Cine-cardiac magnetic resonance to distinguish between ischemic and non-ischemic cardiomyopathies: a machine learning approach.

OBJECTIVE: This work aimed to derive a machine learning (ML) model for the differentiation between ischemic cardiomyopathy (ICM) and non-ischemic cardiomyopathy (NICM) on non-contrast cardiovascular magnetic resonance (CMR). METHODS: This retrospective study evaluated CMR scans of 107 consecutive patients (49 ICM, 58 NICM), including atrial and ventricular strain parameters. We used these data to compare an explainable tree-based gradient boosting additive model with four traditional ML models for the differentiation of ICM and NICM. The models were trained and internally validated with repeated cross-validation according to discrimination and calibration. Furthermore, we examined important variables for distinguishing between ICM and NICM. RESULTS: A total of 107 patients and 38 variables were available for the analysis. Of those, 49 were ICM (34 males, mean age 60 ± 9 years) and 58 patients were NICM (38 males, mean age 56 ± 19 years). After 10 repetitions of the tenfold cross-validation, the proposed model achieved the highest area under curve (0.82, 95% CI [0.47-1.00]) and lowest Brier score (0.19, 95% CI [0.13-0.27]), showing competitive diagnostic accuracy and calibration. At the Youden's index, sensitivity was 0.72 (95% CI [0.68-0.76]), the highest of all. Analysis of predictions revealed that both atrial and ventricular strain CMR parameters were important for the identification of ICM patients. CONCLUSION: The current study demonstrated that using a ML model, multi chamber myocardial strain, and function on non-contrast CMR parameters enables the discrimination between ICM and NICM with competitive diagnostic accuracy. CLINICAL RELEVANCE STATEMENT: A machine learning model based on non-contrast cardiovascular magnetic resonance parameters may discriminate between ischemic and non-ischemic cardiomyopathy enabling wider access to cardiovascular magnetic resonance examinations with lower costs and faster imaging acquisition. KEY POINTS: • The exponential growth in cardiovascular magnetic resonance examinations may require faster and more cost-effective protocols. • Artificial intelligence models can be utilized to distinguish between ischemic and non-ischemic etiologies. • Machine learning using non-contrast CMR parameters can effectively distinguish between ischemic and non-ischemic cardiomyopathies.

Value of a short non-contrast CMR protocol in MINOCA.

OBJECTIVES: To evaluate the diagnostic performance of a short non-contrast CMR (ShtCMR) protocol relative to a matched standard comprehensive CMR (StdCMR) protocol in patients with myocardial infarction with non-obstructive coronary arteries (MINOCA). METHODS: This multicenter retrospective study included patients with a working diagnosis of MINOCA who underwent a StdCMR between January 2019 and December 2020. An expert and a non-expert reader performed a blinded reading with the ShtCMR (long-axis cine images, T2w-STIR, T1- and T2-mapping). A consensus reading of the StdCMR (reference standard) was performed at least 3 months after the ShtCMR reading session. Readers were asked to report the following: (1) diagnosis; (2) level of confidence in their diagnosis with the ShtCMR; (3) number of myocardial segments involved, and (4) functional parameters. RESULTS: A total of 179 patients were enrolled. The ShtCMR lasted 21 ± 9 min and the StdCMR 45 ± 11 min (p < 0.0001). ShtCMR allowed reaching the same diagnosis as StdCMR in 85% of patients when interpreted by expert readers (rising from 66% for poor confidence to 99% for good, p = 0.0001) and in 73% (p = 0.01) by non-expert ones (60% for poor vs 89% for good confidence, p = 0.0001). Overall, the ShtCMR overestimated the ejection fraction, underestimated cardiac volumes (p < 0.01), and underestimated the number of segments involved by pathology (p = 0.0008) when compared with the StdCMR. CONCLUSION: The ShtCMR was found to be a debatable alternative to the StdCMR in patients with MINOCA. Nevertheless, when an experienced reader reaches a good or very good diagnostic confidence using the ShtCMR, the reader may choose to stop the examination, reducing the length of the CMR without affecting the patient's diagnosis. CLINICAL RELEVANCE STATEMENT: A short non-contrast CMR protocol may be a viable alternative to standard protocols in selected CMR studies of patients with MINOCA, allowing for faster diagnosis while reducing time and resources and increasing the number of patients who can be scanned. KEY POINTS: • The ShtCMR lasted 21 ± 9 min and the StdCMR 45 ± 11 min (p < 0.0001). • In 57% of patients with MINOCA, the experienced reader considers that contrast medium is probably not necessary for diagnosis without affecting the patient's diagnosis (99% of agreement rate between ShtCMR and StdCMR).

Base-to-apex Gradient Pattern Assessed by Cardiovascular Magnetic Resonance in Takotsubo Cardiomyopathy.

OBJECTIVES: The purpose of this study was to investigate the base-to-apex gradient strain pattern as a noncontrast cardiovascular magnetic resonance (CMR) parameter in patients with Takotsubo cardiomyopathy (TTC) and determine whether this pattern may help discriminate TTC from patients with anterior myocardial infarction (AMI). MATERIALS AND METHODS: A total of 80 patients were included in the analysis: 30 patients with apical ballooning TTC and 50 patients with AMI. Global and regional ventricular function, including longitudinal (LS), circumferential (CS), and radial strain (RS), were assessed using CMR. The base-to-apex LS, RS, and CS gradients, defined as the peak gradient difference between averaged basal and apical strain, were calculated. RESULTS: The base-to-apex RS gradient was impaired in TTC patients compared with the AMI group (14.04 ± 15.50 vs. -0.43 ± 11.59, P=0.001). Conversely, there were no significant differences in the base-to-apex LS and CS gradients between the AMI group and TTC patients (0.14 ± 2.71 vs. -1.5 ± 3.69, P=0.054: -0.99 ± 6.49 vs. ±1.4 ± 5.43, P=0.47, respectively). Beyond the presence and extension of LGE, base-to-apex RS gradient was the only independent discriminator between TTC and AMI (OR 1.28; 95% CI 1.08, 1.52, P=0.006) in multivariate logistic regression analysis. CONCLUSION: The findings of this study suggest that the pattern of regional myocardial strain impairment could serve as an additional noncontrast CMR tool to refine the diagnosis of TTC. A pronounced base-to-apex RS gradient may be a specific left ventricle strain pattern of TTC.