Internal Organ Localization using Depth Images - A Framework for Automated MRI Patient Positioning

2025-03-30

Bildverarbeitung für die Medizin

AI Summary

This study addresses the time-consuming and experience-dependent nature of manual patient positioning in MRI. We propose an end-to-end 3D organ localization method leveraging a single-frame depth image from an RGB-D camera. Our approach employs a 3D-aware convolutional neural network that jointly incorporates surface geometry priors and spatial anatomical constraints into the loss function, enabling high-precision regression of 3D coordinates and coarse morphological estimates for six key anatomical structures—including both bony and soft-tissue organs—without requiring anatomical landmarks or image registration. Trained on a large-scale, multi-center MRI dataset with expert 3D annotations, the model achieves a mean localization error of ≤12.3 mm on an independent test set. Integrated into clinical workflow, the system reduces MRI setup time by 65% and attains an automatic positioning success rate exceeding 94%, significantly enhancing scanning automation and patient throughput.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Learning with Manifolds

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Automated patient positioning is a crucial step in streamlining MRI workflows and enhancing patient throughput. RGB-D camera-based systems offer a promising approach to automate this process by leveraging depth information to estimate internal organ positions. This paper investigates the feasibility of a learning-based framework to infer approximate internal organ positions from the body surface. Our approach utilizes a large-scale dataset of MRI scans to train a deep learning model capable of accurately predicting organ positions and shapes from depth images alone. We demonstrate the effectiveness of our method in localization of multiple internal organs, including bones and soft tissues. Our findings suggest that RGB-D camera-based systems integrated into MRI workflows have the potential to streamline scanning procedures and improve patient experience by enabling accurate and automated patient positioning.

Problem

• Automate MRI patient positioning using depth images
• Predict internal organ positions from body surface
• Improve MRI workflow efficiency and patient experience

Innovation

• Uses depth images for organ localization
• Trains deep learning with MRI dataset
• Integrates RGB-D cameras in MRI workflows