Spinal Cord Segmentation in Ultrasound Medical Imagery

Automated Method for Growing Rod Length Measurement on Ultrasound Images in Children With Early Onset Scoliosis

OBJECTIVE

To develop and validate machine learning algorithms to automatically extract the rod length of the magnetically controlled growing rod from ultrasound images (US) in a pilot study.

METHODS

Two machine-learning (ML) models, called the "Boundary model" and "Rod model," were developed to identify specific rod segments on ultrasound images. The models were developed utilizing Mask Regional Convolutional Neural Networks (Mask RCNN). Ninety US images were acquired from 23 participants who had early onset scoliosis (EOS) surgeries; among those, 70 were used for model development, including training and validation, and 20 were used for testing by comparing the AI-based vs. manual measurements.

RESULTS

The average precision (AP) of the ML models was 88.5% and 60.2%, respectively. The inter-method correlation coefficient (ICC) was 0.98, and the mean absolute difference ± standard deviation (MAD ± SD) between AI and manual measurements was 0.86 ± 1.0 mm. The Bland-Altman analysis showed no bias, and 90% of the data were within the 95% confidence interval. The automated method was reliable, accurate, and fast. Measurements were displayed in 4.6 seconds after the US image was inputted.

CONCLUSION

This was the first AI-based method to measure the MCGR rod length on US images automatically.

Vision Transformers in Image Restoration: A Survey

The Vision Transformer (ViT) architecture has been remarkably successful in image restoration. For a while, Convolutional Neural Networks (CNN) predominated in most computer vision tasks. Now, both CNN and ViT are efficient approaches that demonstrate powerful capabilities to restore a better version of an image given in a low-quality format. In this study, the efficiency of ViT in image restoration is studied extensively. The ViT architectures are classified for every task of image restoration. Seven image restoration tasks are considered: Image Super-Resolution, Image Denoising, General Image Enhancement, JPEG Compression Artifact Reduction, Image Deblurring, Removing Adverse Weather Conditions, and Image Dehazing. The outcomes, the advantages, the limitations, and the possible areas for future research are detailed. Overall, it is noted that incorporating ViT in the new architectures for image restoration is becoming a rule. This is due to some advantages compared to CNN, such as better efficiency, especially when more data are fed to the network, robustness in feature extraction, and a better feature learning approach that sees better the variances and characteristics of the input. Nevertheless, some drawbacks exist, such as the need for more data to show the benefits of ViT over CNN, the increased computational cost due to the complexity of the self-attention block, a more challenging training process, and the lack of interpretability. These drawbacks represent the future research direction that should be targeted to increase the efficiency of ViT in the image restoration domain.

Rule and Neural Network-Based Image Segmentation of Mice Vertebrae Images

Background

Image segmentation is a fundamental technique that allows researchers to process images from various sources into individual components for certain applications, such as visual or numerical evaluations. Image segmentation is beneficial when studying medical images for healthcare purposes. However, existing semantic image segmentation models like the U-net are computationally intensive. This work aimed to develop less complicated models that could still accurately segment images.

Methodology

Rule-based and linear layer neural network models were developed in Mathematica and trained on mouse vertebrae micro-computed tomography scans. These models were tasked with segmenting the cortical shell from the whole bone image. A U-net model was also set up for comparison.

Results

It was found that the linear layer neural network had comparable accuracy to the U-net model in segmenting the mice vertebrae scans.

Conclusions

This work provides two separate models that allow for automated segmentation of mouse vertebral scans, which could be potentially valuable in applications such as pre-processing the murine vertebral scans for further evaluations of the effect of drug treatment on bone micro-architecture.

LVNet: Lightweight Model for Left Ventricle Segmentation for Short Axis Views in Echocardiographic Imaging

Lightweight segmentation models are becoming more popular for fast diagnosis on small and low cost medical imaging devices. This study focuses on the segmentation of the left ventricle (LV) in cardiac ultrasound (US) images. A new lightweight model [LV network (LVNet)] is proposed for segmentation, which gives the benefits of requiring fewer parameters but with improved segmentation performance in terms of Dice score (DS). The proposed model is compared with state-of-the-art methods, such as UNet, MiniNetV2, and fully convolutional dense dilated network (FCdDN). The model proposed comes with a post-processing pipeline that further enhances the segmentation results. In general, the training is done directly using the segmentation mask as the output and the US image as the input of the model. A new strategy for segmentation is also introduced in addition to the direct training method used. Compared with the UNet model, an improvement in DS performance as high as 5% for segmentation with papillary (WP) muscles was found, while showcasing an improvement of 18.5% when the papillary muscles are excluded. The model proposed requires only 5% of the memory required by a UNet model. LVNet achieves a better trade-off between the number of parameters and its segmentation performance as compared with other conventional models. The developed codes are available at https://github.com/navchetanawasthi/Left_Ventricle_Segmentation.

COVID-19 Diagnosis in Chest X-rays Using Deep Learning and Majority Voting

The COVID-19 disease has spread all over the world, representing an intriguing challenge for humanity as a whole. The efficient diagnosis of humans infected by COVID-19 still remains an increasing need worldwide. The chest X-ray imagery represents, among others, one attractive means to detect COVID-19 cases efficiently. Many studies have reported the efficiency of using deep learning classifiers in diagnosing COVID-19 from chest X-ray images. They conducted several comparisons among a subset of classifiers to identify the most accurate. In this paper, we investigate the potential of the combination of state-of-the-art classifiers in achieving the highest possible accuracy for the detection of COVID-19 from X-ray. For this purpose, we conducted a comprehensive comparison study among 16 state-of-the-art classifiers. To the best of our knowledge, this is the first study considering this number of classifiers. This paper’s innovation lies in the methodology that we followed to develop the inference system that allows us to detect COVID-19 with high accuracy. The methodology consists of three steps: (1) comprehensive comparative study between 16 state-of-the-art classifiers; (2) comparison between different ensemble classification techniques, including hard/soft majority, weighted voting, Support Vector Machine, and Random Forest; and (3) finding the combination of deep learning models and ensemble classification techniques that lead to the highest classification confidence on three classes. We found that using the Majority Voting approach is an adequate strategy to adopt in general cases for this task and may achieve an average accuracy up to 99.314%.

In-Series U-Net Network to 3D Tumor Image Reconstruction for Liver Hepatocellular Carcinoma Recognition

Cancer is one of the common diseases. Quantitative biomarkers extracted from standard-of-care computed tomography (CT) scan can create a robust clinical decision tool for the diagnosis of hepatocellular carcinoma (HCC). According to the current clinical methods, the situation usually accounts for high expenditure of time and resources. To improve the current clinical diagnosis and therapeutic procedure, this paper proposes a deep learning-based approach, called Successive Encoder-Decoder (SED), to assist in the automatic interpretation of liver lesion/tumor segmentation through CT images. The SED framework consists of two different encoder-decoder networks connected in series. The first network aims to remove unwanted voxels and organs and to extract liver locations from CT images. The second network uses the results of the first network to further segment the lesions. For practical purpose, the predicted lesions on individual CTs were extracted and reconstructed on 3D images. The experiments conducted on 4300 CT images and LiTS dataset demonstrate that the liver segmentation and the tumor prediction achieved 0.92 and 0.75 in Dice score, respectively, by as-proposed SED method.