Internal Medical Device Detection from 2D Images Using 3D Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 2D medical imaging technologies inadequately represent the position and orientation of invasive medical devices, leading to increased procedure duration and patient radiation exposure.
Innovation Solution
A system utilizing machine learning algorithms to infer 3D position and orientation of medical devices from 2D images, enhancing existing imaging devices with 3D information without additional hardware, by training models on annotated images to predict device orientation and position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If 2D imaging technologies are used to visualize medical devices, then device visualization is provided, but accurate representation of device position and orientation is lost
Solution Approach 1:
The patent creates a digital copy (virtual model) of the medical device that mirrors its physical properties and appearance. This virtual model is then rendered from multiple angles and perspectives to reconstruct 3D spatial information from 2D images, allowing accurate representation of device position and orientation without adding complex hardware imaging systems.
Solution Approach 2:
The patent transforms 2D image data into 3D spatial information by introducing a virtual modeling dimension. Through computer-generated rendering and perspective transformation, the system recovers depth, orientation, and position data that would normally require complex 3D imaging hardware, effectively adding dimensional information through computational methods.
2Productivity
If procedure duration is reduced, then productivity improves, but measurement precision of device position may deteriorate
Solution Approach 1:
The patent performs preliminary actions by pre-defining device geometry, markers, and spatial relationships in a virtual model before the actual procedure begins. During the procedure, the system rapidly matches real-time 2D images against this pre-prepared virtual model, enabling fast 3D position calculation without compromising accuracy, thus achieving both speed and precision.
3Object-affected harmful factors
If radiation exposure is reduced, then harmful factors decrease, but imaging quality may worsen
Solution Approach 1:
The patent introduces a virtual model as an intermediary between the 2D imaging system and the 3D reconstruction process. Instead of relying on high-radiation 3D imaging modalities, the system uses the virtual model to interpret and enhance 2D fluoroscopic images, achieving accurate 3D device positioning with minimal radiation exposure by computationally deriving spatial information rather than capturing it directly through additional radiation.
Data Source
AI summary
A system, method, and computer program product for image-based detection of an object internal to a patient is disclosed. A model can be trained for an internal object, such as an invasive medical device, the trained model being generated from one or more machine learning algorithms that are trained on annotated images of the object with spatial information of the object. An imaging computer system can receive one or more images of the internal object captured by an imaging device positioned external to the patient. The imaging computer system can further detect, based on applying the trained model to the one or more images of the object, the internal object within the patient. A display can output the one or more images and the identifying information for the object.