Fully Automated Left Atrium Segmentation from Anatomical Cine Long-axis MRI Sequences using Deep Convolutional Neural Network with Unscented Kalman Filter

TL;DR

This paper introduces a fully automated deep learning and Bayesian filtering method for accurate left atrium segmentation in cine MRI sequences, improving robustness and performance over traditional techniques.

Contribution

It presents a novel approach combining neural networks and unscented Kalman filtering with sequence type classification for precise cardiac chamber segmentation.

Findings

Achieved a mean Dice coefficient of 94.1% for 2-chamber sequences.

Outperformed state-of-the-art methods in segmentation accuracy.

Effective with varying training dataset sizes from 20 to 80 patients.

Abstract

This study proposes a fully automated approach for the left atrial segmentation from routine cine long-axis cardiac magnetic resonance image sequences using deep convolutional neural networks and Bayesian filtering. The proposed approach consists of a classification network that automatically detects the type of long-axis sequence and three different convolutional neural network models followed by unscented Kalman filtering (UKF) that delineates the left atrium. Instead of training and predicting all long-axis sequence types together, the proposed approach first identifies the image sequence type as to 2, 3 and 4 chamber views, and then performs prediction based on neural nets trained for that particular sequence type. The datasets were acquired retrospectively and ground truth manual segmentation was provided by an expert radiologist. In addition to neural net based classification and segmentation, another neural net is trained and utilized to select image sequences for further processing using UKF to impose temporal consistency over cardiac cycle. A cyclic dynamic model with time-varying angular frequency is introduced in UKF to characterize the variations in cardiac motion during image scanning. The proposed approach was trained and evaluated separately with varying amount of training data with images acquired from 20, 40, 60 and 80 patients. Evaluations over 1515 images with equal number of images from each chamber group acquired from an additional 20 patients demonstrated that the proposed model outperformed state-of-the-art and yielded a mean Dice coefficient value of 94.1%, 93.7% and 90.1% for 2, 3 and 4-chamber sequences, respectively, when trained with datasets from 80 patients.