Radiomics of Non-Contrast-Enhanced T1 Mapping: Diagnostic and Predictive Performance for Myocardial Injury in Acute ST-Segment-Elevation Myocardial Infarction

Abstract

Objective: To evaluate the feasibility of texture analysis on non-contrast-enhanced T1 maps of cardiac magnetic resonance (CMR) imaging for the diagnosis of myocardial injury in acute myocardial infarction (MI).

Materials and methods: This study included 68 patients (57 males and 11 females; mean age, 55.7 ± 10.5 years) with acute ST-segment-elevation MI who had undergone 3T CMR after a percutaneous coronary intervention. Forty patients of them also underwent a 6-month follow-up CMR. The CMR protocol included T2-weighted imaging, T1 mapping, rest first-pass perfusion, and late gadolinium enhancement. Radiomics features were extracted from the T1 maps using open-source software. Radiomics signatures were constructed with the selected strongest features to evaluate the myocardial injury severity and predict the recovery of left ventricular (LV) longitudinal systolic myocardial contractility.

Results: A total of 1088 segments of the acute CMR images were analyzed; 103 (9.5%) segments showed microvascular obstruction (MVO), and 557 (51.2%) segments showed MI. A total of 640 segments were included in the 6-month follow-up analysis, of which 160 (25.0%) segments showed favorable recovery of LV longitudinal systolic myocardial contractility. Combined radiomics signature and T1 values resulted in a higher diagnostic performance for MVO compared to T1 values alone (area under the curve [AUC] in the training set; 0.88, 0.72, p = 0.031: AUC in the test set; 0.86, 0.71, p002). Combined radiomics signature and T1 values also provided a higher predictive value for LV longitudinal systolic myocardial contractility recovery compared to T1 values (AUC in the training set; 0.76, 0.55, p < 0.001: AUC in the test set; 0.77, 0.60, p < 0.001).

Conclusion: The combination of radiomics of non-contrast-enhanced T1 mapping and T1 values could provide higher diagnostic accuracy for MVO. Radiomics also provides incremental value in the prediction of LV longitudinal systolic myocardial contractility at six months.

Keywords: Machine learning; Magnetic resonance imaging; Myocardial infarction; Myocardial reperfusion injury.

Fig. 1. Image shows the three different sections (base, mid-cavity, apex) of one post-ST-segment-elevation myocardial infarction patient.

From top to bottom, numbers within the panels indicate LGE, native T1 mapping, ROI segmentation, and feature extraction. The infarct segment showed high T1 values compared to the adjacent and remote segments, as demonstrated in the bull's eye maps of T1 values and LGE. A total of 1088 segments were divided into a training set and test set at a random ratio of 7:3 for further analysis. LGE = late gadolinium enhancement, ROI = region of interest

Fig. 2. Graph shows ROC analyses for positive MVO versus negative MVO in the training (A) and test (B) datasets.

ROC analysis indicates that combined T1 values and radiomics features had a higher accuracy for diagnosing segments with MVO compared to T1 values alone, visual assessment, and radiomics signature alone in the training or test datasets. AUC = area under the curve, MVO = microvascular obstruction, ROC = receiver operating characteristics

Fig. 3. Graph shows ROC analyses for peak LGE transmurality level = 0 versus peak LGE transmurality level ≥ 1 in the training (A) and test (B) datasets.

ROC analysis indicates that combined T1 values and radiomics features had the highest accuracy for diagnosing segments with LGE in the training or test datasets.

Fig. 4. Graph shows ROC analyses for non-irreversible myocardial injury versus irreversible myocardial injury in the training (A) and test (B) datasets.

ROC analysis indicates that baseline LGE percentage predicted irreversible injury myocardial segments with the highest accuracy. Combined T1 values and radiomics features had a higher accuracy for predicting segments with irreversible myocardial injury compared to T1 values or radiomics signature alone in the training or test datasets.

Fig. 5. Graph shows ROC analyses for non-favorable SLS recovery versus favorable SLS recovery in the training (A) and test (B) datasets.

ROC analysis indicates that combined T1 values and radiomics features had the acceptable accuracy for predicting segments with favorable SLS recovery in the training or test datasets. SLS = segmental longitudinal strain