intelligence (AI) measurement of thoracic aorta diameter on routine chest CT. METHODS: A single-centre retrospective study involving three cohorts. 210 consecutive ECG-gated CT aorta scans (mean age 75 ± 13) underwent automated analysis (AI-Rad Companion Chest CT, Siemens) and were compared to a reference standard of specialist cardiothoracic radiologists for accuracy measuring aortic diameter. A repeated measures analysis tested reporting consistency in a second cohort (29 patients, mean age 61 ± 17) of immediate sequential pre-contrast and contrast CT aorta acquisitions. Potential clinical impact was assessed in a third cohort of 197 routine CT chests (mean age 66 ± 15) to document potential clinical impact. RESULTS: AI analysis produced a full report in 387/436 (89%) and a partial report in 421/436 (97%). Manual vs AI agreement was good to excellent (ICC 0.76-0.92). Repeated measures analysis of expert and AI reports for the ascending aorta were moderate to good (ICC 0.57-0.88). AI diagnostic performance crossed the threshold for maximally accepted limits of agreement (>5 mm) at the aortic root on ECG-gated CTs. AI newly identified aortic dilatation in 27% of patients on routine thoracic imaging with a specificity of 99% and sensitivity of 77%. CONCLUSION: AI has good agreement with expert readers at the mid-ascending aorta and has high specificity, but low sensitivity, at detecting dilated aortas on non-dedicated chest CTs. ADVANCES IN KNOWLEDGE: An AI tool may improve the detection of previously unknown thoracic aorta dilatation on chest CTs vs current routine reporting.

Multimodality imaging of Abernethy malformation.

Abernethy malformation, or congenital extrahepatic portosystemic venous shunt, is a rare anomaly involving the portal venous system. Despite its rarity, it is increasingly being reported, and therefore, it is important to diagnose given the potential adverse clinical consequences if left untreated. It has a spectrum of presentations, ranging from complete lack of symptoms, to causing hepatic carcinoma, hepatic encephalopathy, severe pulmonary hypertension, and diffuse pulmonary arteriovenous malformation. We herein describe the case and echocardiographic, computed tomography, and magnetic resonance imaging findings of a transgender individual, with this anomaly detected incidentally during adulthood.

Contrast circulation in adult fontan patients using MR Time Resolved Angiography: Application for CT pulmonary angiography.

BACKGROUND: When patients with Fontan circulation require a computed tomographic pulmonary angiogram (CTPA), there are significant challenges in achieving adequate contrast opacification due to the altered anatomical connections. This study used Time Resolved Angiography with Interleaved Stochastic Trajectories (TWIST) Magnetic Resonance Angiography (MRA) to examine contrast circulation in a cohort of patients with Fontan circulation who were having routine MRI follow up to inform the contrast timing of any subsequent CT. METHODS: This is a single centre, cross-sectional, observational, retrospective study. The time to peak (TTP) signal intensity from the MRA was recorded using regions of interest on the aorta, pulmonary arteries, cavae and Fontan conduit. Patients were grouped by ejection fraction, global longitudinal strain, indexed stroke volume and cardiac index to examine if these cardiac performance parameters affected the mean TTP. Statistical analysis was performed to find the mean TTP for each of the vessels, which was consequently compared between the different cardiac performance parameters. RESULTS: 35 patients were included in the study. Mean TTP contrast enhancement was 31s in the thoracic aorta, 46s in the right pulmonary artery, 41s in the left pulmonary artery and 55s in the Fontan conduit. Cardiac performance shows no statistically significant relationship to the peak contrast enhancement whether measured by ejection fraction, global longitudinal strain, stroke volume index or cardiac index. CONCLUSION: The mean optimal timing for a single-phase examination of the Fontan circulation, following an upper limb injection, was 55 s following start of contrast injection irrespective of cardiac performance. In TWIST MRA, the IV bolus is 4-5 s duration. A longer bolus is required for CTA, around 20s, suggesting an additional delay will be required. We propose that an optimal single phase CTPA to be protocolled at 70 s following the start of contrast injection, assuming adequate iodinated contrast dose.

Repaired coarctation of the aorta, persistent arterial hypertension and the selfish brain.

BACKGROUND: It has been estimated that 20-30% of repaired aortic coarctation (CoA) patients develop hypertension, with significant cardiovascular morbidity and mortality. Vertebral artery hypoplasia (VAH) with an incomplete posterior circle of Willis (ipCoW; VAH + ipCoW) is associated with increased cerebrovascular resistance before the onset of increased sympathetic nerve activity in borderline hypertensive humans, suggesting brainstem hypoperfusion may evoke hypertension to maintain cerebral blood flow: the "selfish brain" hypothesis. We now assess the "selfish brain" in hypertension post-CoA repair. METHODS: Time-of-flight cardiovascular magnetic resonance angiography from 127 repaired CoA patients (34 ± 14 years, 61% male, systolic blood pressure (SBP) 138 ± 19 mmHg, diastolic blood pressure (DBP) 76 ± 11 mmHg) was compared with 33 normotensive controls (42 ± 14 years, 48% male, SBP 124 ± 10 mmHg, DBP 76 ± 8 mmHg). VAH was defined as < 2 mm and ipCoW as hypoplasia of one or both posterior communicating arteries. RESULTS: VAH + ipCoW was more prevalent in repaired CoA than controls (odds ratio: 5.8 [1.6-20.8], p = 0.007), after controlling for age, sex and body mass index (BMI). VAH + ipCoW was an independent predictor of hypertension (odds ratio: 2.5 [1.2-5.2], p = 0.017), after controlling for age, gender and BMI. Repaired CoA subjects with VAH + ipCoW were more likely to have difficult to treat hypertension (odds ratio: 3.3 [1.01-10.7], p = 0.049). Neither age at time of CoA repair nor any specific repair type were significant predictors of VAH + ipCoW in univariate regression analysis. CONCLUSIONS: VAH + ipCoW predicts arterial hypertension and difficult to treat hypertension in repaired CoA. It is unrelated to age at time of repair or repair type. CoA appears to be a marker of wider congenital cerebrovascular problems. Understanding the "selfish brain" in post-CoA repair may help guide management. JOURNAL SUBJECT CODES: High Blood Pressure; Hypertension; Magnetic Resonance Imaging (MRI); Cardiovascular Surgery; Cerebrovascular Malformations.

Cardiovascular Manifestations and Complications of Loeys-Dietz Syndrome: CT and MR Imaging Findings.

Loeys-Dietz syndrome (LDS) is a recently described genetic connective tissue disorder with a wide spectrum of multisystem involvement. LDS is characterized by rapidly progressive aortic and peripheral arterial aneurysmal disease. LDS and the other inherited aortopathies such as Marfan syndrome have overlapping phenotypic features. However, LDS is characterized by a more aggressive vascular course; patient morbidity and mortality occur at an early age, with complications developing at relatively smaller aortic dimensions. In addition, there is more diffuse arterial involvement in LDS, with a large proportion of patients developing aneurysms of the iliac, mesenteric, and intracranial arteries. Early diagnosis and careful follow-up are essential for ensuring timely intervention in patients with arterial disease. Cross-sectional angiography has an important role in the baseline assessment, follow-up, and evaluation of acute complications of LDS, the thresholds and considerations of which differ from those of other inherited aortopathies. In this article, LDS is compared with other genetic vascular connective tissue disorders. In addition, the genetic, histopathologic, and cardiovascular manifestations of this disease process are reviewed, with a focus on computed tomographic and magnetic resonance imaging findings. Online DICOM image stacks and supplemental material are available for this article. ©RSNA, 2018.

Impact of Pretreatment Interstitial Lung Disease on Radiation Pneumonitis and Survival in Patients Treated With Lung Stereotactic Body Radiation Therapy (SBRT).

INTRODUCTION: The purpose of this study was to determine the impact of interstitial lung disease (ILD) on radiation pneumonitis (RP) and overall survival (OS) in lung stereotactic body radiation therapy (SBRT). METHODS: Patients treated with lung SBRT from 2004 to 2015 were included. Pretreatment computed tomography scans were reviewed and classified for interstitial changes by thoracic radiologists using American Thoracic Society guidelines and Washko and Kazerooni scores. RP was scored prospectively using Common Terminology Criteria for Adverse Events, version 3.0. Pretreatment imaging characteristics, clinical variables, and dosimetry were assessed by univariate (UVA) and multivariate analysis (MVA). OS was assessed by the log-rank test, and the impact of ILD on OS was assessed by Cox regression. RESULTS: Of the 537 patients assessed, 39 had interstitial changes (13 usual interstitial pneumonia [UIP], 24 possible UIP, and 2 inconsistent with UIP). RP was significantly higher in patients with ILD than in patients without ILD (grade ≥ 2, 20.5% vs. 5.8%; P < .01; grade ≥ 3, 10.3% vs. 1.0%; P < .01). Two of 3 grade 5 RP had imaging features of ILD. On UVA, ILD, Washko score, lung parameters performance status, and dose were significant predictors of grade ≥ 2 RP. On MVA, ILD (odds ratio, 5.81; 95% confidence interval, 2.28-14.83; P < .01) and mean lung dose (odds ratio, 1.40; 95% confidence interval, 1.14-1.71; P < .01) were predictors of RP. ILD did not significantly affect OS on UVA or MVA. Median survival was 27.4 months in the ILD cohort and 34.8 in the ILD-negative cohort (P = .17). DISCUSSION: ILD is a significant risk factor for RP in patients treated with lung SBRT. Computed tomography scans should be reviewed for evidence of ILD prior to SBRT.

ECG strain pattern in hypertension is associated with myocardial cellular expansion and diffuse interstitial fibrosis: a multi-parametric cardiac magnetic resonance study.

AIMS: In hypertension, the presence of left ventricular (LV) strain pattern on 12-lead electrocardiogram (ECG) carries adverse cardiovascular prognosis. The underlying mechanisms are poorly understood. We investigated whether hypertensive ECG strain is associated with myocardial interstitial fibrosis and impaired myocardial strain, assessed by multi-parametric cardiac magnetic resonance (CMR). METHODS AND RESULTS: A total of 100 hypertensive patients [50 ± 14 years, male: 58%, office systolic blood pressure (SBP): 170 ± 30 mmHg, office diastolic blood pressure (DBP): 97 ± 14 mmHg) underwent ECG and 1.5T CMR and were compared with 25 normotensive controls (46 ± 14 years, 60% male, SBP: 124 ± 8 mmHg, DBP: 76 ± 7 mmHg). Native T1 and extracellular volume fraction (ECV) were calculated with the modified look-locker inversion-recovery sequence. Myocardial strain values were estimated with voxel-tracking software. ECG strain (n = 20) was associated with significantly higher indexed LV mass (LVM) (119 ± 32 vs. 80 ± 17 g/m2, P < 0.05) and ECV (30 ± 4 vs. 27 ± 3%, P < 0.05) compared with hypertensive subjects without ECG strain (n = 80). ECG strain subjects had significantly impaired circumferential strain compared with hypertensive subjects without ECG strain and controls (-15.2 ± 4.7 vs. -17.0 ± 3.3 vs. -17.3 ± 2.4%, P < 0.05, respectively). In subgroup analysis, comparing ECG strain subjects to hypertensive subjects with elevated LVM but no ECG strain, a significantly higher ECV (30 ± 4 vs. 28 ± 3%, P < 0.05) was still observed. Indexed LVM was the only variable independently associated with ECG strain in multivariate logistic regression analysis [odds ratio (95th confidence interval): 1.07 (1.02-1.12), P < 0.05). CONCLUSION: In hypertension, ECG strain is a marker of advanced LVH associated with increased interstitial fibrosis and associated with significant myocardial circumferential strain impairment.

Protective Capnothorax During Transthoracic Needle Biopsy.

PURPOSE: Transthoracic needle biopsy (TTNB) is an established procedure in the management of pulmonary nodules. The most common complications are directly related to crossing the lung or visceral pleura during the biopsy. In this study, we describe the use of carbon dioxide instead of room air to create a protective "capnothorax" during TTNB. MATERIALS AND METHODS: Five patients underwent creation of a capnothorax during TTNB. Parameters recorded were location and size of target, distance from pleura, length of procedure, volume of carbon dioxide, periprocedural complications and biopsy result. RESULTS: Induction of capnothorax was successful in all cases. In two patients, a continuous infusion of carbon dioxide was required to maintain an adequate volume of intrapleural gas. In two patients, the carbon dioxide resolved spontaneously and in the remaining patients it was aspirated at the end of the procedure. All biopsies were diagnostic with no periprocedural or postprocedural complications. CONCLUSION: This study suggests that protective iatrogenic capnothorax is a safe and effective technique during TTNB. The intrinsic properties and availability of carbon dioxide make it an attractive alternative to room air.

Mitral Valve Diverticulum Presenting with Severe Mitral Stenosis: Case Report.

The case is presented of a non-infectious anterior mitral valve leaflet diverticulum, which appeared as symptomatic mitral stenosis. Unlike previous reports, there was no histological myxomatous degeneration of the valve. To the authors' knowledge, this is the first time a mitral valve diverticulum resulting in severe mitral stenosis has been reported in the literature.

Comprehensive characterisation of hypertensive heart disease left ventricular phenotypes.

OBJECTIVE: Myocardial intracellular/extracellular structure and aortic function were assessed among hypertensive left ventricular (LV) phenotypes using cardiovascular magnetic resonance (CMR). METHODS: An observational study from consecutive tertiary hypertension clinic patients referred for CMR (1.5 T) was performed. Four LV phenotypes were defined: (1) normal with normal indexed LV mass (LVM) and LVM to volume ratio (M/V), (2) concentric remodelling with normal LVM but elevated M/V, (3) concentric LV hypertrophy (LVH) with elevated LVM but normal indexed end-diastolic volume (EDV) or (4) eccentric LVH with elevated LVM and EDV. Extracellular volume fraction was measured using T1-mapping. Circumferential strain was calculated by voxel-tracking. Aortic distensibility was derived from high-resolution aortic cines and contemporaneous blood pressure measurements. RESULTS: 88 hypertensive patients (49±14 years, 57% men, systolic blood pressure (SBP): 167±30 mm Hg, diastolic blood pressure (DBP): 96±14 mm Hg) were compared with 29 age-matched/sex-matched controls (47±14 years, 59% men, SBP: 128±12 mm Hg, DBP: 79±10 mm Hg). LVH resulted from increased myocardial cell volume (eccentric LVH: 78±19 mL/m(2) vs concentric LVH: 73±15 mL/m(2) vs concentric remodelling: 55±9 mL/m(2), p<0.05, respectively) and interstitial fibrosis (eccentric LVH: 33±10 mL/m(2) vs concentric LVH: 30±10 mL/m(2) vs concentricremodelling: 19±2 mL/m(2), p<0.05, respectively). LVH had worst circumferential impairment (eccentric LVH: -12.8±4.6% vs concentric LVH: -15.5±3.1% vs concentric remodelling: -17.1±3.2%, p<0.05, respectively). Concentric remodelling was associated with reduced aortic distensibility, but not with large intracellular/interstitial expansion or myocardial dysfunction versus controls. CONCLUSIONS: Myocardial interstitial fibrosis varies across hypertensive LV phenotypes with functional consequences. Eccentric LVH has the most fibrosis and systolic impairment. Concentric remodelling is only associated with abnormal aortic function. Understanding these differences may help tailor future antihypertensive treatments.

Endovascular closure of thoracic aortic pseudoaneurysms: A combined device occlusion and coil embolization technique in patients unsuitable for surgery or stenting.

OBJECTIVES: Our aim was to retrospectively evaluate non-stent graft closure of ascending aortic pseudoaneurysms at our center over a 10-year period, and describe a combined device occlusion and coil embolization technique. BACKGROUND: Aortic pseudoaneurysms (APAs) are a rare complication post cardiothoracic surgery, but can have fatal complications. There is increasing use of percutaneous interventional techniques for occlusion of aortic pseudoaneurysms in patients who are considered unsuitable for surgery. Stent graft deployment may not be possible depending on the specific anatomy and pathology. METHODS AND RESULTS: Retrospective evaluation of the catheter laboratory database was performed at our center and anonymized data was obtained for patients who had nonstent endovascular treatment of APAs. Twelve patients were identified with a mean age of 63 ± 16 years. Seven patients had the combined occlusion and embolization technique, only 1/7 (14.3%) died from complications related to APAs. Five patients had occlusion device only, 3/5 (60%) died of complications related to their APA. The mean survival for the patients who had a combination procedure was 33.2±.22.6 months (range, 1 - 60 months), compared to 2.7 ± 2.6 months with device closure only (note 2 patients had short follow up of <3 months). CONCLUSIONS: We evaluate non-stent graft percutaneous closure of APAs in a high-risk patient group and provide data on the use of a novel combined occlusion device and coil embolization technique. We feel this is a viable approach to APA closure in this population but this will require larger clinical studies in the future. © 2016 Wiley Periodicals, Inc.

Extra-cardiac findings in cardiovascular magnetic resonance: what the imaging cardiologist needs to know.

Cardiovascular magnetic resonance (CMR) is an established non-invasive technique to comprehensively assess cardiovascular structure and function in a variety of acquired and inherited cardiac conditions. A significant amount of the neck, thorax and upper abdomen are imaged at the time of routine clinical CMR, particularly in the initial multi-slice axial and coronal images. The discovery of unsuspected disease at the time of imaging has ethical, financial and medico-legal implications. Extra-cardiac findings at the time of CMR are common, can be important and can change clinical management. Certain patient groups undergoing CMR are at particular risk of important extra-cardiac findings as several of the cardiovascular risk factors for atherosclerosis are also risk factors for malignancy. Furthermore, the presence of certain extra-cardiac findings may contribute to the interpretation of the primary cardiac pathology as some cardiac conditions have multi-systemic extra-cardiac involvement. The aim of this review is to give an overview of the type of extra-cardiac findings that may become apparent on CMR, subdivided by anatomical location. We focus on normal variant anatomy that may mimic disease, common incidental extra-cardiac findings and important imaging signs that help distinguish sinister pathology from benign disease. We also aim to provide a framework to the approach and potential further diagnostic work-up of incidental extra-cardiac findings discovered at the time of CMR. However, it is beyond the scope of this review to discuss and determine the clinical significance of extracardiac findings at CMR.

Prevalence and predictors of asymmetric hypertensive heart disease: insights from cardiac and aortic function with cardiovascular magnetic resonance.

AIMS: We sought to determine the prevalence of asymmetric hypertensive heart disease (HHD) overlapping morphologically with hypertrophic cardiomyopathy (HCM) and to determine predictors of this pattern of hypertensive remodelling. METHODS AND RESULTS: One hundred and fifty hypertensive patients underwent 1.5 T cardiovascular magnetic resonance imaging. Twenty-one patients were excluded due to concomitant cardiac pathology that may confound the hypertrophic response, e.g. myocardial infarction, moderate-severe valvular disease, or other cardiomyopathy. Asymmetric HHD was defined as a segmental wall thickness of ≥15 mm and >1.5-fold the opposing wall in ≥1 myocardial segments, measured from short-axis cine stack at end-diastole. Ambulatory blood pressure, myocardial replacement fibrosis, aortic distensibility and aortoseptal angle were investigated as predictors of asymmetric HHD by multivariate logistic regression. Out of 129 hypertensive subjects (age: 51 ± 15 years, 50% male, systolic blood pressure: 170 ± 30 mmHg, diastolic blood pressure: 97 ± 16 mmHg), asymmetric HHD occurred in 21%. Where present, maximal end-diastolic wall thickness (EDWT) was 17.8 ± 1.9 mm and located exclusively in the basal or mid septum. In asymmetric HHD, aortoseptal angle (114 ± 10° vs. 125 ± 9° vs. 123 ± 12°, P < 0.05, respectively) was significantly reduced compared to concentric left ventricular hypertrophy (LVH) and compared to no LVH, respectively. Aortic distensibility in asymmetric HHD (1.01 ± 0.60 vs. 1.83 ± 1.65 mm2/mmHg × 103, P < 0.05, respectively) was significantly reduced compared to subjects with no LVH. Age (odds ratio [95th confidence interval]: 1.10 [1.02-1.18], P < 0.05) and indexed LV mass (1.09 [0.98-1.28], P < 0.0001) were significant, independent predictors of asymmetric HDD. CONCLUSIONS: Asymmetric HHD morphologically overlapping with HCM, according to the current ESC guidelines, is common. Postulating a diagnosis of HCM on the basis of EDWT of ≥15 mm should be made with caution in the presence of arterial hypertension particular in male subjects with elevated LV mass.

A Novel Cause of Acute Coronary Syndrome Due to Dynamic Extrinsic Coronary Artery Compression by a Rib Exostosis: Multimodality Imaging Diagnosis.

We report a case of acute coronary syndrome secondary to intermittent extrinsic compression of the left anterior descending coronary artery by inward-pointing rib exostosis in an 18-year-old woman during forceful repeated expiration in labour. The diagnosis was achieved using multimodality noninvasive cardiac imaging. In particular, we demonstrated the novel role of expiratory-phase cardiac computed tomography in confirming the anatomical relationship of the bony exostosis to the left anterior descending coronary artery. The case reminds us the heart and mediastinum move dynamically, relative to the bony thorax, throughout the respiratory cycle, and that changes in cardiac physiology in pregnancy may become pathological.

Optimising the imaging plane for right ventricular magnetic resonance volume analysis in adult patients referred for assessment of right ventricular structure and function.

INTRODUCTION: Our aim was to evaluate the reproducibility and accuracy of using short-axis and axial (transaxial) plane for magnetic resonance imaging analysis in adult patients referred for assessment of right ventricular (RV) structure and function. METHODS: Twenty consecutive subjects (10 male, 10 female, mean age 32.2 ± 14.8 years) who were referred for RV assessment and had cardiac magnetic resonance imaging were retrospectively selected. Axial and short-axis manual contouring was performed using cine steady-state free precession sequences by three experienced imaging specialists. The reproducibility of end diastolic volumes, end systolic volumes and ejection fraction was assessed with intraclass correlation coefficients (ICCs) and paired t-tests. Left ventricular stroke volume (LVSV) and RV stroke volumes (RVSV) were compared with concordance correlation coefficients (CCCs) and t-tests to determine accuracy. RESULTS: The concordance between the RVSV and LVSV was good using both methods (axial RVSV CCC = 0.93, short-axis RVSV CCC = 0.86). Paired t-test and analysis of variance showed that the LV/RV stroke volume differences were not significant (p = 0.17). There was slight improvement in interobserver reliability with end systolic volume measurements (axial ICC = 0.92, short-axis ICC = 0.81) but this failed to reach statistical significance (p = 0.37). There was excellent intraobserver variability (ICC > 0.9). CONCLUSION: This study shows that there is no statistically significant difference in reproducibility or accuracy using the short-axis or axial orientations in RV volume analysis in adult patients being referred for RV assessment.