Biophysics-based statistical learning: Application to heart and brain interactions

Abstract

Initiatives such as the UK Biobank provide joint cardiac and brain imaging information for thousands of individuals, representing a unique opportunity to study the relationship between heart and brain. Most of research on large multimodal databases has been focusing on studying the associations among the available measurements by means of univariate and multivariate association models. However, these approaches do not provide insights about the underlying mechanisms and are often hampered by the lack of prior knowledge on the physiological relationships between measurements. For instance, important indices of the cardiovascular function, such as cardiac contractility, cannot be measured in-vivo. While these non-observable parameters can be estimated by means of biophysical models, their personalisation is generally an ill-posed problem, often lacking critical data and only applied to small datasets. Therefore, to jointly study brain and heart, we propose an approach in which the parameter personalisation of a lumped cardiovascular model is constrained by the statistical relationships observed between model parameters and brain-volumetric indices extracted from imaging, i.e. ventricles or white matter hyperintensities volumes, and clinical information such as age or body surface area. We explored the plausibility of the learnt relationships by inferring the model parameters conditioned on the absence of part of the target clinical features, applying this framework in a cohort of more than 3 000 subjects and in a pathological subgroup of 59 subjects diagnosed with atrial fibrillation. Our

Electroacupuncture attenuates myocardial ischemia-reperfusion injury by inhibiting microglial engulfment of dendritic spines

“…44 According to our findings, abnormal microglial and neuronal activities are probably significant risk factors for MIRI development, and crosstalk between the heart and brain supports this notion. 62 Furthermore, we found that acupuncture can effectively reduce the overexpression of NLRP3 and caspase-1 in the FN and the overactivation of FN microglia during MIRI; FN has direct neuronal projection to hypothalamus. 63 Direct cerebellar-hypothalamic projections play an important role in transmission and integration of visceral regulation through sympathetic nerves.…”

Combined In-silico and Machine Learning Approaches Toward Predicting Arrhythmic Risk in Post-infarction Patients

“…Approaches to be used for augmenting patient populations (the space of the patient-specific, image-derived geometries and concomitant features) include shape modeling approaches (Balaban et al, 2021 ) as well as generative adversarial networks (Gholami et al, 2018 ; Shaker et al, 2020 ). Next steps for this and related work research may include combining multi-organ systems for joint study (e.g., Banus et al, 2021 ), to both better constrain the parameter space of a personalized model and to subsequently capture plausible physiologically mechanisms.…”

“…Many studies have also successfully employed ML in arrhythmia risk stratification, including advanced ML-enabled image analysis (Feeny et al, 2020 ; Krittanawong et al, 2020 ; Trayanova, 2021 ). Recently, ML models have been combined with biophysical modeling to assess risk for dangerous arrhythmia as well as to uncover mechanisms of rhythm disturbances and to manage treatment for affected patients (Prakosa et al, 2013 ; Bernard et al, 2018 ; Lozoya et al, 2019 ; Shade et al, 2020 ; Banus et al, 2021 ; Monaci et al, 2021 ; Sermesant et al, 2021 ; Trayanova, 2021 ). Biophysical cardiac computational modeling and ML have also increasingly been combined to focus on drug-induced proarrhythmic risk assessment, as in e.g., Yang et al ( 2020 ) and Sahli-Costabal et al ( 2020 ).…”

Data_Sheet_1.pdf

“…Approaches to be used for augmenting patient populations (the space of the patient-specific, image-derived geometries and concomitant features) include shape modeling approaches (Balaban et al, 2021) as well as generative adversarial networks (Gholami et al, 2018;Shaker et al, 2020). Next steps for this and related work research may include combining multi-organ systems for joint study (e.g., Banus et al, 2021), to both better constrain the parameter space of a personalized model and to subsequently capture plausible physiologically mechanisms.…”

“…Many studies have also successfully employed ML in arrhythmia risk stratification, including advanced ML-enabled image analysis (Feeny et al, 2020;Krittanawong et al, 2020;Trayanova, 2021). Recently, ML models have been combined with biophysical modeling to assess risk for dangerous arrhythmia as well as to uncover mechanisms of rhythm disturbances and to manage treatment for affected patients (Prakosa et al, 2013;Bernard et al, 2018;Lozoya et al, 2019;Shade et al, 2020;Banus et al, 2021;Monaci et al, 2021;Trayanova, 2021). Biophysical cardiac computational modeling and ML have also increasingly been combined to focus on drug-induced proarrhythmic risk assessment, as in e.g., Yang et al (2020) and Sahli-Costabal et al (2020).…”