An optimized EBRSA-Bi LSTM model for highly undersampled rapid CT image reconstruction

Abstract

COVID-19 has spread all over the world, causing serious panic around the globe. Chest computed tomography (CT) images are integral in confirming COVID positive patients. Several investigations were conducted to improve or maintain the image reconstruction quality for the sample image reconstruction. Deep learning (DL) methods have recently been proposed to achieve fast reconstruction, but many have focused on a single domain, such as the image domain of k-space. In this research, the highly under-sampled enhanced battle royale self-attention based bi-directional long short-term (EBRSA-bi LSTM) CT image reconstruction model is proposed to reconstruct the image from the under-sampled data. The research is adapted with two phases, namely, pre-processing and reconstruction. The extended cascaded filter (ECF) is proposed for image pre-processing and tends to suppress the noise and enhance the reconstruction accuracy. In the reconstruction model, the battle royale optimization (BrO) is intended to diminish the loss function of the reconstruction network model and weight updation. The proposed model is tested with two datasets, COVID-CT- and SARS-CoV-2 CT. The reconstruction accuracy of the proposed model with two datasets is 93.5 % and 97.7 %, respectively. Also, the image quality assessment parameters such as Peak-Signal to Noise Ratio (PSNR), Root Mean Square Error (RMSE) and Structural Similarity Index metric (SSIM) are evaluated, and it yields an outcome of (45 and 46 dB), (0.0026 and 0.0022) and (0.992, 0.996) with two datasets.

Keywords: Computed tomography (CT); Deep learning; Image reconstruction; K-space data; Under-sampling.

Fig. 1

Global architecture of the proposed model.

Fig. 2

Proposed reconstruction network model.

Fig. 3

Basic structure of LSTM.

Fig. 4

Input and Filtered image for both datasets.

Fig. 5

k-space generation.

Fig. 6

(a) Analysis of training, testing and validation accuracy, (b) Analysis of training, testing and validation loss.

Fig. 7

(a) Analysis of training, testing and validation accuracy, (b) Analysis of training, testing and validation loss.

Fig. 8

Analysis of PSNR and SSIM metrics with various sampling rates.

Fig. 9

Analysis of RMSE and EPI metrics with various sampling rate.

Fig. 10

Analysis of SI with various sampling rates.

Fig. 11

Input image, k-space data and reconstructed image with varying sampling rate (a) Dataset 1, (b) Dataset 2.

Fig. 12

Performance analysis of PSNR.

Fig. 13

Performance analysis of SSIM.

Fig. 14

Performance analysis of RMSE.

Fig. 15

Accuracy and loss values for training and validation data with varying patch sizes. (a)Training accuracy, (b) Training loss, (c) Validation accuracy, (d) Validation loss.

Fig. 16

Reconstructed results. Input image, filtered image, k-space data and reconstructed image for two datasets (Row1: Dataset1, Row 2: Dataset 2).