Dataset Information

Radiomics and deep learning for myocardial scar screening in hypertrophic cardiomyopathy.

ABSTRACT:

Background

Myocardial scar burden quantified using late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR), has important prognostic value in hypertrophic cardiomyopathy (HCM). However, nearly 50% of HCM patients have no scar but undergo repeated gadolinium-based CMR over their life span. We sought to develop an artificial intelligence (AI)-based screening model using radiomics and deep learning (DL) features extracted from balanced steady state free precession (bSSFP) cine sequences to identify HCM patients without scar.

Methods

We evaluated three AI-based screening models using bSSFP cine image features extracted by radiomics, DL, or combined DL-Radiomics. Images for 759 HCM patients (50 ± 16 years, 66% men) in a multi-center/vendor study were used to develop and test model performance. An external dataset of 100 HCM patients (53 ± 14 years, 70% men) was used to assess model generalizability. Model performance was evaluated using area-under-receiver-operating curve (AUC).

Results

The DL-Radiomics model demonstrated higher AUC compared to DL and Radiomics in the internal (0.83 vs 0.77, p = 0.006 and 0.78, p = 0.05; n = 159) and external (0.74 vs 0.64, p = 0.006 and 0.71, p = 0.27; n = 100) datasets. The DL-Radiomics model correctly identified 43% and 28% of patients without scar in the internal and external datasets compared to 42% and 16% by Radiomics model and 42% and 23% by DL model, respectively.

Conclusions

A DL-Radiomics AI model using bSSFP cine images outperforms DL or Radiomics models alone as a scar screening tool prior to gadolinium administration. Despite its potential, the clinical utility of the model remains limited and further investigation is needed to improve the accuracy and generalizability.

SUBMITTER: Fahmy AS

PROVIDER: S-EPMC9235098 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background Cardiac MRI late gadolinium enhancement (LGE) scar volume is an important marker for outcome prediction in patients with hypertrophic cardiomyopathy (HCM); however, its clinical application is hindered by a lack of measurement standardization. Purpose To develop and evaluate a three-dimensional (3D) convolutional neural network (CNN)-based method for automated LGE scar quantification in patients with HCM. Materials and Methods We retrospectively identified LGE MRI data in a multicenter (n = 7) and multivendor (n = 3) HCM study obtained between November 2001 and November 2011. A deep 3D CNN based on U-Net architecture was used for LGE scar quantification. Independent CNN training and testing data sets were maintained with a 4:1 ratio. Stacks of short-axis MRI slices were split into overlapping substacks that were segmented and then merged into one volume. The 3D CNN per-site and per-vendor performances were evaluated with respect to manual scar quantification performed in a core laboratory setting using Dice similarity coefficient (DSC), Pearson correlation, and Bland-Altman analyses. Furthermore, the performance of 3D CNN was compared with that of two-dimensional (2D) CNN. Results This study included 1073 patients with HCM (733 men; mean age, 49 years ± 17 [standard deviation]). The 3D CNN-based quantification was fast (0.15 second per image) and demonstrated excellent correlation with manual scar volume quantification (r = 0.88, P < .001) and ratio of scar volume to total left ventricle myocardial volume (%LGE) (r = 0.91, P < .001). The 3D CNN-based quantification strongly correlated with manual quantification of scar volume (r = 0.82-0.99, P < .001) and %LGE (r = 0.90-0.97, P < .001) for all sites and vendors. The 3D CNN identified patients with a large scar burden (>15%) with 98% accuracy (202 of 207) (95% confidence interval [CI]: 95%, 99%). When compared with 3D CNN, 2D CNN underestimated scar volume (r = 0.85, P < .001) and %LGE (r = 0.83, P < .001). The DSC of 3D CNN segmentation was comparable among different vendors (P = .07) and higher than that of 2D CNN (DSC, 0.54 ± 0.26 vs 0.48 ± 0.29; P = .02). Conclusion In the hypertrophic cardiomyopathy population, a three-dimensional convolutional neural network enables fast and accurate quantification of myocardial scar volume, outperforms a two-dimensional convolutional neural network, and demonstrates comparable performance across different vendors. © RSNA, 2019 Online supplemental material is available for this article.

Project description:BackgroundIn patients with suspected or known hypertrophic cardiomyopathy (HCM), late gadolinium enhancement (LGE) provides diagnostic and prognostic value. However, contraindications and long-term retention of gadolinium have raised concern about repeated gadolinium administration in this population. Alternatively, native T1 -mapping enables identification of focal fibrosis, the substrate of LGE. However HCM-specific heterogeneous fibrosis distribution leads to subtle T1 -maps changes that are difficult to identify.PurposeTo apply radiomic texture analysis on native T1 -maps to identify patients with a low likelihood of LGE(+), thereby reducing the number of patients exposed to gadolinium administration.Study typeRetrospective interpretation of prospectively acquired data.SubjectsIn all, 188 (54.7 ± 14.4 years, 71% men) with suspected or known HCM.Field strength/sequenceA 1.5T scanner; slice-interleaved native T1 -mapping (STONE) sequence and 3D LGE after administration of 0.1 mmol/kg of gadobenate dimeglumine.AssessmentLeft ventricular LGE images were location-matched with native T1 -maps using anatomical landmarks. Using a split-sample validation approach, patients were randomly divided 3:1 (training/internal validation vs. test cohorts). To balance the data during training, 50% of LGE(-) slices were discarded.Statistical testsFour sets of texture descriptors were applied to the training dataset for capture of spatially dependent and independent pixel statistics. Five texture features were sequentially selected with the best discriminatory capacity between LGE(+) and LGE(-) T1 -maps and tested using a decision tree ensemble (DTE) classifier.ResultsThe selected texture features discriminated between LGE(+) and LGE(-) T1 -maps with a c-statistic of 0.75 (95% confidence interval [CI]: 0.70-0.80) using 10-fold cross-validation during internal validation in the training dataset and 0.74 (95% CI: 0.65-0.83) in the independent test dataset. The DTE classifier provided adequate labeling of all (100%) LGE(+) patients and 37% of LGE(-) patients during testing.Data conclusionRadiomic analysis of native T1 -images can identify ~1/3 of LGE(-) patients for whom gadolinium administration can be safely avoided.Level of evidence2 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2020. J. Magn. Reson. Imaging 2020;52:906-919.

Project description:Background Impaired myocardial blood flow (MBF) in the absence of epicardial coronary disease is a feature of hypertrophic cardiomyopathy (HCM). Although most evident in hypertrophied or scarred segments, reduced MBF can occur in apparently normal segments. We hypothesized that impaired MBF and myocardial perfusion reserve, quantified using perfusion mapping cardiac magnetic resonance, might occur in the absence of overt left ventricular hypertrophy (LVH) and late gadolinium enhancement, in mutation carriers without LVH criteria for HCM (genotype-positive, left ventricular hypertrophy-negative). Methods and Results A single center, case-control study investigated MBF and myocardial perfusion reserve (the ratio of MBF at stress:rest), along with other pre-phenotypic features of HCM. Individuals with genotype-positive, left ventricular hypertrophy-negative (n=50) with likely pathogenic/pathogenic variants and no evidence of LVH, and matched controls (n=28) underwent cardiac magnetic resonance. Cardiac magnetic resonance identified LVH-fulfilling criteria for HCM in 5 patients who were excluded. Individuals with genotype-positive, left ventricular hypertrophy-negative had longer indexed anterior mitral valve leaflet length (12.52±2.1 versus 11.55±1.6 mm/m2, P=0.03), lower left ventricular end-systolic volume (21.0±6.9 versus 26.7±6.2 mm/m2, P≤0.005) and higher left ventricular ejection fraction (71.9±5.5 versus 65.8±4.4%, P≤0.005). Maximum wall thickness was not significantly different (9.03±1.95 versus 8.37±1.2 mm, P=0.075), and no subject had significant late gadolinium enhancement (minor right ventricle‒insertion point late gadolinium enhancement only). Perfusion mapping demonstrated visual perfusion defects in 9 (20%) carriers versus 0 controls (P=0.011). These were almost all septal or near right ventricle insertion points. Globally, myocardial perfusion reserve was lower in carriers (2.77±0.83 versus 3.24±0.63, P=0.009), with a subendocardial:subepicardial myocardial perfusion reserve gradient (2.55±0.75 versus 3.2±0.65, P=<0.005; 3.01±0.96 versus 3.47±0.75, P=0.026) but equivalent MBF (2.75±0.82 versus 2.65±0.69 mL/g per min, P=0.826). Conclusions Regional and global impaired myocardial perfusion can occur in HCM mutation carriers, in the absence of significant hypertrophy or scarring.

Project description:Genetic testing provides valuable insights into family screening strategies, diagnosis, and prognosis in patients with hypertrophic cardiomyopathy (HCM). On the other hand, genetic testing carries socio-economical and psychological burdens. It is therefore important to identify patients with HCM who are more likely to have positive genotype. However, conventional prediction models based on clinical and echocardiographic parameters offer only modest accuracy and are subject to intra- and inter-observer variability. We therefore hypothesized that deep convolutional neural network (DCNN, a type of deep learning) analysis of echocardiographic images improves the predictive accuracy of positive genotype in patients with HCM. In each case, we obtained parasternal short- and long-axis as well as apical 2-, 3-, 4-, and 5-chamber views. We employed DCNN algorithm to predict positive genotype based on the input echocardiographic images. We performed 5-fold cross-validations. We used 2 reference models-the Mayo HCM Genotype Predictor score (Mayo score) and the Toronto HCM Genotype score (Toronto score). We compared the area under the receiver-operating-characteristic curve (AUC) between a combined model using the reference model plus DCNN-derived probability and the reference model. We calculated the p-value by performing 1,000 bootstrapping. We calculated sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). In addition, we examined the net reclassification improvement. We included 99 adults with HCM who underwent genetic testing. Overall, 45 patients (45%) had positive genotype. The new model combining Mayo score and DCNN-derived probability significantly outperformed Mayo score (AUC 0.86 [95% CI 0.79-0.93] vs. 0.72 [0.61-0.82]; p < 0.001). Similarly, the new model combining Toronto score and DCNN-derived probability exhibited a higher AUC compared to Toronto score alone (AUC 0.84 [0.76-0.92] vs. 0.75 [0.65-0.85]; p = 0.03). An improvement in the sensitivity, specificity, PPV, and NPV was also achieved, along with significant net reclassification improvement. In conclusion, compared to the conventional models, our new model combining the conventional and DCNN-derived models demonstrated superior accuracy to predict positive genotype in patients with HCM.

Project description:ObjectivesThis study aimed to determine whether texture analysis (TA) and machine learning-based classifications can be applied in differential diagnosis of cardiac amyloidosis (CA) and hypertrophic cardiomyopathy (HCM) using non-contrast cine cardiac magnetic resonance (CMR) images.MethodsIn this institutional review board-approved study, we consecutively enrolled 167 patients with CA (n = 85), HCM (n = 82), and 84 patients with normal CMR served as controls. All cases were randomized into training [119 patients (70%)] and validation [48 patients (30%)] groups. A total of 275 texture features were extracted from cine images. Based on regression analysis with the least absolute shrinkage and selection operator (LASSO), nine machine learning models were established and their diagnostic performance determined.ResultsNineteen radiomics texture features derived from cine images were used to differentiate CA and HCM. In the validation cohort, the support vector machine (SVM), which had an accuracy of 0.85, showed the best performance (MCC = 0.637). Gray level non-uniformity (GLevNonU) was the single most effective feature. The combined model of radiomics texture features and conventional MR metrics had superior discriminatory performance (AUC = 0.89) over conventional MR metrics model (AUC = 0.79). Moreover, results showed that GLevNonU levels in HCM patients were significantly higher compared with levels in CA patients and control groups (P < 0.001). A cut-off of GLevNonU ≥ 25 was shown to differentiate between CA and HCM patients, with an area under the curve (AUC) of 0.86 (CI:0.804-0.920). Multiple comparisons tests showed that GLevNonU was significantly greater in LGE+, relative to LGE-patient groups (CA+ vs. CA- and HCM+ vs. HCM-, P = 0.01, 0.001, respectively).ConclusionMachine learning-based classifiers can accurately differentiate between CA and HCM on non-contrast cine images. The radiomics-MR combined model can be used to improve the discriminatory performance. TA may be used to assess myocardial microstructure changes that occur during different stages of cardiomyopathies.