Artificial Neural Network Predicts Artificial Heart Flow
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing artificial heart pumps, particularly those of the centrifugal type, face issues with blood stagnation near ball bearings, leading to coagulation, and prior flow estimation methods for magnetically levitated pumps are inaccurate due to dynamic forces perturbing the rotor position.
Innovation Solution
A system utilizing an artificial neural network to predict characteristics of artificial hearts, such as fluid flow and differential pressure, by training on input vectors including rotor speed, position, and current, allowing for adjustments to weights based on error rates, thereby improving estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If ball bearings are used to support the rotor, then the structure is simple and cost-effective, but blood flow stagnates near the bearing causing coagulation
Solution Approach 1:
The patent replaces the mechanical ball bearing support system with a magnetic field-based levitation system. Electromagnets generate magnetic fields that levitate and position the rotor without physical contact, eliminating the mechanical contact that causes blood stagnation and coagulation while maintaining structural support.
Solution Approach 2:
The patent introduces magnetic fields as an intermediary between the stator and rotor. The magnetic field acts as a non-contact mediator that provides the necessary support and control forces without requiring direct mechanical contact, thereby preventing blood stagnation in the bearing region.
2Object-affected harmful factors
If magnetic levitation is used to eliminate blood stagnation, then blood coagulation is reduced, but dynamic forces perturb rotor position reducing flow estimation accuracy
Solution Approach 1:
The patent implements a feedback control system using sensors to detect rotor position and an artificial neural network to process this data. The system continuously monitors rotor position perturbations caused by dynamic forces and adjusts electromagnetic control signals to compensate for these deviations, maintaining accurate flow estimation despite magnetic levitation instabilities.
Solution Approach 2:
The patent employs dynamic electromagnetic control to actively adjust the magnetic field in response to changing operating conditions. By continuously regulating the electromagnetic forces based on real-time rotor position feedback, the system adapts to dynamic perturbations and maintains precise rotor positioning for accurate flow measurement.
3Device complexity
If linear equations with least-squared methods are used for flow estimation, then the model is simple, but accuracy is insufficient due to rotor position perturbations
Solution Approach 1:
The patent transforms the flow estimation approach from simple linear parameter fitting to a neural network model that learns complex non-linear relationships between input parameters (voltage, current, frequency, time) and flow characteristics. This parameter transformation enables accurate estimation despite rotor position perturbations by capturing the true non-linear dynamics of the system.
Solution Approach 2:
The patent replaces the mathematical modeling approach (linear equations with least-squared fitting) with an artificial neural network system. This substitution enables the model to automatically learn and adapt to the complex non-linear relationships in the system, providing accurate flow estimation even when rotor position is perturbed by dynamic forces.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system provides accurate estimates of flow rates, differential pressure, and patient conditions, enhancing the performance and hemocompatibility of magnetically levitated artificial hearts by minimizing errors and predicting patient-specific parameters.
Implementation Method 1
an artificial neural network configured to receive an input vector of a predetermined length to train the artificial neural network, produce an output vector based on the input vector, and compare the output vector with a target vector of the predetermined length
Implementation Method 2
artificial heart pumps that include rotors suspended in a non-contacting state by magnetic forces have been used
Implementation Method 3
constantly maintain the rotor in the proper attitude by regulating the current supplied to the electromagnets so as to control their magnetic field
Implementation Method 4
an impeller that rotates integrally with the rotor for imparting centrifugal force to the blood flowing in through the blood flow channel formed in the casing
Data Source
AI summary
A system configured to predict characteristics of an artificial heart is described. The system includes a processor and memory in electronic communication with the processor, and an artificial neural network configured to receive an input vector of a predetermined length to train the artificial neural network, produce an output vector based on the input vector, and compare the output vector with a target vector of the predetermined length. When the output vector does not match the target vector within a predetermined error rate, the network is configured to adjust at least one weight, and when the output vector matches the target vector within the predetermined error rate, the network is configured to execute the input vector to produce an estimate at least one characteristic of the artificial heart.