Physics-informed neural networks (PINNs) for 4D hemodynamics prediction: An investigation of optimal framework based on vascular morphology

Removing 65 Years of Approximation in Rotating Ring Disk Electrode Theory with Physics-Informed Neural Networks

The rotating Ring Disk Electrode (RRDE), since its introduction in 1959 by Frumkin and Nekrasov, has become indispensable with diverse applications in electrochemistry, catalysis, and material science. The collection efficiency (N ) is an important parameter extracted from the ring and disk currents of the RRDE, providing valuable information about reaction mechanism, kinetics, and pathways. The theoretical prediction of N is a challenging task: requiring solution of the complete convective diffusion mass transport equation with complex velocity profiles. Previous efforts, including by Albery and Bruckenstein who developed the most widely used analytical equations, heavily relied on approximations by removing radial diffusion and using approximate velocity profiles. 65 years after the introduction of RRDE, we employ a physics-informed neural network to solve the complete convective diffusion mass transport equation, to reveal the formerly neglected edge effects and velocity corrections on N , and to provide a guideline where conventional approximation is applicable.

Symptom clusters and unplanned hospital readmission in Chinese patients with acute myocardial infarction on admission

Backgroud

Acute myocardial infarction (AMI) has a high morbidity rate, high mortality rate, high readmission rate, high health care costs, and a high symptomatic, psychological, and economic burden on patients. Patients with AMI usually present with multiple symptoms simultaneously, which are manifested as symptom clusters. Symptom clusters have a profound impact on the quality of survival and clinical outcomes of AMI patients.

Objective

The purpose of this study was to analyze unplanned hospital readmissions among cluster groups within a 1-year follow-up period, as well as to identify clusters of acute symptoms and the characteristics associated with them that appeared in patients with AMI.

Methods

Between October 2021 and October 2022, 261 AMI patients in China were individually questioned for symptoms using a structured questionnaire. Mplus 8.3 software was used to conduct latent class analysis in order to find symptom clusters. Univariate analysis is used to examine characteristics associated with each cluster, and multinomial logistic regression is used to analyze a cluster membership as an independent predictor of hospital readmission after 1-year.

Results

Three unique clusters were found among the 11 acute symptoms: the typical chest symptom cluster (64.4%), the multiple symptom cluster (29.5%), and the atypical symptom cluster (6.1%). The cluster of atypical symptoms was more likely to have anemia and the worse values of Killip class compared with other clusters. The results of multiple logistic regression indicated that, in comparison to the typical chest cluster, the atypical symptom cluster substantially predicted a greater probability of 1-year hospital readmission (odd ratio 8.303, 95% confidence interval 2.550-27.031, P < 0.001).

Conclusion

Out of the 11 acute symptoms, we have found three clusters: the typical chest symptom, multiple symptom, and atypical symptom clusters. Compared to patients in the other two clusters, those in the atypical symptom cluster-which included anemia and a large percentage of Killip class patients-had worse clinical indicators at hospital readmission during the duration of the 1-year follow-up. Both anemia and high Killip classification suggest that the patient's clinical presentation is poor and therefore the prognosis is worse. Intensive treatment should be considered for anemia and high level of Killip class patients with atypical presentation. Clinicians should focus on patients with atypical symptom clusters, enhance early recognition of symptoms, and develop targeted symptom management strategies to alleviate their discomfort in order to improve symptomatic outcomes.

Performance of Fourier-based activation function in physics-informed neural networks for patient-specific cardiovascular flows

BACKGROUND AND OBJECTIVES

Physics-informed neural networks (PINNs) can be used to inversely model complex physical systems by encoding the governing partial differential equations and training data into the neural network. However, neural networks are known to be biased towards learning less complex functions, called spectral bias. This has important implications in modeling cardiovascular flows, where spatial frequencies can vary substantially across anatomies and pathologies (e.g., aneurysms or stenoses). Recent evidence suggests that Fourier-based activation functions have desirable properties, and can potentially reduce spectral bias; however, the performance and adequacy of such Fourier activation functions have not yet been evaluated in patient-specific cardiovascular flow applications.

METHODS

The performance of sine activation function was evaluated against tanh and swish activation functions in a 1D advection-diffusion problem, an eccentric 2D stenosis model (Re=5000), and a patient-specific 3D aortic model (Re=823) under pulsatile flow conditions. CFD simulations were performed at high spatio-temporal resolution and data points were extracted for training the neural network. The number of training data points were normalized by L/D. The performance of the PINNs framework was evaluated with increasing number of training data points and across all three activation functions.

RESULTS

Our results demonstrate that sine activation function presents desirable characteristics, such as monotonic reduction in errors, relatively faster convergence, and accurate eigen spectra at higher modes, compared to tanh and swish activation functions. Interestingly, for all activation functions, the domain-averaged errors tended to asymptote at ≈15-20% despite substantial increase in training point density. For 2D eccentric stenosis, errors asymptoted at a sensor point density of 40L/D. For 3D patient-specific aorta, this asymptote was achieved at 180L/D for all three activation functions with an error of ≈15% although sine activation function demonstrated relatively faster convergence.

CONCLUSIONS

We have demonstrated that Fourier-based activation functions have higher performance in terms of accuracy and convergence properties for cardiovascular flow applications; however, inherent challenges of neural networks (e.g., spectral bias) can limit the accuracy to ≈15% under physiological, 3D patient-specific blood flow conditions.

Quantitative analysis of molecular transport in the extracellular space using physics-informed neural network

The brain extracellular space (ECS), an irregular, extremely tortuous nanoscale space located between cells or between cells and blood vessels, is crucial for nerve cell survival. It plays a pivotal role in high-level brain functions such as memory, emotion, and sensation. However, the specific form of molecular transport within the ECS remain elusive. To address this challenge, this paper proposes a novel approach to quantitatively analyze the molecular transport within the ECS by solving an inverse problem derived from the advection-diffusion equation (ADE) using a physics-informed neural network (PINN). PINN provides a streamlined solution to the ADE without the need for intricate mathematical formulations or grid settings. Additionally, the optimization of PINN facilitates the automatic computation of the diffusion coefficient governing long-term molecule transport and the velocity of molecules driven by advection. Consequently, the proposed method allows for the quantitative analysis and identification of the specific pattern of molecular transport within the ECS through the calculation of the Péclet number. Experimental validation on two datasets of magnetic resonance images (MRIs) captured at different time points showcases the effectiveness of the proposed method. Notably, our simulations reveal identical molecular transport patterns between datasets representing rats with tracer injected into the same brain region. These findings highlight the potential of PINN as a promising tool for comprehensively exploring molecular transport within the ECS.

Assessment of intracranial aneurysm rupture risk using a point cloud-based deep learning model

Background and Purpose: Precisely assessing the likelihood of an intracranial aneurysm rupturing is critical for guiding clinical decision-making. The objective of this study is to construct and validate a deep learning framework utilizing point clouds to forecast the likelihood of aneurysm rupturing. Methods: The dataset included in this study consisted of a total of 623 aneurysms, with 211 of them classified as ruptured and 412 as unruptured, which were obtained from two separate projects within the AneuX morphology database. The HUG project, which included 124 ruptured aneurysms and 340 unruptured aneurysms, was used to train and internally validate the model. For external validation, another project named @neurIST was used, which included 87 ruptured and 72 unruptured aneurysms. A standardized method was employed to isolate aneurysms and a segment of their parent vessels from the original 3D vessel models. These models were then converted into a point cloud format using open3d package to facilitate training of the deep learning network. The PointNet++ architecture was utilized to process the models and generate risk scores through a softmax layer. Finally, two models, the dome and cut1 model, were established and then subjected to a comprehensive comparison of statistical indices with the LASSO regression model built by the dataset authors. Results: The cut1 model outperformed the dome model in the 5-fold cross-validation, with the mean AUC values of 0.85 and 0.81, respectively. Furthermore, the cut1 model beat the morphology-based LASSO regression model with an AUC of 0.82. However, as the original dataset authors stated, we observed potential generalizability concerns when applying trained models to datasets with different selection biases. Nevertheless, our method outperformed the LASSO regression model in terms of generalizability, with an AUC of 0.71 versus 0.67. Conclusion: The point cloud, as a 3D visualization technique for intracranial aneurysms, can effectively capture the spatial contour and morphological aspects of aneurysms. More structural features between the aneurysm and its parent vessels can be exposed by keeping a portion of the parent vessels, enhancing the model's performance. The point cloud-based deep learning model exhibited good performance in predicting rupture risk while also facing challenges in generalizability.

Prediction of Hemodynamic-Related Hemolysis in Carotid Stenosis and Aiding in Treatment Planning and Risk Stratification Using Computational Fluid Dynamics

Atherosclerosis affects human health in many ways, leading to disability or premature death due to ischemic heart disease, stroke, or limb ischemia. Poststenotic blood flow disruption may also play an essential role in artery wall impairment linked with hemolysis related to shear stress. The maximum shear stress in the atherosclerotic plaque area is the main parameter determining hemolysis risk. In our work, a 3D internal carotid artery model was built from CT scans performed on patients qualified for percutaneous angioplasty due to its symptomatic stenosis. The obtained stenosis geometries were used to conduct a series of computer simulations to identify critical parameters corresponding to the increase in shear stress in the arteries. Stenosis shape parameters responsible for the increase in shear stress were determined. The effect of changes in the carotid artery size, length, and degree of narrowing on the change in maximum shear stress was demonstrated. Then, a correlation for the quick initial diagnosis of atherosclerotic stenoses regarding the risk of hemolysis was developed. The developed relationship for rapid hemolysis risk assessment uses information from typical non-invasive tests for treated patients. Practical guidelines have been developed regarding which stenosis shape parameters pose a risk of hemolysis, which may be adapted in medical practice.