Browsing by Author "Dharmakumar, Rohan"
Now showing 1 - 10 of 15
Results Per Page
Sort Options
Item A Patch-Wise Deep Learning Approach for Myocardial Blood Flow Quantification with Robustness to Noise and Nonrigid Motion(IEEE, 2021) Youssef, Khalid; Heydari, Bobby; Rivero, Luis Zamudio; Beaulieu, Taylor; Cheema, Karandeep; Dharmakumar, Rohan; Sharif, Behzad; Medicine, School of MedicineQuantitative analysis of dynamic contrast-enhanced cardiovascular MRI (cMRI) datasets enables the assessment of myocardial blood flow (MBF) for objective evaluation of ischemic heart disease in patients with suspected coronary artery disease. State-of-the-art MBF quantification techniques use constrained deconvolution and are highly sensitive to noise and motion-induced errors, which can lead to unreliable outcomes in the setting of high-resolution MBF mapping. To overcome these limitations, recent iterative approaches incorporate spatial-smoothness constraints to tackle pixel-wise MBF mapping. However, such iterative methods require a computational time of up to 30 minutes per acquired myocardial slice, which is a major practical limitation. Furthermore, they cannot enforce robustness to residual nonrigid motion which can occur in clinical stress/rest studies of patients with arrhythmia. We present a non-iterative patch-wise deep learning approach for pixel-wise MBF quantification wherein local spatio-temporal features are learned from a large dataset of myocardial patches acquired in clinical stress/rest cMRI studies. Our approach is scanner-independent, computationally efficient, robust to noise, and has the unique feature of robustness to motion-induced errors. Numerical and experimental results obtained using real patient data demonstrate the effectiveness of our approach.Clinical Relevance- The proposed patch-wise deep learning approach significantly improves the reliability of high-resolution myocardial blood flow quantification in cMRI by improving its robustness to noise and nonrigid myocardial motion and is up to 300-fold faster than state-of-the-art iterative approaches.Item Artificial Intelligence for Contrast-Free MRI: Scar Assessment in Myocardial Infarction Using Deep Learning-Based Virtual Native Enhancement(American Heart Association, 2022-11-15) Zhang, Qiang; Burrage, Matthew K.; Shanmuganathan, Mayooran; Gonzales, Ricardo A.; Lukaschuk, Elena; Thomas, Katharine E.; Mills, Rebecca; Pelado, Joana Leal; Nikolaidou, Chrysovalantou; Popescu, Iulia A.; Lee, Yung P.; Zhang, Xinheng; Dharmakumar, Rohan; Myerson, Saul G.; Rider, Oliver; Oxford Acute Myocardial Infarction (OxAMI) Study; Channon, Keith M.; Neubauer, Stefan; Piechnik, Stefan K.; Ferreira, Vanessa M.; Medicine, School of MedicineBackground: Myocardial scar is currently assessed non-invasively using cardiovascular magnetic resonance (CMR) late gadolinium enhancement (LGE) as an imaging gold-standard. However, a contrast-free approach would provide many advantages, including a faster and cheaper scan, without contrast-associated problems. Methods: Virtual Native Enhancement (VNE) is a novel technology that can produce virtual LGE-like images, without the need for contrast. VNE combines cine imaging and native T1-maps to produce LGE-like images using artificial intelligence (AI). VNE was developed for patients with prior myocardial infarction on 4271 datasets (912 patients), where each dataset is comprised of slice position-matched cine, T1-maps and LGE images. After quality control, 3002 datasets (775 patients) were used for development, and 291 datasets (68 patients) for testing. The VNE generator was trained using generative adversarial networks, employing two adversarial discriminators to improve the image quality. The left ventricle was contoured semi-automatically. Myocardial scar volume was quantified using the full width at half maximum method. Scar transmurality was measured using the centerline chord method and visualized on bull’s eye plots. Lesion quantification by VNE and LGE were compared using linear regression, Pearson correlation (R) and intraclass correlation coefficients (ICC). Proof-of-principle histopathological comparison of VNE in a porcine model of myocardial infarction was also performed. Results: VNE provided significantly better image quality than LGE on blinded analysis by 5 independent operators on 291 datasets (all p<0.001). VNE correlated strongly with LGE in quantifying scar size (R=0.89, ICC=0.94) and transmurality (R=0.84, ICC=0.90) in 66 patients (277 test datasets). Two CMR experts reviewed all test image slices and reported an overall accuracy of 84% of VNE in detecting scar when compared with LGE, with specificity of 100% and sensitivity of 77%. VNE also showed excellent visuospatial agreement with histopathology in 2 cases of a porcine model of myocardial infarction. Conclusions: VNE demonstrated high agreement with LGE-CMR for myocardial scar assessment in patients with prior myocardial infarction in visuospatial distribution and lesion quantification, with superior image quality. VNE is a potentially transformative AI-based technology, with promise to reduce scan times and costs, increase clinical throughput, and improve the accessibility of CMR in the very-near future.Item Assessment of intramyocardial hemorrhage with dark-blood T2*-weighted cardiovascular magnetic resonance(Elsevier, 2021-07-15) Guan, Xingmin; Chen, Yinyin; Yang, Hsin‑Jung; Zhang, Xinheng; Ren, Daoyuan; Sykes, Jane; Butler, John; Han, Hui; Zeng, Mengsu; Prato, Frank S.; Dharmakumar, Rohan; Medicine, School of MedicineBackground: Intramyocardial hemorrhage (IMH) within myocardial infarction (MI) is associated with major adverse cardiovascular events. Bright-blood T2*-based cardiovascular magnetic resonance (CMR) has emerged as the reference standard for non-invasive IMH detection. Despite this, the dark-blood T2*-based CMR is becoming interchangeably used with bright-blood T2*-weighted CMR in both clinical and preclinical settings for IMH detection. To date however, the relative merits of dark-blood T2*-weighted with respect to bright-blood T2*-weighted CMR for IMH characterization has not been studied. We investigated the diagnostic capacity of dark-blood T2*-weighted CMR against bright-blood T2*-weighted CMR for IMH characterization in clinical and preclinical settings. Materials and methods: Hemorrhagic MI patients (n = 20) and canines (n = 11) were imaged in the acute and chronic phases at 1.5 and 3 T with dark- and bright-blood T2*-weighted CMR. Imaging characteristics (Relative signal-to-noise (SNR), Relative contrast-to-noise (CNR), IMH Extent) and diagnostic performance (sensitivity, specificity, accuracy, area-under-the-curve, and inter-observer variability) of dark-blood T2*-weighted CMR for IMH characterization were assessed relative to bright-blood T2*-weighted CMR. Results: At both clinical and preclinical settings, compared to bright-blood T2*-weighted CMR, dark-blood T2*-weighted images had significantly lower SNR, CNR and reduced IMH extent (all p < 0.05). Dark-blood T2*-weighted CMR also demonstrated weaker sensitivity, specificity, accuracy, and inter-observer variability compared to bright-blood T2*-weighted CMR (all p < 0.05). These observations were consistent across infarct age and imaging field strengths. Conclusion: While IMH can be visible on dark-blood T2*-weighted CMR, the overall conspicuity of IMH is significantly reduced compared to that observed in bright-blood T2*-weighted images, across infarct age in clinical and preclinical settings at 1.5 and 3 T. Hence, bright-blood T2*-weighted CMR would be preferable for clinical use since dark-blood T2*-weighted CMR carries the potential to misclassify hemorrhagic MIs as non-hemorrhagic MIs.Item Chronic heart failure following hemorrhagic myocardial infarction: mechanism, treatment and outlook(Shared Science Publishers, 2023-02-13) Chan, Shing Fai; Vora, Keyur; Dharmakumar, Rohan; Medicine, School of MedicineMyocardial infarction (MI), the blockage of arterial blood supply of the heart, is among the most common causes of death worldwide. Even when patients receive immediate treatment by re-opening blocked arteries, they often develop chronic heart failure (CHF) in the aftermath of MI events. Yet, the factors that contribute to the development of MI-associated CHF are poorly understood. In our recent study (Nat Commun 13:6394), we link intramyocardial hemorrhage, an injury which can occur during reperfusion of areas affected by MI, to an increased risk of CHF. Mechanistically, our data suggest that an iron-induced adverse cascade of events after hemorrhagic MI drives fatty degeneration of infarcted tissue, which ultimately contributes to negative cardiac remodeling. In this Microreview, we discuss the implications of our findings regarding the molecular mechanism, more targeted treatment options as well as perspectives in the clinical care of CHF after hemorrhagic MI.Item Clinical Outcomes of Patients With Acute Myocardial Infarction in Health Professional Shortage Areas in Indiana(Elsevier, 2024-03-08) Gunderman, David J.; Kumar, Ashish; Munguia-Vazquez, Raymundo; Vora, Keyur; Shah, Chirag; Lambert, Nathan; Cavanaugh, Brendan; Dharmakumar, Rohan; Kalra, Ankur; Medicine, School of MedicineItem Deep Learning-Based Segmentation and Uncertainty Assessment for Automated Analysis of Myocardial Perfusion MRI Datasets Using Patch-Level Training and Advanced Data Augmentation(IEEE, 2021) Yalcinkaya, Dilek Mirgun; Youssef, Khalid; Heydari, Bobby; Zamudio, Luis; Dharmakumar, Rohan; Sharif, Behzad; Medicine, School of MedicineIn this work, we develop a patch-level training approach and a task-driven intensity-based augmentation method for deep-learning-based segmentation of motion-corrected perfusion cardiac magnetic resonance imaging (MRI) datasets. Further, the proposed method generates an image-based uncertainty map thanks to a novel spatial sliding-window approach used during patch-level training, hence allowing for uncertainty quantification. Using the quantified uncertainty, we detect the out-of-distribution test data instances so that the end-user can be alerted that the test data is not suitable for the trained network. This feature has the potential to enable a more reliable integration of the proposed deep learning-based framework into clinical practice. We test our approach on external MRI data acquired using a different acquisition protocol to demonstrate the robustness of our performance to variations in pulse-sequence parameters. The presented results further demonstrate that our deep-learning image segmentation approach trained with the proposed data-augmentation technique incorporating spatiotemporal (2D+time) patches is superior to the state-of-the-art 2D approach in terms of generalization performance.Item Disentangle, Align and Fuse for Multimodal and Semi-Supervised Image Segmentation(IEEE, 2021) Chartsias, Agisilaos; Papanastasiou, Giorgos; Wang, Chengjia; Semple, Scott; Newby, David E.; Dharmakumar, Rohan; Tsaftaris, Sotirios A.; Medicine, School of MedicineMagnetic resonance (MR) protocols rely on several sequences to assess pathology and organ status properly. Despite advances in image analysis, we tend to treat each sequence, here termed modality, in isolation. Taking advantage of the common information shared between modalities (an organ's anatomy) is beneficial for multi-modality processing and learning. However, we must overcome inherent anatomical misregistrations and disparities in signal intensity across the modalities to obtain this benefit. We present a method that offers improved segmentation accuracy of the modality of interest (over a single input model), by learning to leverage information present in other modalities, even if few (semi-supervised) or no (unsupervised) annotations are available for this specific modality. Core to our method is learning a disentangled decomposition into anatomical and imaging factors. Shared anatomical factors from the different inputs are jointly processed and fused to extract more accurate segmentation masks. Image misregistrations are corrected with a Spatial Transformer Network, which non-linearly aligns the anatomical factors. The imaging factor captures signal intensity characteristics across different modality data and is used for image reconstruction, enabling semi-supervised learning. Temporal and slice pairing between inputs are learned dynamically. We demonstrate applications in Late Gadolinium Enhanced (LGE) and Blood Oxygenation Level Dependent (BOLD) cardiac segmentation, as well as in T2 abdominal segmentation. Code is available at https://github.com/vios-s/multimodal_segmentation.Item Disparities in Prescription Patterns of Cardioprotective Medications in Postacute Myocardial Infarction Patients in Indiana(Elsevier, 2024-09-06) Gunderman, David; Kumar, Ashish; Munguia-Vazquez, Raymundo; Vora, Keyur P.; Shah, Chirag D.; Dharmakumar, Rohan; Kalra, Ankur; Medicine, School of MedicineItem Enabling Reliable Visual Detection of Chronic Myocardial Infarction with Native T1 Cardiac MRI Using Data-Driven Native Contrast Mapping(Radiological Society of North America, 2024) Youssef, Khalid; Zhang, Xinheng; Yoosefian, Ghazal; Chen, Yinyin; Chan, Shing Fai; Yang, Hsin-Jung; Vora, Keyur; Howarth, Andrew; Kumar, Andreas; Sharif, Behzad; Dharmakumar, Rohan; Medicine, School of MedicinePurpose: To investigate whether infarct-to-remote myocardial contrast can be optimized by replacing generic fitting algorithms used to obtain native T1 maps with a data-driven machine learning pixel-wise approach in chronic reperfused infarct in a canine model. Materials and Methods: A controlled large animal model (24 canines, equal male and female animals) of chronic myocardial infarction with histologic evidence of heterogeneous infarct tissue composition was studied. Unsupervised clustering techniques using self-organizing maps and t-distributed stochastic neighbor embedding were used to analyze and visualize native T1-weighted pixel-intensity patterns. Deep neural network models were trained to map pixel-intensity patterns from native T1-weighted image series to corresponding pixels on late gadolinium enhancement (LGE) images, yielding visually enhanced noncontrast maps, a process referred to as data-driven native mapping (DNM). Pearson correlation coefficients and Bland-Altman analyses were used to compare findings from the DNM approach against standard T1 maps. Results: Native T1-weighted images exhibited distinct pixel-intensity patterns between infarcted and remote territories. Granular pattern visualization revealed higher infarct-to-remote cluster separability with LGE labeling as compared with native T1 maps. Apparent contrast-to-noise ratio from DNM (mean, 15.01 ± 2.88 [SD]) was significantly different from native T1 maps (5.64 ± 1.58; P < .001) but similar to LGE contrast-to-noise ratio (15.51 ± 2.43; P = .40). Infarcted areas based on LGE were more strongly correlated with DNM compared with native T1 maps (R2 = 0.71 for native T1 maps vs LGE; R2 = 0.85 for DNM vs LGE; P < .001). Conclusion: Native T1-weighted pixels carry information that can be extracted with the proposed DNM approach to maximize image contrast between infarct and remote territories for enhanced visualization of chronic infarct territories.Item Impact of Intramyocardial Hemorrhage on Clinical Outcomes in ST-Elevation Myocardial Infarction: A Systematic Review and Meta-analysis(Elsevier, 2022-08-26) Vyas, Rohit; Changal, Khalid H.; Bhuta, Sapan; Pasadyn, Vanessa; Katterle, Konrad; Niedoba, Matthew J.; Vora, Keyur; Dharmakumar, Rohan; Gupta, Rajesh; Medicine, School of MedicineBackground: Intramyocardial hemorrhage (IMH) occurs after ST-elevation myocardial infarction (STEMI) and has been documented using cardiac magnetic resonance imaging. The prevalence and prognostic significance of IMH are not well described, and the small sample size has limited prior studies. Methods: We performed a comprehensive literature search of multiple databases to identify studies that compared outcomes in STEMI patients with or without IMH. The outcomes studied were major adverse cardiovascular events (MACE), infarct size, thrombolysis in myocardial infarction (TIMI) flow after percutaneous coronary intervention (PCI), left ventricular end-diastolic volume (LVEDV), left ventricular ejection fraction (LVEF), and mortality. Odds ratios (ORs) and standardized mean differences with corresponding 95% CIs were calculated using a random effects model. Results: Eighteen studies, including 2824 patients who experienced STEMI (1078 with IMH and 1746 without IMH), were included. The average prevalence of IMH was 39%. There is a significant association between IMH and subsequent MACE (OR, 2.63; 95% CI, 1.79-3.86; P < .00001), as well as IMH and TIMI grade <3 after PCI (OR, 1.75; 95% CI, 1.14-2.68; P = .05). We also found a significant association between IMH and the use of glycoprotein IIb/IIIa inhibitors (OR, 2.34; 95% CI, 1.42-3.85; P = .0008). IMH has a positive association with infarct size (standardized mean difference, 2.19; 95% CI, 1.53-2.86; P < .00001) and LVEDV (standardized mean difference, 0.7; 95% CI, 0.41-0.99; P < .00001) and a negative association with LVEF (standardized mean difference, -0.89; 95% CI, -1.15 to -0.63; P = .01). Predictors of IMH include male sex, smoking, and left anterior descending infarct. Conclusions: Intramyocardial hemorrhage is prevalent in approximately 40% of patients who experience STEMI. IMH is a significant predictor of MACE and is associated with larger infarct size, higher LVEDV, and lower LVEF after STEMI.