Development and validation of AI-derived segmentation of four-chamber cine cardiac magnetic resonance

Abstract

Background: Cardiac magnetic resonance (CMR) in the four-chamber plane offers comprehensive insight into the volumetrics of the heart. We aimed to develop an artificial intelligence (AI) model of time-resolved segmentation using the four-chamber cine.

Methods: A fully automated deep learning algorithm was trained using retrospective multicentre and multivendor data of 814 subjects. Validation, reproducibility, and mortality prediction were evaluated on an independent cohort of 101 subjects.

Results: The mean age of the validation cohort was 54 years, and 66 (65%) were males. Left and right heart parameters demonstrated strong correlations between automated and manual analysis, with a ρ of 0.91-0.98 and 0.89-0.98, respectively, with minimal bias. All AI four-chamber volumetrics in repeatability analysis demonstrated high correlation (ρ = 0.99-1.00) and no bias. Automated four-chamber analysis underestimated both left ventricular (LV) and right ventricular (RV) volumes compared to ground-truth short-axis cine analysis. Two correction factors for LV and RV four-chamber analysis were proposed based on systematic bias. After applying the correction factors, a strong correlation and minimal bias for LV volumetrics were observed. During a mean follow-up period of 6.75 years, 16 patients died. On stepwise multivariable analysis, left atrial ejection fraction demonstrated an independent association with death in both manual (hazard ratio (HR) = 0.96, p = 0.003) and AI analyses (HR = 0.96, p < 0.001).

Conclusion: Fully automated four-chamber CMR is feasible, reproducible, and has the same real-world prognostic value as manual analysis. LV volumes by four-chamber segmentation were comparable to short-axis volumetric assessment.

Trials registration: ClinicalTrials.gov: NCT05114785.

Relevance statement: Integrating fully automated AI in CMR promises to revolutionise clinical cardiac assessment, offering efficient, accurate, and prognostically valuable insights for improved patient care and outcomes.

Key points: • Four-chamber cine sequences remain one of the most informative acquisitions in CMR examination. • This deep learning-based, time-resolved, fully automated four-chamber volumetric, functional, and deformation analysis solution. • LV and RV were underestimated by four-chamber analysis compared to ground truth short-axis segmentation. • Correction bias for both LV and RV volumes by four-chamber segmentation, minimises the systematic bias.

Keywords: Artificial intelligence, Deep learning, Heart diseases, Magnetic resonance imaging (cine), Prognosis.

Fig. 1

Overview of study flow chart. AI Artificial intelligence, LA Left atrium, LV Left ventricle, RA Right atrium, RV Right ventricle

Fig. 2

Examples of the manual and automated four-chamber segmentation methods and their correlation. The AI method segments all time frames; however, only end-diastole and end-systole frames are demonstrated. The coloured contours are as follows: green for the LV epicardium, red for the LV endocardium, pink for the LA, yellow for the RV endocardium, and turquoise contours for the RA. AI Artificial intelligence, EDV End-diastolic volume, EF Ejection fraction, ESV End-systolic volume, LA Left atrium, LV Left ventricle, ρ Spearman correlation coefficient, RA Right atrium, RV Right ventricle

Fig. 3

Quantification results of mean LV and RV volumes of automated four-chamber and short-axis segmentation methods over time. a AI-generated contours of short-axis cine stack of images using standard endocardial and epicardial contour methods. b AI-generated four-chamber cine segmentation contours using standard endocardial and epicardial contours methods. c The AI-generated four-chamber cine segmentation results yielded slightly lower left and right ventricular volumes than the ground truth. d Quantification results of automated four-chamber mean left and right ventricular volumes after applying 14.56 mL and 50.78 mL correction factors and short-axis segmentation results over time. The AI-generated four-chamber cine LV segmentation results were similar, and RV yielded slightly higher volumes than the ground truth. AI Artificial intelligence, 4CH Four-chamber, LV Left ventricle, RV Right ventricle, SAX Short-axis

Fig. 4

Survival analysis. a Kaplan–Meier analysis demonstrates that patients with left atrial ejection fraction < 55% had a higher risk of death. b The risk of death is higher when using artificial intelligence-generated segmentation. EF Ejection fraction, LA Left atrium

Fig. 5

Kaplan-Meir analysis demonstrates that AI-generated LA ejection fraction < 55% is independently associated with risk of death after adjusting for manual LA EF in the regression model. AI Artificial intelligence, EF Ejection fraction, LA Left atrium