Establishing minimally important differences for cardiac MRI end-points in pulmonary arterial hypertension.

BACKGROUND: Cardiac magnetic resonance (CMR) is the gold standard technique to assess biventricular volumes and function, and is increasingly being considered as an end-point in clinical studies. Currently, with the exception of right ventricular (RV) stroke volume and RV end-diastolic volume, there is only limited data on minimally important differences (MIDs) reported for CMR metrics. Our study aimed to identify MIDs for CMR metrics based on US Food and Drug Administration recommendations for a clinical outcome measure that should reflect how a patient "feels, functions or survives". METHODS: Consecutive treatment-naïve patients with pulmonary arterial hypertension (PAH) between 2010 and 2022 who had two CMR scans (at baseline prior to treatment and 12 months following treatment) were identified from the ASPIRE registry. All patients were followed up for 1 additional year after the second scan. For both scans, cardiac measurements were obtained from a validated fully automated segmentation tool. The MID in CMR metrics was determined using two distribution-based (0.5sd and minimal detectable change) and two anchor-based (change difference and generalised linear model regression) methods benchmarked to how a patient "feels" (emPHasis-10 quality of life questionnaire), "functions" (incremental shuttle walk test) or "survives" for 1-year mortality to changes in CMR measurements. RESULTS: 254 patients with PAH were included (mean±sd age 53±16 years, 79% female and 66% categorised as intermediate risk based on the 2022 European Society of Cardiology/European Respiratory Society risk score). We identified a 5% absolute increase in RV ejection fraction and a 17 mL decrease in RV end-diastolic or end-systolic volumes as the MIDs for improvement. Conversely, a 5% decrease in RV ejection fraction and a 10 mL increase in RV volumes were associated with worsening. CONCLUSIONS: This study establishes clinically relevant CMR MIDs for how a patient "feels, functions or survives" in response to PAH treatment. These findings provide further support for the use of CMR as a clinically relevant clinical outcome measure and will aid trial size calculations for studies using CMR.

Plain language summaryPulmonary arterial hypertension (PAH) is a disease of the vessels of the lung that causes their narrowing and stiffening. As a result, the heart pumping blood into these diseased lung vessels has to work harder and eventually gets worn out. PAH can affect patients' ability to function in daily activities and impact their quality of life. It also reduces their life expectancy dramatically. Patients are, therefore, often monitored and undergo several investigations to adapt treatment according to their situation. These investigations include a survey of how a patient feels (the emPHasis-10 questionnaire), functions (walking test) and how well the heart is coping with the disease (MRI of the heart). Until now, it is unclear how changes on MRI of the heart reflect changes in how a patient feels and functions. Our study identified patients that had the emPHasis-10 questionnaire, walking test and MRI of the heart at both the time of PAH diagnosis and one year later. This allowed us to compare how the changes in the different tests relate to each other. And because previous research identified thresholds for important changes in the emPHasis-10 questionnaire and the walking tests, we were able to use these tests as a benchmark for changes in the MRI of the heart. Our study identified thresholds for change on heart MRI that might indicate whether a patient has improved or worsened. This finding might have implications for how patients are monitored in clinical practice and future research on PAH treatments.

Identification of Cardiac Magnetic Resonance Imaging Thresholds for Risk Stratification in Pulmonary Arterial Hypertension.

Rationale: Pulmonary arterial hypertension (PAH) is a life-shortening condition. The European Society of Cardiology and European Respiratory Society and the REVEAL (North American Registry to Evaluate Early and Long-Term PAH Disease Management) risk score calculator (REVEAL 2.0) identify thresholds to predict 1-year mortality.Objectives: This study evaluates whether cardiac magnetic resonance imaging (MRI) thresholds can be identified and used to aid risk stratification and facilitate decision-making.Methods: Consecutive patients with PAH (n = 438) undergoing cardiac MRI were identified from the ASPIRE (Assessing the Spectrum of Pulmonary Hypertension Identified at a Referral Center) MRI database. Thresholds were identified from a discovery cohort and evaluated in a test cohort.Measurements and Main Results: A percentage-predicted right ventricular end-systolic volume index threshold of 227% or a left ventricular end-diastolic volume index of 58 ml/m2 identified patients at low (<5%) and high (>10%) risk of 1-year mortality. These metrics respectively identified 63% and 34% of patients as low risk. Right ventricular ejection fraction >54%, 37-54%, and <37% identified 21%, 43%, and 36% of patients at low, intermediate, and high risk, respectively, of 1-year mortality. At follow-up cardiac MRI, patients who improved to or were maintained in a low-risk group had a 1-year mortality <5%. Percentage-predicted right ventricular end-systolic volume index independently predicted outcome and, when used in conjunction with the REVEAL 2.0 risk score calculator or a modified French Pulmonary Hypertension Registry approach, improved risk stratification for 1-year mortality.Conclusions: Cardiac MRI can be used to risk stratify patients with PAH using a threshold approach. Percentage-predicted right ventricular end-systolic volume index can identify a high percentage of patients at low-risk of 1-year mortality and, when used in conjunction with current risk stratification approaches, can improve risk stratification. This study supports further evaluation of cardiac MRI in risk stratification in PAH.

Avoiding skull radiographs in infants with suspected inflicted injury who also undergo head CT: "a no-brainer?"

OBJECTIVES: To assess whether head CT with 3D reconstruction can replace skull radiographs (SXR) in the imaging investigation of suspected physical abuse (SPA)/abusive head trauma (AHT). METHODS: PACS was interrogated for antemortem skeletal surveys performed for SPA, patients younger than 2 years, SXR and CT performed within 4 days of each other. Paired SXR and CT were independently reviewed. One reviewer analysed CT without and (3 months later) with 3D reconstructions. SXR and CT expert consensus review formed the gold standard. Observer reliability was calculated. RESULTS: A total of 104 SXR/CT examination pairs were identified, mean age 6.75 months (range 4 days to 2 years); 21 (20%) had skull fractures; two fractures on CT were missed on SXR. There were no fractures on SXR that were not seen on CT. For SXR and CT, respectively: PPV reviewer 1, 95% confidence interval (CI) 48-82% and 85-100%; reviewer 2, 67-98% and 82-100%; and NPV reviewer 1, 95%, CI 88-98% and 96-100%; reviewer 2, 88-97% and 88-98%. Inter- and intra-observer reliability were respectively the following: SXR, excellent (kappa = 0.831) and good (kappa = 0.694); CT, excellent (kappa = 0.831) and perfect (kappa = 1). All results were statistically significant (p < 0.001). CONCLUSIONS: CT has greater diagnostic accuracy than SXR in detecting skull fractures which is increased on concurrent review of 3D reconstructions and should be performed in every case of SPA/AHT. SXR does not add further diagnostic information and can be omitted from the skeletal survey when CT with 3D reconstruction is going to be, or has been, performed. KEY POINTS: • Head CT with 3D reconstruction is more sensitive and specific for the diagnosis of skull fractures. • Skull radiographs can be safely omitted from the initial skeletal survey performed for suspected physical abuse when head CT with 3D reconstruction is going to be, or has been, performed.

Airway Microstructure in Idiopathic Pulmonary Fibrosis: Assessment at Hyperpolarized 3He Diffusion-weighted MRI.

Background MRI with inhaled hyperpolarized helium 3 (3He) allows for functional and structural imaging of the lungs. Hyperpolarized gas diffusion-weighted (DW) MRI provides noninvasive and quantitative assessment of microstructural acinar changes in the lungs. Purpose To investigate whether microstructural imaging metrics from in-vivo hyperpolarized 3He DW MRI are sensitive to longitudinal changes in a cohort of participants with idiopathic pulmonary fibrosis (IPF) and to evaluate the reproducibility of these metrics and their correlation with existing clinical measures of IPF disease severity. Materials and Methods In this prospective study, 18 participants with IPF underwent 3He DW MRI at 1.5 T and 11 participants underwent an identical same-day examination for reproducibility assessment. Thirteen participants returned for 6- and 12-month follow-up examinations. Pulmonary function tests, including diffusing capacity of the lungs for carbon monoxide and forced vital capacity, were performed at each examination. The apparent diffusion coefficient (ADC) and stretched exponential model-derived mean diffusive length scale (LmD) from DW MRI was compared with baseline CT fibrosis scores and pulmonary function tests by using Spearman rank correlation coefficient. Longitudinal changes in DW MRI and pulmonary function test measurements were assessed with Friedman tests and post hoc Dunn test. Results 3He ADC and LmD were reproducible (mean Bland-Altman analysis bias, 0.002 cm2 · sec-1 and -1.5 µm, respectively). Elevated ADC and LmD regions qualitatively corresponded to fibrotic regions at CT. ADC and LmD correlated with diffusing capacity of the lungs for carbon monoxide (respectively: r = -0.56, P = .017; and r = -0.54, P = .02) and CT fibrosis score (respectively: r = 0.71, P = .001; and r = 0.65, P = .003). LmD increased by 12 µm after 12 months (P = .001) whereas mean ADC (P = .17), forced vital capacity (P = .12), and diffusing capacity of the lungs for carbon monoxide (P > .99) were not statistically different between examinations. Conclusion Helium 3 diffusion-weighted MRI-derived mean diffusive length scale demonstrates longitudinal changes in lungs affected by idiopathic pulmonary fibrosis. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Altes and Flors in this issue.

Diagnosis of Pulmonary Hypertension with Cardiac MRI: Derivation and Validation of Regression Models.

Purpose To derive and test multiparametric cardiac MRI models for the diagnosis of pulmonary hypertension (PH). Materials and Methods Images and patient data from consecutive patients suspected of having PH who underwent cardiac MRI and right-sided heart catheterization (RHC) between 2012 and 2016 were retrospectively reviewed. Of 2437 MR images identified, 603 fit the inclusion criteria. The mean patient age was 61 years (range, 18-88 years; mean age of women, 60 years [range, 18-84 years]; mean age of men, 62 years [range, 22-88 years]). In the first 300 patients (derivation cohort), cardiac MRI metrics that showed correlation with mean pulmonary arterial pressure (mPAP) were used to create a regression algorithm. The performance of the model was assessed in the 303-patient validation cohort by using receiver operating characteristic (ROC) and χ2 analysis. Results In the derivation cohort, cardiac MRI mPAP model 1 (right ventricle and black blood) was defined as follows: -179 + loge interventricular septal angle × 42.7 + log10 ventricular mass index (right ventricular mass/left ventricular mass) × 7.57 + black blood slow flow score × 3.39. In the validation cohort, cardiac MRI mPAP model 1 had strong agreement with RHC-measured mPAP, an intraclass coefficient of 0.78, and high diagnostic accuracy (area under the ROC curve = 0.95; 95% confidence interval [CI]: 0.93, 0.98). The threshold of at least 25 mm Hg had a sensitivity of 93% (95% CI: 89%, 96%), specificity of 79% (95% CI: 65%, 89%), positive predictive value of 96% (95% CI: 93%, 98%), and negative predictive value of 67% (95% CI: 53%, 78%) in the validation cohort. A second model, cardiac MRI mPAP model 2 (right ventricle pulmonary artery), which excludes the black blood flow score, had equivalent diagnostic accuracy (ROC difference: P = .24). Conclusion Multiparametric cardiac MRI models have high diagnostic accuracy in patients suspected of having pulmonary hypertension. Published under a CC BY 4.0 license. Online supplemental material is available for this article. See also the editorial by Colletti in this issue.

Identifying At-Risk Patients with Combined Pre- and Postcapillary Pulmonary Hypertension Using Interventricular Septal Angle at Cardiac MRI.

Purpose To assess interventricular septal (IVS) angle in the identification of combined pre- and postcapillary pulmonary hypertension (Cpc-PH) in patients with pulmonary hypertension (PH) due to left-sided heart disease. Materials and Methods In this retrospective study, consecutive, incident patients suspected of having PH underwent same-day right-sided heart catheterization (RHC) and MRI at a PH referral center between April 2012 and April 2017. The diagnostic accuracy of the IVS angle to identify Cpc-PH in patients with pulmonary arterial wedge pressure (PAWP) greater than 15 mmHg was assessed by using receiver operator characteristic curves, sensitivity, specificity, and negative and positive predictive values. IVS angle also was assessed as a predictor of all-cause mortality by using Cox uni- and multivariable proportional hazards regression. Results A total of 708 patients underwent same-day MRI and RHC, and 171 patients had PAWP greater than 15 mmHg. Mean age was 70 years (range, 21-90 years) (women: mean age, 69 years; range, 21-88 years) (men: mean age, 71 years; range, 43-90 years). Systolic IVS angle correlated with diastolic pulmonary gradient (DPG) (r = 0.739, P < .001). Receiver operating characteristic curve analysis showed septal angle enabled identification of Cpc-PH (DPG ≥ 7), with an area under the receiver operating characteristic curve of 0.911 (P < .001). A 160° threshold, derived from the first half of patients with raised PAWP, enabled identification of a DPG of at least 7 mmHg with 67% sensitivity and 93% specificity (P < .001) in the second cohort of patients with raised PAWP. IVS angle was predictive of all-cause mortality (standardized univariable hazard ratio, 1.615; P < .01). Conclusion The systolic interventricular septal angle is elevated in patients with combined pre- and postcapillary pulmonary hypertension and enables one to predict those patients who have PH due to left-sided heart disease who have an increased risk of death. Published under a CC BY 4.0 license. Online supplemental material is available for this article.

The impact of patient choice on survival in chronic thromboembolic pulmonary hypertension.

Pulmonary endarterectomy (PEA) is the gold standard treatment for operable chronic thromboembolic pulmonary hypertension (CTEPH). However, a proportion of patients with operable disease decline surgery. There are currently no published data on this patient group. The aim of this study was to identify outcomes and prognostic factors in a large cohort of consecutive patients with CTEPH.Data were collected for consecutive, treatment-naive CTEPH patients at the Pulmonary Vascular Disease Unit of the Royal Hallamshire Hospital (Sheffield, UK) between 2001 and 2014.Of 550 CTEPH patients (mean±sd age 63±15 years, follow-up 4±3 years), 49% underwent surgery, 32% had technically operable disease and did not undergo surgery (including patient choice n=72 and unfit for surgery n=63), and 19% had inoperable disease due to disease distribution. The 5-year survival was superior in patients undergoing PEA (83%) versus technically operable disease who did not undergo surgery (53%) and inoperable due to disease distribution (59%) (p<0.001). Survival was superior in patients following PEA compared with those offered but declining surgery (55%) (p<0.001). In patients offered PEA, independent prognostic factors included mixed venous oxygen saturation, gas transfer and patient decision to proceed to surgery.Outcomes in CTEPH following PEA are excellent and superior to patients declining surgery, and strongly favour consideration of a surgical intervention in eligible patients.

Novel imaging techniques in pulmonary hypertension.

PURPOSE OF REVIEW: Pulmonary hypertension is a life-shortening condition, which may be idiopathic but is more frequently seen in association with other conditions. Current guidelines recommend cardiac catheterization to confirm the diagnosis of pulmonary hypertension. Evidence suggests an increasing role for noninvasive imaging modalities in the initial diagnostic and prognostic assessment and evaluation of treatment response. RECENT FINDINGS: In this review we examine the evidence for current noninvasive imaging methodologies: echocardiography computed tomography and MRI in the diagnostic and prognostic assessment of suspected pulmonary hypertension and explore the potential utility of modeling and machine-learning approaches. SUMMARY: Noninvasive imaging allows a comprehensive assessment of patients with suspected pulmonary hypertension. It plays a key part in the initial diagnostic and prognostic assessment and machine-learning approaches show promise in the diagnosis of pulmonary hypertension.

Lung perfusion: MRI vs. SPECT for screening in suspected chronic thromboembolic pulmonary hypertension.

PURPOSE: To assess the diagnostic accuracy of magnetic resonance imaging (MRI) perfusion against perfusion single photon emission tomography (SPECT) screening for chronic thromboembolic pulmonary hypertension (CTEPH). Ventilation/perfusion (V/Q) scintigraphy is recommended to screen for suspected CTEPH. It has previously been shown that 3D dynamic contrast-enhanced (DCE) lung perfusion MRI has a similar sensitivity for diagnosing CTEPH in comparison to planar perfusion scintigraphy; however, planar scintigraphy has now been largely replaced by SPECT, due to higher spatial resolution and sensitivity. MATERIALS AND METHODS: Consecutive patients with suspected CTEPH or unexplained pulmonary hypertension attending a referral center, who underwent lung DCE perfusion MRI at 1.5T, perfusion SPECT, and computed tomography pulmonary angiography (CTPA) within 14 days of right heart catheterization, from April 2013 to April 2014, were included. DCE-MR, SPECT, and CTPA were independently analyzed by two blinded radiologists. Disagreements were corrected by consensus. The gold standard reference for the diagnosis of chronic thromboemboli was based on a review of multimodality imaging and clinical findings. RESULTS: In all, 74 patients with suspected CTEPH underwent all three modalities. Forty-six were diagnosed with CTEPH (36) or chronic thromboembolic disease (CTED) (10). 3D DCE perfusion MRI correctly identified all patients (sensitivity of 100%), compared with a 97% sensitivity for SPECT. CONCLUSION: DCE lung perfusion MRI has increased sensitivity when compared with perfusion scintigraphy in screening for CTEPH. As MRI does not use ionizing radiation, it should be considered as a first-line imaging modality in suspected CTEPH. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2017;46:1693-1697.

Targeted Lung Health Check: breast related incidental findings-imaging appearances and lessons learned.

OBJECTIVE: The introduction of Targeted Lung Health Checks (TLHC) to screen for lung cancer has highlighted that incidental findings are common and require management strategies. This study analyses retrospectively, incidentally detected breast lesions reported as part of the TLHC referred to the Breast Cancer clinicians. METHODS: All participants with incidental breast nodules referred to the Breast Cancer team in the first year of screening were reviewed. RESULTS: Fifty-two participants (48 female; 92.3%) were referred to the Breast Multidisciplinary Team Meeting for assessment of 43 breast nodules, 8 breast asymmetry/dense breasts, and 2 likely breast related metastatic disease. One participant declined breast team referral. For the 42 breast nodules investigated, the final diagnoses were 5 breast carcinomas, 10 normal breast tissue, and 27 benign nodules. One male patient was diagnosed with breast carcinoma. The 29 breast nodules classified as smooth and well defined were all benign. No malignancy was demonstrated in the group with asymmetric or dense breast tissue. Metastatic breast carcinoma was confirmed in two participants. Twenty-six out of thirty-seven (54%) females had prior breast screening mammograms precluding further investigation. CONCLUSION: Incidental breast nodules are common on THLC scans. Smooth, sharply defined breast nodules are likely to be benign but low-dose CT is poor at accurately assessing breast nodules. Agreed breast referral pathways prior to starting the Lung Cancer Screening programme are recommended. Access to screening mammograms can reduce referrals to the Breast clinic. ADVANCES IN KNOWLEDGE: Lessons learned from TLHC pilot studies can be useful to sites commencing national TLHC programme.

Voxel-Wise Comparison of Co-Registered Quantitative CT and Hyperpolarised Gas Diffusion-Weighted MRI Measurements in IPF.

The patterns of idiopathic pulmonary fibrosis (IPF) lung disease that directly correspond to elevated hyperpolarised gas diffusion-weighted (DW) MRI metrics are currently unknown. This study aims to develop a spatial co-registration framework for a voxel-wise comparison of hyperpolarised gas DW-MRI and CALIPER quantitative CT patterns. Sixteen IPF patients underwent 3He DW-MRI and CT at baseline, and eleven patients had a 1-year follow-up DW-MRI. Six healthy volunteers underwent 129Xe DW-MRI at baseline only. Moreover, 3He DW-MRI was indirectly co-registered to CT via spatially aligned 3He ventilation and structural 1H MRI. A voxel-wise comparison of the overlapping 3He apparent diffusion coefficient (ADC) and mean acinar dimension (LmD) maps with CALIPER CT patterns was performed at baseline and after 1 year. The abnormal lung percentage classified with the LmD value, based on a healthy volunteer 129Xe LmD, and CALIPER was compared with a Bland-Altman analysis. The largest DW-MRI metrics were found in the regions classified as honeycombing, and longitudinal DW-MRI changes were observed in the baseline-classified reticular changes and ground-glass opacities regions. A mean bias of -15.3% (95% interval -56.8% to 26.2%) towards CALIPER was observed for the abnormal lung percentage. This suggests DW-MRI may detect microstructural changes in areas of the lung that are determined visibly and quantitatively normal by CT.

Evaluating the performance of artificial intelligence software for lung nodule detection on chest radiographs in a retrospective real-world UK population.

OBJECTIVES: Early identification of lung cancer on chest radiographs improves patient outcomes. Artificial intelligence (AI) tools may increase diagnostic accuracy and streamline this pathway. This study evaluated the performance of commercially available AI-based software trained to identify cancerous lung nodules on chest radiographs. DESIGN: This retrospective study included primary care chest radiographs acquired in a UK centre. The software evaluated each radiograph independently and outputs were compared with two reference standards: (1) the radiologist report and (2) the diagnosis of cancer by multidisciplinary team decision. Failure analysis was performed by interrogating the software marker locations on radiographs. PARTICIPANTS: 5722 consecutive chest radiographs were included from 5592 patients (median age 59 years, 53.8% women, 1.6% prevalence of cancer). RESULTS: Compared with radiologist reports for nodule detection, the software demonstrated sensitivity 54.5% (95% CI 44.2% to 64.4%), specificity 83.2% (82.2% to 84.1%), positive predictive value (PPV) 5.5% (4.6% to 6.6%) and negative predictive value (NPV) 99.0% (98.8% to 99.2%). Compared with cancer diagnosis, the software demonstrated sensitivity 60.9% (50.1% to 70.9%), specificity 83.3% (82.3% to 84.2%), PPV 5.6% (4.8% to 6.6%) and NPV 99.2% (99.0% to 99.4%). Normal or variant anatomy was misidentified as an abnormality in 69.9% of the 943 false positive cases. CONCLUSIONS: The software demonstrated considerable underperformance in this real-world patient cohort. Failure analysis suggested a lack of generalisability in the training and testing datasets as a potential factor. The low PPV carries the risk of over-investigation and limits the translation of the software to clinical practice. Our findings highlight the importance of training and testing software in representative datasets, with broader implications for the implementation of AI tools in imaging.

The reproducibility of manual RV/LV ratio measurement on CT pulmonary angiography.

Objectives: Right ventricular (RV) dysfunction carries elevated risk in acute pulmonary embolism (PE). An increased ratio between the size of the right and left ventricles (RV/LV ratio) is a biomarker of RV dysfunction. This study evaluated the reproducibility of RV/LV ratio measurement on CT pulmonary angiography (CTPA). Methods: 20 inpatient CTPA scans performed to assess for acute PE were retrospectively identified from a tertiary UK centre. Each scan was evaluated by 14 radiologists who provided a qualitative overall opinion on the presence of RV dysfunction and measured the RV/LV ratio. Using a threshold of 1.0, the RV/LV ratio measurements were classified as positive (≥1.0) or negative (<1.0) for RV dysfunction. Interobserver agreement was quantified using the Fleiss κ and intraclass correlation coefficient (ICC). Results: Qualitative opinion of RV dysfunction showed weak agreement (κ = 0.42, 95% CI 0.37-0.46). The mean RV/LV ratio measurement for all cases was 1.28 ± 0.68 with significant variation between reporters (p < 0.001). Although agreement for RV/LV measurement was good (ICC = 0.83, 95% CI 0.73-0.91), categorisation of RV dysfunction according to RV/LV ratio measurements showed weak agreement (κ = 0.46, 95% CI 0.41-0.50). Conclusion: Both qualitative opinion and quantitative manual RV/LV ratio measurement show poor agreement for identifying RV dysfunction on CTPA. Advances in knowledge: Caution should be exerted if using manual RV/LV ratio measurements to inform clinical risk stratification and management decisions.

Fully automatic cardiac four chamber and great vessel segmentation on CT pulmonary angiography using deep learning.

Introduction: Computed tomography pulmonary angiography (CTPA) is an essential test in the work-up of suspected pulmonary vascular disease including pulmonary hypertension and pulmonary embolism. Cardiac and great vessel assessments on CTPA are based on visual assessment and manual measurements which are known to have poor reproducibility. The primary aim of this study was to develop an automated whole heart segmentation (four chamber and great vessels) model for CTPA. Methods: A nine structure semantic segmentation model of the heart and great vessels was developed using 200 patients (80/20/100 training/validation/internal testing) with testing in 20 external patients. Ground truth segmentations were performed by consultant cardiothoracic radiologists. Failure analysis was conducted in 1,333 patients with mixed pulmonary vascular disease. Segmentation was achieved using deep learning via a convolutional neural network. Volumetric imaging biomarkers were correlated with invasive haemodynamics in the test cohort. Results: Dice similarity coefficients (DSC) for segmented structures were in the range 0.58-0.93 for both the internal and external test cohorts. The left and right ventricle myocardium segmentations had lower DSC of 0.83 and 0.58 respectively while all other structures had DSC >0.89 in the internal test cohort and >0.87 in the external test cohort. Interobserver comparison found that the left and right ventricle myocardium segmentations showed the most variation between observers: mean DSC (range) of 0.795 (0.785-0.801) and 0.520 (0.482-0.542) respectively. Right ventricle myocardial volume had strong correlation with mean pulmonary artery pressure (Spearman's correlation coefficient = 0.7). The volume of segmented cardiac structures by deep learning had higher or equivalent correlation with invasive haemodynamics than by manual segmentations. The model demonstrated good generalisability to different vendors and hospitals with similar performance in the external test cohort. The failure rates in mixed pulmonary vascular disease were low (<3.9%) indicating good generalisability of the model to different diseases. Conclusion: Fully automated segmentation of the four cardiac chambers and great vessels has been achieved in CTPA with high accuracy and low rates of failure. DL volumetric biomarkers can potentially improve CTPA cardiac assessment and invasive haemodynamic prediction.

Machine learning cardiac-MRI features predict mortality in newly diagnosed pulmonary arterial hypertension.

Aims: Pulmonary arterial hypertension (PAH) is a rare but serious disease associated with high mortality if left untreated. This study aims to assess the prognostic cardiac magnetic resonance (CMR) features in PAH using machine learning. Methods and results: Seven hundred and twenty-three consecutive treatment-naive PAH patients were identified from the ASPIRE registry; 516 were included in the training, and 207 in the validation cohort. A multilinear principal component analysis (MPCA)-based machine learning approach was used to extract mortality and survival features throughout the cardiac cycle. The features were overlaid on the original imaging using thresholding and clustering of high- and low-risk of mortality prediction values. The 1-year mortality rate in the validation cohort was 10%. Univariable Cox regression analysis of the combined short-axis and four-chamber MPCA-based predictions was statistically significant (hazard ratios: 2.1, 95% CI: 1.3, 3.4, c-index = 0.70, P = 0.002). The MPCA features improved the 1-year mortality prediction of REVEAL from c-index = 0.71 to 0.76 (P ≤ 0.001). Abnormalities in the end-systolic interventricular septum and end-diastolic left ventricle indicated the highest risk of mortality. Conclusion: The MPCA-based machine learning is an explainable time-resolved approach that allows visualization of prognostic cardiac features throughout the cardiac cycle at the population level, making this approach transparent and clinically interpretable. In addition, the added prognostic value over the REVEAL risk score and CMR volumetric measurements allows for a more accurate prediction of 1-year mortality risk in PAH.

Right ventricular remodelling in pulmonary arterial hypertension predicts treatment response.

OBJECTIVES: To determine the prognostic value of patterns of right ventricular adaptation in patients with pulmonary arterial hypertension (PAH), assessed using cardiac magnetic resonance (CMR) imaging at baseline and follow-up. METHODS: Patients attending the Sheffield Pulmonary Vascular Disease Unit with suspected pulmonary hypertension were recruited into the ASPIRE (Assessing the Spectrum of Pulmonary hypertension Identified at a REferral Centre) Registry. With exclusion of congenital heart disease, consecutive patients with PAH were followed up until the date of census or death. Right ventricular end-systolic volume index adjusted for age and sex and ventricular mass index were used to categorise patients into four different volume/mass groups: low-volume-low-mass, low-volume-high-mass, high-volume-low-mass and high-volume-high-mass. The prognostic value of the groups was assessed with one-way analysis of variance and Kaplan-Meier plots. Transition of the groups was studied. RESULTS: A total of 505 patients with PAH were identified, 239 (47.3%) of whom have died at follow-up (median 4.85 years, IQR 4.05). The mean age of the patients was 59±16 and 161 (32.7%) were male. Low-volume-low-mass was associated with CMR and right heart catheterisation metrics predictive of improved prognosis. There were 124 patients who underwent follow-up CMR (median 1.11 years, IQR 0.78). At both baseline and follow-up, the high-volume-low-mass group had worse prognosis than the low-volume-low-mass group (p<0.001). With PAH therapy, 73.5% of low-volume-low-mass patients remained in this group, whereas only 17.4% of high-volume-low-mass patients transitioned into low-volume-low-mass. CONCLUSIONS: Right ventricular adaptation assessed using CMR has prognostic value in patients with PAH. Patients with maladaptive remodelling (high-volume-low-mass) are at high risk of treatment failure.

Imaging and Risk Stratification in Pulmonary Arterial Hypertension: Time to Include Right Ventricular Assessment.

Background: Current European Society of Cardiology and European Respiratory Society guidelines recommend regular risk stratification with an aim of treating patients with pulmonary arterial hypertension (PAH) to improve or maintain low-risk status (<5% 1-year mortality). Methods: Consecutive patients with PAH who underwent cardiac magnetic resonance imaging (cMRI) were identified from the Assessing the Spectrum of Pulmonary hypertension Identified at a Referral centre (ASPIRE) registry. Kaplan-Meier survival curves, locally weighted scatterplot smoothing regression and multi-variable logistic regression analysis were performed. Results: In 311 consecutive, treatment-naïve patients with PAH undergoing cMRI including 121 undergoing follow-up cMRI, measures of right ventricular (RV) function including right ventricular ejection fraction (RVEF) and RV end systolic volume and right atrial (RA) area had prognostic value. However, only RV metrics were able to identify a low-risk status. Age (p < 0.01) and RVEF (p < 0.01) but not RA area were independent predictors of 1-year mortality. Conclusion: This study highlights the need for guidelines to include measures of RV function rather than RA area alone to aid the risk stratification of patients with PAH.

Lung MRI with hyperpolarised gases: current & future clinical perspectives.

The use of pulmonary MRI in a clinical setting has historically been limited. Whilst CT remains the gold-standard for structural lung imaging in many clinical indications, technical developments in ultrashort and zero echo time MRI techniques are beginning to help realise non-ionising structural imaging in certain lung disorders. In this invited review, we discuss a complementary technique - hyperpolarised (HP) gas MRI with inhaled 3He and 129Xe - a method for functional and microstructural imaging of the lung that has great potential as a clinical tool for early detection and improved understanding of pathophysiology in many lung diseases. HP gas MRI now has the potential to make an impact on clinical management by enabling safe, sensitive monitoring of disease progression and response to therapy. With reference to the significant evidence base gathered over the last two decades, we review HP gas MRI studies in patients with a range of pulmonary disorders, including COPD/emphysema, asthma, cystic fibrosis, and interstitial lung disease. We provide several examples of our experience in Sheffield of using these techniques in a diagnostic clinical setting in challenging adult and paediatric lung diseases.