Pulse transit time estimation of aortic pulse wave velocity and blood pressure using machine learning and simulated training data.

Recent developments in cardiovascular modelling allow us to simulate blood flow in an entire human body. Such model can also be used to create databases of virtual subjects, with sizes limited only by computational resources. In this work, we study if it is possible to estimate cardiovascular health indices using machine learning approaches. In particular, we carry out theoretical assessment of estimating aortic pulse wave velocity, diastolic and systolic blood pressure and stroke volume using pulse transit/arrival timings derived from photopletyshmography signals. For predictions, we train Gaussian process regression using a database of virtual subjects generated with a cardiovascular simulator. Simulated results provides theoretical assessment of accuracy for predictions of the health indices. For instance, aortic pulse wave velocity can be estimated with a high accuracy (r > 0.9) when photopletyshmography is measured from left carotid artery using a combination of foot-to-foot pulse transmit time and peak location derived for the predictions. Similar accuracy can be reached for diastolic blood pressure, but predictions of systolic blood pressure are less accurate (r > 0.75) and the stroke volume predictions are mostly contributed by heart rate.

Deep convolutional neural networks for estimating porous material parameters with ultrasound tomography.

The feasibility of data based machine learning applied to ultrasound tomography is studied to estimate water-saturated porous material parameters. In this work, the data to train the neural networks is simulated by solving wave propagation in coupled poroviscoelastic-viscoelastic-acoustic media. As the forward model, a high-order discontinuous Galerkin method is considered, while deep convolutional neural networks are used to solve the parameter estimation problem. In the numerical experiment, the material porosity and tortuosity is estimated, while the remaining parameters which are of less interest are successfully marginalized in the neural networks-based inversion. Computational examples confirm the feasibility and accuracy of this approach.

MRI contrasts in high rank rotating frames.

PURPOSE: MRI relaxation measurements are performed in the presence of a fictitious magnetic field in the recently described technique known as RAFF (Relaxation Along a Fictitious Field). This method operates in the 2(nd) rotating frame (rank n = 2) by using a nonadiabatic sweep of the radiofrequency effective field to generate the fictitious magnetic field. In the present study, the RAFF method is extended for generating MRI contrasts in rotating frames of ranks 1 ≤ n ≤ 5. The developed method is entitled RAFF in rotating frame of rank n (RAFFn). THEORY AND METHODS: RAFFn pulses were designed to generate fictitious fields that allow locking of magnetization in rotating frames of rank n. Contrast generated with RAFFn was studied using Bloch-McConnell formalism together with experiments on human and rat brains. RESULTS: Tolerance to B0 and B1 inhomogeneities and reduced specific absorption rate with increasing n in RAFFn were demonstrated. Simulations of exchange-induced relaxations revealed enhanced sensitivity of RAFFn to slow exchange. Consistent with such feature, an increased grey/white matter contrast was observed in human and rat brain as n increased. CONCLUSION: RAFFn is a robust and safe rotating frame relaxation method to access slow molecular motions in vivo.

Deep learning for prediction of cardiac indices from photoplethysmographic waveform: A virtual database approach.

Deep learning methods combined with large datasets have recently shown significant progress in solving several medical tasks. However, collecting and annotating large datasets can be a very cumbersome and expensive task. We tackle these problems with a virtual database approach where training data is generated using computer simulations of related phenomena. Specifically, we concentrate on the following problem: can cardiovascular indices such as aortic elasticity, diastolic and systolic blood pressures, and blood flow from heart be predicted continuously using wearable photoplethysmographic sensors? We simulate the blood flow using a haemodynamic model consisting of the entire human circulation. Repeated evaluation of the simulator allows us to create a database of "virtual subjects" with size that is only limited by available computational resources. Using this database, we train neural networks to predict the cardiac indices from photoplethysmographic signal waveform. We consider two approaches: neural networks based on predefined input features and deep convolutional neural networks taking waveform directly as the input. The performance of the methods is demonstrated using numerical examples, thus carrying out a preliminary assessment of the approaches. The results show improvements in accuracy compared with the previous methods. The improvements are especially significant with indices related to aortic elasticity and maximum blood flow. The proposed approach would provide new means to measure cardiovascular health continuously, for example, with a simple wrist device.

Determination of heterogeneous thermal parameters using ultrasound induced heating and MR thermal mapping.

In this paper, a method for the determination of spatially varying thermal conductivity and perfusion coefficients of tissue is proposed. The temperature evolution in tissue is modelled with the Pennes bioheat equation. The main motivation here is a model-based optimal control for ultrasound surgery, in which the tissue properties are needed when the treatment is planned. The overview of the method is as follows. The same ultrasound transducers, which are eventually used for the treatment, are used to inflict small temperature changes in tissue. This temperature evolution is monitored using MR thermal imaging, and the tissue properties are then estimated on the basis of these measurements. Furthermore, an approach to choose transducer excitations for the determination procedure is also considered. The purpose of this paper is to introduce a method and therefore simulations are used to verify the method. Furthermore, computations are accomplished in a 2D spatial domain.