Patient-Specific Tissue Viability Prediction from Single-Scan Vessel Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing imaging techniques for assessing ischemia are costly, expose patients to radiation, and may not be available at all facilities, necessitating a more efficient and accessible method to estimate tissue viability using patient-specific vascular and anatomical models.
Innovation Solution
A system and method utilizing patient-specific physiological parameters and machine learning to estimate tissue viability from a single scan, incorporating patient-specific vascular and anatomical models, enabling accurate assessment of tissue anatomy and viability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specialized imaging techniques (CEMRI, FDG-PET, stress echo/MRI, multidetector CT, dual energy CT) are used to assess ischemia, then measurement precision of tissue viability is improved, but financial cost and radiation exposure increase
Solution Approach 1:
The patent creates a computational model that replicates the functionality of expensive specialized imaging techniques. By using machine learning algorithms trained on data from techniques like FDG-PET and CEMRI, the system generates virtual images and viability assessments that copy the diagnostic value of these gold-standard methods without requiring the actual expensive equipment or radiation exposure.
Solution Approach 2:
The patent replaces physical imaging mechanisms (radiation-based PET, magnetic resonance, ultrasound) with a computational information processing system. The machine learning model processes conventional angiographic images and physiological parameters to substitute for the mechanical and physical systems of specialized imaging equipment, eliminating radiation exposure while maintaining diagnostic capability.
2Measurement precision
If specialized imaging techniques are used to assess ischemia, then measurement precision of tissue viability is improved, but financial cost increases
Solution Approach 1:
The system creates computational replicas of expensive imaging studies. By training machine learning models on datasets from specialized imaging techniques and then using these models to generate viability assessments from常规 angiography, the patent copies the diagnostic value of expensive studies at minimal computational cost, eliminating the need to repeat costly imaging procedures.
Solution Approach 2:
The patent uses inexpensive conventional angiographic images and readily available physiological parameters as input data, replacing the need for expensive specialized imaging studies. The computational model processes these cheap, easily obtainable data sources to produce assessments that would otherwise require costly imaging, making the process economically viable for routine use.
3Measurement precision
If multiple imaging techniques are used to derive vascular and anatomical models, then measurement precision of tissue viability is improved, but device complexity increases
Solution Approach 1:
The patent merges multiple data sources (conventional angiographic images, physiological parameters, anatomical information) into a single integrated machine learning model. This unified computational approach combines the information that would otherwise require separate imaging studies, processing everything through one system to generate comprehensive tissue viability assessments without needing multiple specialized devices.
Solution Approach 2:
The machine learning model serves multiple functions: it processes conventional angiographic images, integrates physiological parameters, generates tissue viability maps, and provides diagnostic assessments all in one system. This universal computational platform replaces the need for multiple specialized imaging devices, each designed for a single specific function, with one multi-functional intelligent system.
Data Source
AI summary
Systems and methods are disclosed for using patient-specific anatomical models and physiological parameters to predict viability of a target tissue or vessel to guide diagnosis or treatment of cardiovascular disease. One method includes: receiving a patient-specific vessel model and a patient-specific tissue model of a patient anatomy; receiving one or more patient-specific physiological parameters (e.g. blood flow, anatomical characteristics, etc.) for one or more physiological states; estimating a viability characteristic of the patient-specific tissue or vessel model (e.g., via a trained machine learning algorithm), using the patient-specific physiological parameters; and outputting the viability characteristic to an electronic storage medium or display.