3D-Printable Patient-Specific Anatomy Models Using ML Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods lack a secure and efficient platform for generating and delivering customizable 3D printed models of patient-specific anatomy from 2D medical images, with a need for improved accuracy and insight into patient anatomy or pathology.
Innovation Solution
A computer-implemented method using machine learning-based image segmentation to generate 3D printable models from 2D medical images, incorporating voxel classification, multi-channel training, and cryptographic file signing for secure distribution, enabling remote printing and verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If machine learning based image segmentation is used to generate 3D printable models from 2D medical images, then manufacturing precision and measurement precision are improved, but device complexity and difficulty of detecting and measuring increase
Solution Approach 1:
The patent replaces manual segmentation processes with automated machine learning algorithms. The system uses trained neural networks to automatically identify and segment anatomical structures from 2D medical images, converting a complex manual task into an automated computational process that achieves high precision without requiring manual intervention for each segmentation step.
Solution Approach 2:
The system creates a digital copy of the patient's anatomy through 3D modeling from 2D images. This virtual replica is then used for surgical planning, education, and patient communication without exposing actual patient data, allowing multiple users to access the same anatomical information securely.
2Productivity
If automated 3D model generation is implemented, then productivity is improved, but reliability and security of patient data processing deteriorate
Solution Approach 1:
The system performs preliminary actions by pre-processing 2D medical images to enhance anatomical features before segmentation. This includes adjusting image contrast, filtering noise, and aligning images from different modalities (CT, MRI, PET) to ensure accurate segmentation and model generation from the outset.
Solution Approach 2:
The patent introduces an intermediary security layer using cryptographic hashing and digital signatures. Each generated 3D model is signed with a unique cryptographic hash that verifies the model's integrity and authenticates its source, creating a secure chain of trust between the processing system and the final output without exposing patient data.
3Measurement precision
If multi-channel training and voxel classification are used, then measurement precision is improved, but use of energy and computing resources increase
Solution Approach 1:
The system applies local quality by focusing computational resources on specific regions of interest within the 3D model rather than processing the entire volume uniformly. The machine learning algorithms identify and segment only the anatomical structures relevant to the specific clinical question, reducing unnecessary computational energy while maintaining high precision for critical features.
Solution Approach 2:
The patent uses partial action by processing only the necessary portions of the medical images and 3D models required for the specific diagnostic or surgical planning task. The system dynamically adjusts the level of processing based on the clinical requirements, applying full multi-channel analysis only when necessary and using simplified methods for routine cases.
Data Source
AI summary
A computer implemented method for generating a 3D printable model of a patient specific anatomic feature from 2D medical images is provided. A 3D image is automatically generated from a set of 2D medical images. A machine learning based image segmentation technique is used to segment the generated 3D image. A 3D printable model of the patient specific anatomic feature is created from the segmented 3D image.