3D Probabilistic Segmentation and Volumetry from 2D projection images

TL;DR

This paper introduces probabilistic 2D-to-3D convolutional neural networks for reconstructing volumetric images from 2D X-ray projections, achieving high accuracy on CT data and providing confidence measures.

Contribution

It presents a novel end-to-end trainable method for 3D reconstruction from 2D images with performance evaluation and confidence estimation capabilities.

Findings

Achieved Dice score of 0.91 on CT scans

Effective for large connected structures

Limited in reconstructing fine structures

Abstract

X-Ray imaging is quick, cheap and useful for front-line care assessment and intra-operative real-time imaging (e.g., C-Arm Fluoroscopy). However, it suffers from projective information loss and lacks vital volumetric information on which many essential diagnostic biomarkers are based on. In this paper we explore probabilistic methods to reconstruct 3D volumetric images from 2D imaging modalities and measure the models' performance and confidence. We show our models' performance on large connected structures and we test for limitations regarding fine structures and image domain sensitivity. We utilize fast end-to-end training of a 2D-3D convolutional networks, evaluate our method on 117 CT scans segmenting 3D structures from digitally reconstructed radiographs (DRRs) with a Dice score of $0.91 \pm 0.0013$. Source code will be made available by the time of the conference.