Radiation therapy (RT) plays an integral role in treating thoracic cancers, despite the risk of radiation-induced cardiotoxicity. We hypothesize that our newly developed magnetic resonance imaging (MRI)-based contractility index (ContractiX) is a sensitive marker for early detection of RT-induced cardiotoxicity in a preclinical rat model of thoracic cancer RT. Adult salt-sensitive rats received image-guided heart RT and were imaged with MRI at 8 weeks and 10 weeks post-RT or sham. The MRI exam included cine and tagging sequences to measure left-ventricular ejection fraction (LVEF), mass, myocardial strain, and ContractiX. Furthermore, ventricular torsion, diastolic strain rate, and mechanical dyssynchrony were measured. Statistical analyses were performed between the sham, 8 weeks post-RT, and 10 weeks post-RT MRI parameters. The results showed that both LVEF and myocardial mass increased post-RT. Peak systolic strain and ContractiX significantly decreased post-RT, with a more relative reduction in ContractiX compared to strain. ContractiX showed an inverse nonlinear relationship with LVEF and continuously decreased with time post-RT. While early diastolic strain rate and mechanical dyssynchrony significantly changed post-RT, ventricular torsion changes were not significant post-RT. In conclusion, ContractiX measured via non-contrast MRI is a sensitive early marker for the detection of subclinical cardiac dysfunction post-RT, and it is superior to other MRI cardiac measures.

Radiation therapy (RT) is widely used to treat thoracic cancers but carries a risk of radiation-induced heart disease (RIHD). This study aimed to detect early markers of RIHD using machine learning (ML) techniques and cardiac MRI in a rat model. SS.BN3 consomic rats, which have a more subtle RIHD phenotype compared to Dahl salt-sensitive (SS) rats, were treated with localized cardiac RT or sham at 10 weeks of age. Cardiac MRI was performed 8 and 10 weeks post-treatment to assess global and regional cardiac function. ML algorithms were applied to differentiate sham-treated and irradiated rats based on early changes in myocardial function. Despite normal global left ventricular ejection fraction in both groups, strain analysis showed significant reductions in the anteroseptal and anterolateral segments of irradiated rats. Gradient boosting achieved an F1 score of 0.94 and an ROC value of 0.95, while random forest showed an accuracy of 88%. These findings suggest that ML, combined with cardiac MRI, can effectively detect early preclinical changes in RIHD, particularly alterations in regional myocardial contractility, highlighting the potential of these techniques for early detection and monitoring of radiation-induced cardiac dysfunction.

Mechanical Thrombectomy (MT) is the standard of care in the interventional management of Acute Ischemic Stroke (AIS). The NVQI-QOD registry records detailed patient characteristics, pre-operative imaging, procedure metrics, and post-operative outcomes of neurointerventional surgical procedures. Although these data are highly informative, there is substantial uncertainty in all medical interventions, so patient outcomes remain variable after intervention. In this work, we leverage a probabilistic machine learning paradigm to predict MT outcomes in the context of this inherent uncertainty. Using data from the NVQI-QOD AIS registry, we identified three groups of feature variables: those available prior to MT (Group Preop), post MT (Group Postop), and at discharge (Group DC). Using Probabilistic Neural Networks (PNNs) and XGBoost, we predicted 1) the change in NIH Stroke Scale from presentation to discharge (∆NIHSS), and 2) a binary measure of functional outcome, which was aggregated from the 90-day follow-up Modified Rankin Scale (mRS). Both XGBoost and the PNN are capable of binary probabilistic classification of mRS scores, with accuracies ranging from 0.69 using preoperative feature variables to 0.80 when utilizing input features that are available at the time of discharge. XGBoost and the PNN had similar mean squared error performance for the ∆NIHSS regression task as well, however the PNN can also perform probabilistic regression, predicting distributions of ∆NIHSS with means and standard deviations (SDs). Feature importance analysis showed that predictions of both ∆NIHSS and mRS severity primarily depended upon the presenting NIHSS, Pre mRS, and patient age. The probabilistic machine learning paradigm allows for quantification of predictive uncertainty through outcome probability distributions and may offer clinicians critical insights beyond traditional deterministic methods. With the clinical information available prior to MT, patients with the worst predicted outcomes will have a nearly ∼50% chance of neurological improvement, while those with the best anticipated outcomes have a > 98% probability for improvement, reinforcing both the safety and profound benefits of MT.

Radiation therapy (RT) plays a vital role in managing thoracic cancers, though it can lead to adverse effects, including significant cardiotoxicity. Understanding the risk factors like hypertension in RT is important for patient prognosis and management. A Dahl salt-sensitive (SS) female rat model was used to study hypertension effect on RT-induced cardiotoxicity. Rats were fed a high-salt diet to induce hypertension and then divided into RT and sham groups. The RT group received 24 Gy of whole-heart irradiation. Cardiac function was evaluated using MRI and blood pressure measurements at baseline, 8 weeks and 12 weeks post-RT. Histological examination was performed after the last timepoint or animal death. The hypertensive RT rats demonstrated significant decreases in left-ventricular ejection fraction (EF) (45 ± 7.2%) compared to sham (68 ± 7.3%). Furthermore, circumferential (Ecc) and radial (Err) myocardial strains were significantly reduced (Ecc: −7.4 ± 2.0% RT rats vs. −11 ± 2.4% sham; Err: 15 ± 6.5% RT rats vs. 23 ± 8.9% sham). Histological analysis revealed significant pathophysiological remodeling post-RT, including nuclear size, interstitial fibrosis, necrosis, and the presence of inflammatory cells. This study provides valuable insights into the cardiotoxic effects of RT in the context of hypertension, highlighting the potential of using MRI for improved risk assessment with potential for future clinical translation.

Background/Objectives: Magnetic resonance imaging (MRI) provides a non-invasive means for a comprehensive assessment of the effect of radiation therapy (RT) on heart function. This study aims to determine RT induced cardiotoxicity in thoracic cancer patients using cardiac MRI. Methods: Cardiac MRI was performed at baseline and at six months post-treatment in patients undergoing standard-of-care RT for lung or esophageal cancers at a single institution. Parameters included regional myocardial strain in the longitudinal, circumferential, and radial directions as well as myocardium T1, T2, and extracellular-volume (ECV) maps. Cardiac segmental doses were extracted from the RT planning scans. The relationship between changes in segmental MRI parameters at six months and segmental heart RT dose were investigated. Results: Twelve patients underwent baseline MRI and four completed the follow-up MRI. Five of the segmental strain parameters showed notable changes between baseline and six-month follow-up. Increased doses in the heart base and apex were associated with moderate-to-large and mild deteriorations, respectively, in strain for all regions. Increased doses in the mid-ventricular regions were associated with improved strain in all regions. The segmental analysis revealed that myocardial regions nurtured by the left coronary artery are more negatively affected by radiation compared to those nurtured by the right coronary artery. Conclusions: Alterations in regional tissue and strain parameters on MRI vary according to local myocardial RT dose, suggesting there may be heterogeneity of radiation sensitivity for the heart substructures and regions. Changes in segmental strain parameters may reflect post-RT cardiac remodeling, but larger confirmatory studies are required.

To develop and assess MyoNet, a deep learning (DL)-based network for measuring myocardial regional function from cine cardiac magnetic resonance (CMR) images, and compare its efficacy with ResMyoNet as an efficient alternative to SinMod-derived reference. MyoNet was tested alongside ResMyoNet on datasets from Dahl salt-sensitive rat models undergoing radiation therapy (RT). Both networks were designed to extract displacement maps from cine images, were specifically optimized for detailed myocardial deformation, employed advanced convolution operations with alternating kernel sizes for spatial and temporal analysis, and robust loss functions. MyoNet demonstrated superior performance in myocardial strain measurement, achieving high consistency with the SinMod-derived reference strains. It outperformed ResMyoNet, achieving higher performance metrics, including SSIM of 0.961 and 0.960, ICC of 0.973 and 0.975, and Pearson CC of 0.973 and 0.953 for circumferential (Ecc) and radial (Err) strains, respectively. Its accuracy and efficiency in generating strain measurements were validated through comprehensive statistical analyses. MyoNet offers a significant advancement in myocardial strain analysis from cine CMR images, potentially revolutionizing cardiac imaging in pre-clinical studies. Its ability to provide detailed and reliable measurements positions it as a valuable tool for clinical applications, particularly in monitoring the cardiac health of cancer patients.