arXiv subjects

Hiroshi Suito

Publications and source records attributed to Hiroshi Suito.

7 recordsLinked to original sources

arXiv2025

Multiscale modeling of blood circulation with cerebral autoregulation and network pathway analysis for hemodynamic redistribution in the vascular network with anatomical variations and stenosis conditions

Cerebral hemodynamics is fundamentally regulated by the Circle of Willis (CoW), which redistributes flow through communicating arteries to stabilize perfusion under anatomical variations and vascular stenosis. In this study, we develop a multiscale circulation model by coupling a systemic hemodynamic framework with a cerebral arterial network reconstructed from medical imaging. The model incorporates a cerebral autoregulation mechanism (CAM) and enables quantitative simulation of flow redistribution within the CoW under normal, anatomically varied, and stenotic conditions. Baseline simulations reproduce physiological flow distributions in which communicating arteries remain nearly inactive, showing negligible cross-flow and agreement with clinical measurements. In contrast, anatomical variations reveal distinct collateral activation patterns: the anterior communicating artery (ACoA) emerges as the earliest and most sensitive functional collateral, whereas the posterior communicating arteries (PCoAs) exhibit structure-dependent engagement. Progressive stenosis simulations further demonstrate a transition from a complete CoW to a fetal-type posterior cerebral artery (PCA) configuration, characterized by early ACoA flow reversal followed by ipsilateral PCoA activation, consistent with experimental and transcranial Doppler observations. Finally, a path-based quantitative analysis is introduced to illustrate how the cerebral vascular network dynamically reconfigures collateral pathways in response to structural changes. Overall, the proposed framework provides a physiologically interpretable, image-informed tool for investigating cerebral flow regulation through functional collaterals within the CoW, with potential applications in the diagnosis and treatment planning of cerebrovascular diseases.

physics.med-ph↗

arXiv2025

Ocean wave spectrum reconstruction from HF radar data and its application to wave height estimation

Real-time estimation of ocean wave heights using high-frequency (HF) radar has attracted great attention. This method offers the benefit of easy maintenance by virtue of its ground-based installation. However, it is adversely affected by issues such as low estimation accuracy. As described herein, we propose an algorithm based on the nonnegative sparse regularization method to estimate the energy distribution of the component waves, known as the ocean wave spectrum, from HF radar data. After proving a stability estimate of this algorithm, we perform numerical simulations to verify the proposed method's effectiveness.

math.NA↗

arXiv2024

Cross-sectional shape analysis for risk assessment and prognosis of patients with true lumen narrowing after type-A aortic dissection surgery

Background: For acute type-A aortic dissection (ATAAD) surgery, early post-surgery assessment is crucially important for effective treatment plans, underscoring the need for a framework to identify the risk level of aortic dissection cases. We examined true-lumen narrowing during follow-up examinations, collected morphological data 14 days (early stages) after surgery, and assessed patient risk levels over 2.8 years. Purpose: To establish an implementable framework supported by mathematical techniques to predict the risk of aortic dissection patients experiencing true-lumen narrowing after ATAAD surgery. Materials and Methods: This retrospective study analyzed CT data from 21 ATAAD patients. Forty uniformly distributed cross-sectional shapes (CSSs) are derived from each lumen to account for gradual changes in shape. We introduced the form factor (FF) to assess CSS morphology. Linear discriminant analysis (LDA) is used for the risk classification of aortic dissection patients. Leave-one-patient-out cross-validation (LOPO-CV) is used for risk prediction. Results: For this investigation, we examined data of 21 ATAAD patients categorized into high-risk, medium-risk, and low-risk cases based on clinical observations of the range of true-lumen narrowing. Our risk classification machine-learning (ML) model preserving the model's generalizability. The model's predictions reliably identified low-risk patients, thereby potentially reducing hospital visits. It also demonstrated proficiency in accurately predicting the risk for all high-risk patients. Conclusion: The suggested method anticipates the risk linked to aortic enlargement in patients with a narrowing true lumen in the early stage following ATAAD surgery, thereby aiding follow-up doctors in enhancing patient care.

physics.med-ph↗

arXiv2024

A multicore parallel algorithm for multiscale modelling of an entire human blood circulation network

The presented multi-scale, closed-loop blood circulation model includes arterial, venous, and portal venous systems, heart-pulmonary circulation, and micro-circulation in capillaries. One-dimensional models simulate large blood vessel flow, whereas zerodimensional models are used for simulating blood flow in vascular subsystems corresponding to peripheral arteries and organs. Transmission conditions at bifurcation and confluence are solved using Riemann invariants. Blood circulation simulation in the portal venous system and related organs (liver, stomach, spleen, pancreas, intestine) is particularly targeted. Those organs play important roles in metabolic system dynamics. The proposed efficient parallel algorithms for multicore environments solve these equations much faster than serial computations.

physics.med-ph↗

arXiv2023

Computational analysis for competition flows in arteriovenous fistulas based on non-contrast magnetic resonance imaging

Introduction: Characteristics of hemodynamics strongly affect the patency of arteriovenous fistula (AVF) in hemodialysis patients. Because of pressure balance changes among arteries after AVF construction, regurgitating flow occurs in some patients. Methods: Based on phase-contrast MRI measurements, flow types around the anastomotic site are classified to the three different types of splitting, merging, and one-way, where merging type incorporates regurgitating flow. We have performed computational simulations to analyze characteristic differences among these types. Results: In the merging type, a characteristic spiral flow is observed in AVF causing strong wall shear stress and large pressure drop, whereas the splitting type shows a smooth flow and gives a smaller pressure drop. The one-way case is intermediate between splitting and merging types. Conclusion: Regurgitation brings about high wall shear stress near the anastomotic site because of instabilities induced by merging phenomena, for which type careful follow-up examinations are regarded as necessary.

math.NA↗

arXiv2021

Communication-hiding pipelined BiCGSafe methods for solving large linear systems

Recently, a new variant of the BiCGStab method, known as the pipeline BiCGStab, has been proposed. This method can achieve a higher degree of scalability and speed-up rates through a mechanism in which the communication phase for the computation of the inner product can be overlapped with the computation of the matrix-vector product. On the other hand, there exist several generalized iteration methods with better convergence behavior than BiCGStab such as ssBiCGSafe, BiCGSafe, GPBi-CG. Of these methods, ssBiCGSafe, which requires a single phase of computing inner products per one iteration, is best suited for high-performance computing systems. In this paper, inspired by the success of the pipelined BiCGStab method, we propose variations of the ssBiCGSafe method, in which only one phase of inner product computation per iteration is required and this inner product computation phase can be overlapped with the matrix-vector computation. Through numerical experiments, we show that the proposed methods lead to improvements in convergence behavior and execution time compared to the pipelined BiCGStab and ssBiCGSafe methods.

cs.DC↗

arXiv2020

Artificial intelligence supported anemia control system (AISACS) to prevent anemia in maintenance hemodialysis patients

Anemia, for which erythropoiesis-stimulating agents (ESAs) and iron supplements (ISs) are used as preventive measures, presents important difficulties for hemodialysis patients. Nevertheless, the number of physicians able to manage such medications appropriately is not keeping pace with the rapid increase of hemodialysis patients. Moreover, the high cost of ESAs imposes heavy burdens on medical insurance systems. An artificial-intelligence-supported anemia control system (AISACS) trained using administration direction data from experienced physicians has been developed by the authors. For the system, appropriate data selection and rectification techniques play important roles. Decision making related to ESAs poses a multi-class classification problem for which a two-step classification technique is introduced. Several validations have demonstrated that AISACS exhibits high performance with correct classification rates of 72-87% and clinically appropriate classification rates of 92-98%.

cs.LG↗