Cerebral Blood Flow Prediction From Point-Cloud Vessel Morphology

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for computing cerebral blood flow require generating surface or solid models from brain shapes, which is inefficient and inconvenient when new data is acquired, necessitating the development of a more efficient method.

Innovation Solution

A neural network model is trained using a training dataset comprising morphological data in the form of a point cloud or a one-dimensional network, boundary information, and initial condition information to compute cerebral blood flow data without the need for surface or solid models.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If existing cerebral blood flow computation methods are used, then cerebral blood flow can be computed, but it requires generating surface or solid models which is time-consuming and inefficient

Engineering Contradiction:

Improvecomputation efficiencyVSAvoidtime for model generation

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces the traditional mechanical/mathematical model generation process (surface modeling, solid modeling) with a neural network-based computational system. The neural network is trained on medical image data and directly computes cerebral blood flow without requiring intermediate surface or solid model generation, thus substituting the complex mechanical modeling process with a learned computational approach.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent performs preliminary training of the neural network model using a training dataset that includes medical images and corresponding cerebral blood flow data. This preliminary action prepares the model in advance to directly compute blood flow from new medical images without requiring time-consuming model generation during actual computation.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If new data is acquired, then new cerebral blood flow computation is needed, but existing methods require re-computation using complex software and mathematical techniques

Engineering Contradiction:

Improveability to process new dataVSAvoidcomputation software complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces complex computation software and mathematical techniques with a trained neural network model. The neural network, once trained, can process new medical image data directly through forward propagation, eliminating the need for complex software pipelines and mathematical computations for each new dataset.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates a computational model (neural network) that learns from training data and then copies the learned patterns to predict cerebral blood flow in new, unseen medical images. This allows the system to adapt to new data by applying the learned computational patterns without requiring the same complex processing steps as traditional methods.

Inventive Principle:

Principle #26Copying

3Ease of operation

If surface or solid models are generated, then cerebral blood flow computation can be performed, but the process is inconvenient and requires specialized modeling steps

Engineering Contradiction:

Improveconvenience of computation processVSAvoidmodeling process complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent substitutes the inconvenient surface and solid modeling steps with a neural network-based direct computation approach. The neural network takes medical image data as input and directly outputs cerebral blood flow measurements, eliminating the need for specialized modeling operations and making the process more convenient and accessible.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20260044954A1Method for computing cerebral blood flow data and method for training neural network model for computing cerebral blood flow data

Publication Date: 2026.02.12 NEAR BRAIN INC

AI summary

A method for computing cerebral blood flow data according to an embodiment of the present disclosure includes the steps of: acquiring a trained cerebral blood flow prediction model; acquiring an original medical image; acquiring morphological data corresponding to the cerebral blood vessel region from the original medical image, the morphological data being composed of data in the form of a point cloud; acquiring a boundary area of the cerebral blood vessel region and acquiring boundary information corresponding to the boundary area; acquiring initial condition information or boundary condition information; and inputting the morphological data, the boundary information, and the initial condition information or boundary condition information into the cerebral blood flow prediction model, and acquiring cerebral blood flow data related to the speed or pressure of the cerebral blood flow output through the cerebral blood flow prediction model.