Mechanical Ventilator Configuration from Similar-Patient Models Without CT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mechanical ventilator settings are determined based on patient body size, which can lead to Ventilator Induced Lung Injury (VILI) due to lung heterogeneity, especially in patients with conditions like COPD or pneumonia, as existing methods rely on non-patient-specific CT imaging models.
Innovation Solution
A system that generates patient-specific mechanical ventilation models without requiring CT scans by using deformable image registration with X-ray or ultrasound data, adjusting similar patient data to create a digital twin for personalized ventilator settings, and dynamically updates the model with bedside imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CT imaging is used to construct patient-specific lung models, then measurement precision and model accuracy are improved, but patient exposure to ionizing radiation increases
Solution Approach 1:
The patent creates a digital copy (virtual lung model) of the patient's lungs using alternative imaging methods like X-ray or ultrasound, then uses this copy for simulation and testing of ventilator settings. This eliminates the need for repeated CT scans while maintaining model accuracy, as the digital twin can be updated with new imaging data when available but doesn't require continuous CT exposure.
Solution Approach 2:
The patent introduces an intermediary approach by using lower-radiation imaging modalities (X-ray, ultrasound) as mediators to capture lung morphology and mechanics data. These intermediaries provide sufficient information to construct accurate lung models without the harmful effects of repeated CT radiation exposure.
2Device complexity
If generic ventilator settings based on body size are used, then device complexity and setup time are reduced, but manufacturing precision and patient-specific accuracy deteriorate
Solution Approach 1:
The patent performs preliminary actions by pre-processing patient imaging data to extract lung morphology and mechanical properties, then using this information to pre-configure the digital lung model and recommended ventilator settings before actual mechanical ventilation begins. This preliminary customization reduces setup complexity during critical moments while maintaining high precision.
Solution Approach 2:
The patent dynamically adjusts ventilator setting parameters based on patient-specific lung model characteristics derived from imaging data. The system changes key parameters like tidal volume, pressure limits, and PEEP levels according to the individual patient's lung compliance and resistance, transitioning from generic to personalized settings while managing complexity through automated calculations.
3Ease of manufacture
If non-patient-specific mechanical properties are used in lung models, then ease of manufacture and model construction are improved, but adaptability and patient-specific accuracy deteriorate
Solution Approach 1:
The patent applies local quality by assigning different mechanical properties to different regions of the lung model based on patient-specific imaging data. Instead of using uniform mechanical properties throughout, the system extracts region-specific compliance and resistance values from imaging and incorporates them into the digital model, allowing heterogeneous lung conditions (like COPD or pneumonia affecting specific areas) to be accurately represented.
Solution Approach 2:
The patent makes the lung model dynamic by allowing mechanical properties to change and be updated based on new imaging data or clinical observations. The model can adapt to changing patient conditions over time, transitioning from static generic properties to dynamic patient-specific properties that evolve with the patient's clinical status.
Data Source
AI summary
A non-transitory storage medium stores instructions readable and executable by at least one electronic processor to receive clinical data for a current patient (P); search a database of CT scans and/or patient-specific mechanical ventilation models for other patients using the clinical data for the current patient as a search criterion to identify a similar patient in the database and similar patient data (S) comprising a CT scan and/or a patient-specific mechanical ventilation model for the similar patient; determine a patient-specific mechanical ventilation model for the current patient based on the CT scan and/or patient-specific mechanical ventilation model for the similar patient; generate ventilator configuration data for mechanically ventilating the current patient based on the determined patient-specific mechanical ventilation model for the current patient.