Prostate m-rep Model Mapping MRI to TRUS Biopsy Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for medical image segmentation, particularly in prostate cancer diagnosis, are inefficient due to the inability to accurately differentiate between normal and cancerous tissues during transrectal ultrasound-guided biopsies, leading to high false negative rates and incomplete tumor sampling, which results in inappropriate treatment decisions and increased healthcare costs.
Innovation Solution
The development of a system that maps a model of an anatomical structure from a planning image to an intervention-guiding image, using a patient-specific medial representation object model (m-rep) to accurately identify and overlay intervention target regions on intervention-guiding images, reducing the need for human interaction and minimizing user variability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If transrectal ultrasound imaging is used to guide biopsy procedures, then the overall shape of the prostate can be visualized, but the ability to differentiate between normal and cancerous tissues is lost
Solution Approach 1:
The patent introduces a 3D probabilistic atlas as an intermediary that bridges the gap between planning images (MRI) and intervention images (TRUS). The atlas serves as a mediator that can be deformed and registered across different imaging modalities, allowing cancerous regions identified in MRI to be accurately localized in TRUS images during biopsy procedures.
Solution Approach 2:
The patent segments the prostate into multiple tissue types (glandular, transition zone, peripheral zone) and further divides them into cancerous and non-cancerous regions using probability maps. This segmentation allows the system to differentiate between tissue types and track cancerous regions across different imaging modalities, resolving the contradiction between procedural ease and tissue differentiation capability.
2Adaptability or versatility
If a pre-defined grid pattern is applied for biopsy sampling, then the procedure can be standardized, but the false negative rate increases due to inability to target suspicious regions
Solution Approach 1:
The system performs preliminary action by creating a 3D probabilistic atlas before the biopsy procedure. The atlas is built from planning images and includes probability maps that identify cancerous regions. This preliminary modeling allows the system to pre-identify target regions, which are then accurately localized during the actual biopsy procedure, eliminating the need for grid-based sampling and reducing false negatives.
Solution Approach 2:
The patent implements feedback by using probability maps that provide real-time information about the likelihood of cancer presence in different prostate regions. This feedback mechanism guides the biopsy procedure by highlighting suspicious areas, allowing the system to adapt the sampling strategy based on identified cancerous regions rather than following a fixed grid pattern.
3Measurement precision
If expert human interaction is used for image segmentation, then segmentation accuracy can be maintained, but the process becomes extremely time consuming and expensive
Solution Approach 1:
The patent implements self-service by using automated algorithms to create and deform the 3D probabilistic atlas. The system automatically processes planning images, generates probability maps, and deforms the atlas to match intervention images. This automation eliminates the need for time-consuming expert human interaction while maintaining high segmentation accuracy through sophisticated computational methods.
Solution Approach 2:
The patent replaces the mechanical system of manual expert segmentation with an automated computational system. The 3D probabilistic atlas approach uses algorithmic deformation and registration processes to achieve segmentation, substituting human expertise with automated mathematical models that can process images rapidly and consistently without human intervention.
4Ease of operation
If human experts perform image segmentation, then segmentation can be performed, but inter- and intra-user variabilities adversely affect clinical decisions
Solution Approach 1:
The patent uses parameter changes by deforming the 3D probabilistic atlas using mathematical transformations that account for variations in prostate shape and position. The atlas deformation parameters are adjusted to match the specific patient anatomy, ensuring consistent and reliable segmentation results that are not affected by human expert variability. The probability maps provide standardized quantitative measures that eliminate subjectivity.
Data Source
AI summary
Methods, systems, and computer readable media for mapping a model of an object comprising an anatomical structure in a planning image and an intervention target region within it to intervention-guiding image data are disclosed. According to one method, an initial medial representation object model (m-rep) of an object comprising an anatomical structure is created based on image data of at least a first instance of the object. A patient-specific m-rep is created by deforming the initial m-rep based on planning image data of at least a second instance of the object, wherein the at least second instance of the object is associated with the patient. An intervention target region within the m-rep is identified in an image registered with the planning image. The patient-specific m-rep is correlated to the intervention-guiding image data of the at least second instance of the object, deformed from the planning image. The intervention target region is transferred to the intervention-guiding image according to the transformation between the m-rep in the planning image and the m-rep in the intervention-guiding image.