Machine learning-aided exploration of relationship between strength and elastic properties in ascending thoracic aneurysm

A Comparative Study of Machine Learning and Algorithmic Approaches to Automatically Identify the Yield Point in Normal and Aneurysmal Human Aortic Tissues

The stress-strain curve of biological soft tissues helps characterize their mechanical behavior. The yield point on this curve is when a specimen breaches its elastic range due to irreversible microstructural damage. The yield point is easily found using the offset yield method in traditional engineering materials. However, correctly identifying the yield point in soft tissues can be subjective due to its nonlinear material behavior. The typical method for yield point identification is visual inspection, which is investigator-dependent and does not lend itself to automation of the analysis pipeline. An automated algorithm to identify the yield point objectively assesses soft tissues' biomechanical properties. This study aimed to analyze data from uniaxial extension testing on biological soft tissue specimens and create a machine learning (ML) model to determine a tissue sample's yield point. We present a trained machine learning model from 279 uniaxial extension curves from testing aneurysmal/nonaneurysmal and longitudinal/circumferential oriented tissue specimens that multiple experts labeled through an adjudication process. The ML model showed a median error of 5% in its estimated yield stress compared to the expert picks. The study found that an ML model could accurately identify the yield point (as defined) in various aortic tissues. Future studies will be performed to validate this approach by visually inspecting when damage occurs and adjusting the model using the ML-based approach.

A classification approach to improve out of sample predictability of structure-based constitutive models for ascending thoracic aortic tissue

In this research, a pipeline was developed to assess the out-of-sample predictive capability of structure-based constitutive models of ascending aortic aneurysmal tissue. The hypothesis being tested is that a biomarker can help establish similarities among tissues sharing the same level of a quantifiable property, thus enabling the development of biomarker-specific constitutive models. Biomarker-specific averaged material models were constructed from biaxial mechanical tests of specimens that shared similar biomarker properties such as level of blood-wall shear stress or microfiber (elastin or collagen) degradation in the extracellular matrix. Using a cross-validation strategy commonly used in classification algorithms, biomarker-specific averaged material models were assessed in contrast to individual tissue mechanics of out of sample specimens that fell under the same category but did not contribute to the averaged model's generation. The normalized root means square errors (NRMSE) calculated on out-of-sample data were compared with average models when no categorization was performed versus biomarker-specific models and among different level of a biomarker. Different biomarker levels exhibited statistically different NRMSE when compared among each other, indicating more common features shared by the specimens belonging to the lower error groups. However, no specific biomarkers reached a significant difference when compared to the average model created when No Categorization was performed, possibly on account of unbalanced number of specimens. The method developed could allow for the screening of different biomarkers or combinations/interactions in a systematic manner leading the way to larger datasets and to more individualized constitutive approaches.

Modeling planar response of vascular tissues using quadratic functions of effective strain

Simulation-based studies of the cardiovascular structure such as aorta have become increasingly popular for many biomedical problems such as predictions of aneurysm rupture. A critical step in these simulations is the development of constitutive models that accurately describe the tissue's mechanical behavior. In this work, we present a new constitutive model, which explicitly accounts for the gradual recruitment of collagen fibers. The recruitment is considered using an effective stretch, which is a continuum-scale kinematic variable measuring the uncrimped stretch of the tissue in an average sense. The strain energy of a fiber bundle is described by a quadratic function of the effective strain. Constitutive models formulated in this manner are applied to describe the responses of ascending thoracic aortic aneurysm and porcine thoracic aorta tissues. The heterogeneous properties of the ATAA tissue are extracted from bulge inflation test data, and then used in finite element analysis to simulate the inflation test. The descriptive and predictive capabilities are further assessed using planar testing data of porcine thoracic aortic tissues. It is found that the constitutive model can accurately describe the stress-strain relations. In particular, the finite element simulation replicates the displacement, strain, and stress distributions with excellent fidelity.

Interplay between Artificial Intelligence and Biomechanics Modeling in the Cardiovascular Disease Prediction

Cardiovascular disease (CVD) is the most common cause of morbidity and mortality worldwide, and early accurate diagnosis is the key point for improving and optimizing the prognosis of CVD. Recent progress in artificial intelligence (AI), especially machine learning (ML) technology, makes it possible to predict CVD. In this review, we first briefly introduced the overview development of artificial intelligence. Then we summarized some ML applications in cardiovascular diseases, including ML-based models to directly predict CVD based on risk factors or medical imaging findings and the ML-based hemodynamics with vascular geometries, equations, and methods for indirect assessment of CVD. We also discussed case studies where ML could be used as the surrogate for computational fluid dynamics in data-driven models and physics-driven models. ML models could be a surrogate for computational fluid dynamics, accelerate the process of disease prediction, and reduce manual intervention. Lastly, we briefly summarized the research difficulties and prospected the future development of AI technology in cardiovascular diseases.

Deep Learning for Computational Hemodynamics: A Brief Review of Recent Advances

Computational fluid dynamics (CFD) modeling of blood flow plays an important role in better understanding various medical conditions, designing more effective drug delivery systems, and developing novel diagnostic methods and treatments. However, despite significant advances in computational technology and resources, the expensive computational cost of these simulations still hinders their transformation from a research interest to a clinical tool. This bottleneck is even more severe for image-based, patient-specific CFD simulations with realistic boundary conditions and complex computational domains, which make such simulations excessively expensive. To address this issue, deep learning approaches have been recently explored to accelerate computational hemodynamics simulations. In this study, we review recent efforts to integrate deep learning with CFD and discuss the applications of this approach in solving hemodynamics problems, such as blood flow behavior in aorta and cerebral arteries. We also discuss potential future directions in the field. In this review, we suggest that incorporating physiologic understandings and underlying fluid mechanics laws in deep learning models will soon lead to a paradigm shift in the development novel non-invasive computational medical decisions.

Incorporating fiber recruitment in hyperelastic modeling of vascular tissues by means of kinematic average

We present a framework for considering the gradual recruitment of collagen fibers in hyperelastic constitutive modeling. An effective stretch, which is a response variable representing the true stretch at the tissue-scale, is introduced. Properties of the effective stretch are discussed in detail. The effective stretch and strain invariants derived from it are used in selected hyperelastic constitutive models to describe the tissue response. This construction is investigated in conjunction with Holzapfel-Gasser-Ogden family strain energy functions. The ensuing models are validated against a large body of uniaxial and bi-axial stress-strain response data from human aortic aneurysm tissues. Both the descriptive and the predictive capabilities are examined. The former is evaluated by the quality of constitutive fitting, and the latter is assessed using finite element simulation. The models significantly improve the quality of fitting, and reproduce the experiment displacement, stress, and strain distributions with high fidelity in the finite element simulation.

On the identification of thyroid nodules using semi-supervised deep learning

Detecting malign cases from thyroid nodule examinations is crucial in healthcare particularly to improve the early detection of such cases. However, malign thyroid nodules can be extremely rare and is hard to find using the traditional rule based or expert-based methods. For this reason, the solutions backed by Machine Learning (ML) algorithms are key to improve the detection rates of such rare cases. In this paper, we investigate the application of ML in the healthcare domain for the detection of rare thyroid nodules. The utilized dataset is collected from 636 distinct patients in 99 unique days in Turkey. In addition to the texture feature data of the Ultrasound (US), we have also included the scores of different assessment methods created by different health institutions (e.g., Korean, American and European thyroid societies) as additional features. For detection of extremely rare malign cases, we use auto-encoder based neural network model. Through numerical results, it is shown that the auto-encoder based model can result in an average Recall score of 0.98 and a Sensitivity score of 1.00 for detecting malign and non-malign cases from the healthcare dataset outperforming the traditional classification algorithms that are trained after Synthetic Minority Oversampling Technique (SMOTE) oversampling.

Estimating aortic thoracic aneurysm rupture risk using tension-strain data in physiological pressure range: an in vitro study

Previous studies have shown that the rupture properties of an ascending thoracic aortic aneurysm (ATAA) are strongly correlated with the pre-rupture response features. In this work, we present a two-step machine learning method to predict where the rupture is likely to occur in ATAA and what safety reserve the structure may have. The study was carried out using ATAA specimens from 15 patients who underwent surgical intervention. Through inflation test, full-field deformation data and post-rupture images were collected, from which the wall tension and surface strain distributions were computed. The tension-strain data in the pressure range of 9-18 kPa were fitted to a third-order polynomial to characterize the response properties. It is hypothesized that the region where rupture is prone to initiate is associated with a high level of tension buildup. A machine learning method is devised to predict the peak risk region. The predicted regions were found to match the actual rupture sites in 13 samples out of the total 15. In the second step, another machine learning model is utilized to predict the tissue's rupture strength in the peak risk region. Results suggest that the ATAA rupture risk can be reasonably predicted using tension-strain response in the physiological range. This may open a pathway for evaluating the ATAA rupture propensity using information of in vivo response.

Prediction of local strength of ascending thoracic aortic aneurysms

Knowledges of both local stress and strength are needed for a reliable evaluation of the rupture risk for ascending thoracic aortic aneurysm (ATAA). In this study, machine learning is applied to predict the local strength of ATAA tissues based on tension-strain data collected through in vitro inflation tests on tissue samples. Inputs to machine learning models are tension, strain, slope, and curvature values at two points on the low strain region of the tension-strain curve. The models are trained using data from locations where the tissue ruptured, and subsequently applied to data from intact sites to predict the local rupture strength. The predicted strengths are compared with the known strength at rupture sites as well as the highest tension the tissues experienced at the intact sites. A local rupture index, which is the ratio of the end tension to the predicted rupture strength, is computed. The 'hot spots' of the rupture index are found to match the rupture sites better than those of the peak tension. The study suggests that the strength of ATAA tissue could be reliably predicted from early phase response features defined in this work.

A generic physics-informed neural network-based constitutive model for soft biological tissues

Constitutive modeling is a cornerstone for stress analysis of mechanical behaviors of biological soft tissues. Recently, it has been shown that machine learning (ML) techniques, trained by supervised learning, are powerful in building a direct linkage between input and output, which can be the strain and stress relation in constitutive modeling. In this study, we developed a novel generic physics-informed neural network material (NNMat) model which employs a hierarchical learning strategy by following the steps: (1) establishing constitutive laws to describe general characteristic behaviors of a class of materials; (2) determining constitutive parameters for an individual subject. A novel neural network structure was proposed which has two sets of parameters: (1) a class parameter set for characterizing the general elastic properties; and (2) a subject parameter set (three parameters) for describing individual material response. The trained NNMat model may be directly adopted for a different subject without re-training the class parameters, and only the subject parameters are considered as constitutive parameters. Skip connections are utilized in the neural network to facilitate hierarchical learning. A convexity constraint was imposed to the NNMat model to ensure that the constitutive model is physically relevant. The NNMat model was trained, cross-validated and tested using biaxial testing data of 63 ascending thoracic aortic aneurysm tissue samples, which was compared to expert-constructed models (Holzapfel-Gasser-Ogden, Gasser-Ogden-Holzapfel, and four-fiber families) using the same fitting and testing procedure. Our results demonstrated that the NNMat model has a significantly better performance in both fitting (R² value of 0.9632 vs 0.9019, p=0.0053) and testing (R² value of 0.9471 vs 0.8556, p=0.0203) than the Holzapfel-Gasser-Ogden model. The proposed NNMat model provides a convenient and general methodology for constitutive modeling.

Augmented patient-specific functional medical imaging by implicit manifold learning

This paper uses machine learning to enrich magnetic resonance angiography and magnetic resonance imaging acquisitions. A convolutional neural network is built and trained over a synthetic database linking geometrical parameters and mechanical characteristics of the arteries to blood flow rates and pressures in an arterial network. Once properly trained, the resulting neural network can be used in order to predict blood pressure in cerebral arteries noninvasively in nearly real-time. One challenge here is that not all input variables present in the synthetic database are known from patient-specific medical data. To overcome this challenge, a learning technique, which we refer to as implicit manifold learning, is employed: in this view, the input and output data of the neural network are selected based on their availability from medical measurements rather than being defined from the mechanical description of the arterial system. The results show the potential of the method and that machine learning is an alternative to costly ensemble based inversion involving sophisticated fluid structure models.

A feasibility study of deep learning for predicting hemodynamics of human thoracic aorta

Numerical analysis methods including finite element analysis (FEA), computational fluid dynamics (CFD), and fluid-structure interaction (FSI) analysis have been used to study the biomechanics of human tissues and organs, as well as tissue-medical device interactions, and treatment strategies. However, for patient-specific computational analysis, complex procedures are usually required to set-up the models, and long computing time is needed to perform the simulation, preventing fast feedback to clinicians in time-sensitive clinical applications. In this study, by using machine learning techniques, we developed deep neural networks (DNNs) to directly estimate the steady-state distributions of pressure and flow velocity inside the thoracic aorta. After training on hemodynamic data from CFD simulations, the DNNs take as input a shape of the aorta and directly output the hemodynamic distributions in one second. The trained DNNs are capable of predicting the velocity magnitude field with an average error of 1.9608% and the pressure field with an average error of 1.4269%. This study demonstrates the feasibility and great potential of using DNNs as a fast and accurate surrogate model for hemodynamic analysis of large blood vessels.

Identification of in vivo nonlinear anisotropic mechanical properties of ascending thoracic aortic aneurysm from patient-specific CT scans

Accurate identification of in vivo nonlinear, anisotropic mechanical properties of the aortic wall of individual patients remains to be one of the critical challenges in the field of cardiovascular biomechanics. Since only the physiologically loaded states of the aorta are given from in vivo clinical images, inverse approaches, which take into account of the unloaded configuration, are needed for in vivo material parameter identification. Existing inverse methods are computationally expensive, which take days to weeks to complete for a single patient, inhibiting fast feedback for clinicians. Moreover, the current inverse methods have only been evaluated using synthetic data. In this study, we improved our recently developed multi-resolution direct search (MRDS) approach and the computation time cost was reduced to 1~2 hours. Using the improved MRDS approach, we estimated in vivo aortic tissue elastic properties of two ascending thoracic aortic aneurysm (ATAA) patients from pre-operative gated CT scans. For comparison, corresponding surgically-resected aortic wall tissue samples were obtained and subjected to planar biaxial tests. Relatively close matches were achieved for the in vivo-identified and ex vivo-fitted stress-stretch responses. It is hoped that further development of this inverse approach can enable an accurate identification of the in vivo material parameters from in vivo image data.

Estimation of in vivo constitutive parameters of the aortic wall using a machine learning approach

The patient-specific biomechanical analysis of the aorta requires the quantification of the in vivo mechanical properties of individual patients. Current inverse approaches have attempted to estimate the nonlinear, anisotropic material parameters from in vivo image data using certain optimization schemes. However, since such inverse methods are dependent on iterative nonlinear optimization, these methods are highly computation-intensive. A potential paradigm-changing solution to the bottleneck associated with patient-specific computational modeling is to incorporate machine learning (ML) algorithms to expedite the procedure of in vivo material parameter identification. In this paper, we developed an ML-based approach to estimate the material parameters from three-dimensional aorta geometries obtained at two different blood pressure (i.e., systolic and diastolic) levels. The nonlinear relationship between the two loaded shapes and the constitutive parameters are established by an ML-model, which was trained and tested using finite element (FE) simulation datasets. Cross-validations were used to adjust the ML-model structure on a training/validation dataset. The accuracy of the ML-model was examined using a testing dataset.