Automated calculation of the right ventricle to left ventricle ratio on CT for the risk stratification of patients with acute pulmonary embolism

Abstract

Objectives To assess the feasibility and reliability of the use of artificial intelligence post-processing to calculate the RV:LV diameter ratio on computed tomography pulmonary angiography (CTPA) and to investigate its prognostic value in patients with acute PE. Methods Single-centre, retrospective study of 101 consecutive patients with CTPA-proven acute PE. RV and LV volumes were segmented on 1-mm contrast-enhanced axial slices and maximal ventricular diameters were derived for RV:LV ratio using automated post-processing software (IMBIO LLC, USA) and compared to manual analysis in two observers, via intraclass coefficient correlation analysis. Each CTPA report was analysed for mention of the RV:LV ratio and compared to the automated RV:LV ratio. Thirty-day all-cause mortality post-CTPA was recorded. Results Automated RV:LV analysis was feasible in 87% (n = 88). RV:LV ratios ranged from 0.67 to 2.43, with 64% (n = 65) > 1.0. There was very strong agreement between manual and automated RV:LV ratios (ICC = 0.83, 0.77-0.88). The use of automated analysis led to a change in risk stratification in 45% of patients (n = 40). The AUC of the automated measurement for the prediction of all-cause 30-day mortality was 0.77 (95% CI: 0.62-0.99). ConclusionThe RV:LV ratio on CTPA can be reliably measured automatically in the majority of real-world cases of acute PE, with perfect reproducibility. The routine use of this automated analysis in clinical practice would add important prognostic information in patients with acute PE. Key Points • Automated calculation of the right ventricle to left ventricle ratio was feasible in the majority of patients and demonstrated perfect intraobserver variability. • Automated analysis would have added important prognostic information and altered risk stratification in the majority of patients. • The optimal cut-off value for the automated right ventricle to left ventricle ratio was 1.18, with a sensitivity of 100% and specificity of 54% for the prediction of 30-day mortality.

Teaching Nonradiologists to Identify Right Heart Strain on Computed Tomography Scans of Acute Pulmonary Embolism

“…A retrospective study looking at cardiovascular and thoracic radiologists’ assessments of RV parameters on CTPA consistently showed ICCs >0.95 ( 13 ). Comparatively, a recent study of artificial intelligence software measuring RV/LV ratio in patients with acute pulmonary embolism showed an ICC of 0.83 compared with a radiologist standard ( 14 ). This was defined as a “very good” correlation and is consistent with the correlation we found between our trainees and the standard.…”

Assessment of pulmonary embolism severity and the risk of early death

“…It is reassuring that repeated measurements performed by expert radiologists as well as physicians trained in CTPA interpretation show a satisfactory intra-and interobserver correlation [26]. Among tested parameters, the right ventricle/left ventricle ratio (RV/LV) ratio measured in the transverse view by identifying the maximal distance between the ventricular endocardium and the interventricular septum, perpendicular to the long axis of the heart has the most robust evidence as a prognostic marker and recent reports show that it can be reliably measured automatically in the majority of real-world cases of acute PE, with good reproducibility [80,81]. Values ≥ 1.0 are believed to be appropriate to indicate a poor prognosis, however, it should be noted that they are very common in the initially low-risk subgroup classified using clinical prognostic rules [80,81].…”

“…Among tested parameters, the right ventricle/left ventricle ratio (RV/LV) ratio measured in the transverse view by identifying the maximal distance between the ventricular endocardium and the interventricular septum, perpendicular to the long axis of the heart has the most robust evidence as a prognostic marker and recent reports show that it can be reliably measured automatically in the majority of real-world cases of acute PE, with good reproducibility [80,81]. Values ≥ 1.0 are believed to be appropriate to indicate a poor prognosis, however, it should be noted that they are very common in the initially low-risk subgroup classified using clinical prognostic rules [80,81]. Defining the optimal cutoff for low-risk patients remains to be determined, with some studies conducted on smaller groups of patients suggesting a threshold of 1.2 [82].…”

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

“…Right and left atrium measurements on computed tomography (CT) are also known to have diagnostic and prognostic value for pulmonary vascular disease ( 14 – 18 ). Manual cardiac and pulmonary measurements are limited by their time-consuming nature ( 14 ), human error, observer variability ( 19 , 20 ), and observer experience leading to potentially inaccurate predictions and less frequent use.…”

“…Automated ventricular volume measurement approaches have been developed in CT showing similar accuracy to cardiac magnetic resonance imaging (MRI) ( 21 ), and provide added prognostic value in acute pulmonary embolism ( 19 ). There is the need to develop methods to automatically measure the cardiac volume, myocardial hypertrophy, and great vessels on CTPA to provide a comprehensive cardiopulmonary assessment.…”