Spine localization in X-ray images using interest point detection

Evaluation of U-Net models in automated cervical spine and cranial bone segmentation using X-ray images for traumatic atlanto-occipital dislocation diagnosis

Segmentation of the cervical spine in tandem with three cranial bones, hard palate, basion, and opisthion using X-ray images is crucial for measuring metrics used to diagnose traumatic atlanto-occipital dislocation (TAOD). Previous studies utilizing automated segmentation methods have been limited to segmenting parts of the cervical spine (C3 ~ C7), due to difficulties in defining the boundaries of C1 and C2 bones. Additionally, there has yet to be a study that includes cranial bone segmentations necessary for determining TAOD diagnosing metrics, which are usually defined by measuring the distance between certain cervical (C1 ~ C7) and cranial (hard palate, basion, opisthion) bones. For this study, we trained a U-Net model on 513 sagittal X-ray images with segmentations of both cervical and cranial bones for an automated solution to segmenting important features for diagnosing TAOD. Additionally, we tested U-Net derivatives, recurrent residual U-Net, attention U-Net, and attention recurrent residual U-Net to observe any notable differences in segmentation behavior. The accuracy of U-Net models ranged from 99.07 to 99.12%, and dice coefficient values ranged from 88.55 to 89.41%. Results showed that all 4 tested U-Net models were capable of segmenting bones used in measuring TAOD metrics with high accuracy.

Machine learning for image analysis in the cervical spine: Systematic review of the available models and methods

•Neural network approaches show the most potential for automated image analysis of thecervical spine.•Fully automatic convolutional neural network (CNN) models are promising Deep Learning methods for segmentation.•In cervical spine analysis, the biomechanical features are most often studied using finiteelement models.•The application of artificial neural networks and support vector machine models looks promising for classification purposes.•This article provides an overview of the methods for research on computer aided imaging diagnostics of the cervical spine.

Real-time unstructured road detection based on CNN and Gibbs energy function

Road detection algorithms with high robustness as well as timeliness are the basis for developing intelligent assisted driving systems. To improve the robustness as well as the timeliness of unstructured road detection, a new algorithm is proposed in this paper. First, for the first frame in the video, the homography matrix H is estimated based on the improved random sample consensus (RANSAC) algorithm for different regions in the image, and the features of H are automatically extracted using convolutional neural network (CNN), which in turn enables road detection. Secondly, in order to improve the rate of subsequent similar frame detection, the color as well as texture features of the road are extracted from the detection results of the first frame, and the corresponding Gaussian mixture models (GMMs) are constructed based on Orchard-Bouman, and then the Gibbs energy function is used to achieve road detection in subsequent frames. Finally, the above algorithm is verified in a real unstructured road scene, and the experimental results show that the algorithm is 98.4% accurate and can process 58 frames per second with 1024×960 pixels.

AM-UNet: automated mini 3D end-to-end U-net based network for brain claustrum segmentation

Recent advances in deep learning (DL) have provided promising solutions to medical image segmentation. Among existing segmentation approaches, the U-Net-based methods have been used widely. However, very few U-Net-based studies have been conducted on automatic segmentation of the human brain claustrum (CL). The CL segmentation is challenging due to its thin, sheet-like structure, heterogeneity of its image modalities and formats, imperfect labels, and data imbalance. We propose an automatic optimized U-Net-based 3D segmentation model, called AM-UNet, designed as an end-to-end process of the pre and post-process techniques and a U-Net model for CL segmentation. It is a lightweight and scalable solution which has achieved the state-of-the-art accuracy for automatic CL segmentation on 3D magnetic resonance images (MRI). On the T1/T2 combined MRI CL dataset, AM-UNet has obtained excellent results, including Dice, Intersection over Union (IoU), and Intraclass Correlation Coefficient (ICC) scores of 82%, 70%, and 90%, respectively. We have conducted the comparative evaluation of AM-UNet with other pre-existing models for segmentation on the MRI CL dataset. As a result, medical experts confirmed the superiority of the proposed AM-UNet model for automatic CL segmentation. The source code and model of the AM-UNet project is publicly available on GitHub: https://github.com/AhmedAlbishri/AM-UNET.

Automatic Lumbar MRI Detection and Identification Based on Deep Learning

The aim of this research is to automatically detect lumbar vertebras in MRI images with bounding boxes and their classes, which can assist clinicians with diagnoses based on large amounts of MRI slices. Vertebras are highly semblable in appearance, leading to a challenging automatic recognition. A novel detection algorithm is proposed in this paper based on deep learning. We apply a similarity function to train the convolutional network for lumbar spine detection. Instead of distinguishing vertebras using annotated lumbar images, our method compares similarities between vertebras using a beforehand lumbar image. In the convolutional neural network, a contrast object will not update during frames, which allows a fast speed and saves memory. Due to its distinctive shape, S1 is firstly detected and a rough region around it is extracted for searching for L1-L5. The results are evaluated with accuracy, precision, mean, and standard deviation (STD). Finally, our detection algorithm achieves the accuracy of 98.6% and the precision of 98.9%. Most failed results are involved with wrong S1 locations or missed L5. The study demonstrates that a lumbar detection network supported by deep learning can be trained successfully without annotated MRI images. It can be believed that our detection method will assist clinicians to raise working efficiency.

Automatic Lumbar Spine Tracking Based on Siamese Convolutional Network

Deep learning has demonstrated great success in various computer vision tasks. However, tracking the lumbar spine by digitalized video fluoroscopic imaging (DVFI), which can quantitatively analyze the motion mode of the spine to diagnose lumbar instability, has not yet been well developed due to the lack of steady and robust tracking method. The aim of this work is to automatically track lumbar vertebras with rotated bounding boxes in DVFI sequences. Instead of distinguishing vertebras using annotated lumbar images or sequences, we train a full-convolutional siamese neural network offline to learn generic image features with transfer learning. The siamese network is trained to learn a similarity function that compares the labeled target from the initial frame with the candidate patches from the current frame. The similarity function returns a high score if the two images depict the same object. Once learned, the similarity function is used to track a previously unseen object without any adapting online. Our tracker is performed by evaluating the candidate rotated patches sampled around the previous target's position and presents rotated bounding boxes to locate the lumbar spine from L1 to L4. Results indicate that the proposed tracking method can track the lumbar vertebra steadily and robustly. The study demonstrates that the lumbar tracker based on siamese convolutional network can be trained successfully without annotated lumbar sequences.

Image Segmentation and Analysis of Flexion-Extension Radiographs of Cervical Spines

We present a new analysis tool for cervical flexion-extension radiographs based on machine vision and computerized image processing. The method is based on semiautomatic image segmentation leading to detection of common landmarks such as the spinolaminar (SL) line or contour lines of the implanted anterior cervical plates. The technique allows for visualization of the local curvature of these landmarks during flexion-extension experiments. In addition to changes in the curvature of the SL line, it has been found that the cervical plates also deform during flexion-extension examination. While extension radiographs reveal larger curvature changes in the SL line, flexion radiographs on the other hand tend to generate larger curvature changes in the implanted cervical plates. Furthermore, while some lordosis is always present in the cervical plates by design, it actually decreases during extension and increases during flexion. Possible causes of this unexpected finding are also discussed. The described analysis may lead to a more precise interpretation of flexion-extension radiographs, allowing diagnosis of spinal instability and/or pseudoarthrosis in already seemingly fused spines.

A fast method for spine localization in x-ray images

Detection of spines in medical images are important tasks in medical applications. These tasks are relatively easy for CT/MR images because the bones are easily distinguishable from other tissues. However, they are difficult for x-ray images due to bone and soft tissue overlapping. This paper illustrates a method for detecting the medial axis of spine in x-ray images. Given an initial point on the spine in the x-ray image manually or automatically, the method iteratively searches for good feature points on the spine to locate the medial axis. As a result, the effort of determining the relevant medical information, such as Cobb's angle, can be minimized. The proposed method is fast and efficient. In average it took less than 1 second for localizing the spine on a 3000×1000 gray scale x-ray image.

A framework of vertebra segmentation using the active shape model-based approach

We propose a medical image segmentation approach based on the Active Shape Model theory. We apply this method for cervical vertebra detection. The main advantage of this approach is the application of a statistical model created after a training stage. Thus, the knowledge and interaction of the domain expert intervene in this approach. Our application allows the use of two different models, that is, a global one (with several vertebrae) and a local one (with a single vertebra). Two modes of segmentation are also proposed: manual and semiautomatic. For the manual mode, only two points are selected by the user on a given image. The first point needs to be close to the lower anterior corner of the last vertebra and the second near the upper anterior corner of the first vertebra. These two points are required to initialize the segmentation process. We propose to use the Harris corner detector combined with three successive filters to carry out the semiautomatic process. The results obtained on a large set of X-ray images are very promising.

Detection of vertebral plateaus in lateral lumbar spinal X-ray images with Gabor filters

A few recent studies have proposed computed-aided methods for the detection and analysis of vertebral bodies in radiographic images. This paper presents a method based on Gabor filters. Forty-one lateral lumbar spinal X-ray images from different patients were included in the study. For each image, a radiologist manually delineated the vertebral plateaus of L1, L2, L3, and L4 using a software tool for image display and mark-up. Each original image was filtered with a bank of 180 Gabor filters. The angle of the Gabor filter with the highest response at each pixel was used to derive a measure of the strength of orientation or alignment. In order to limit the spatial extent of the image data and the derived features in further analysis, a semi-automated procedure was applied to the original image. A neural network utilizing the logistic sigmoid function was trained with pixel intensity from the original image, the result of manual delineation of the plateaus, the Gabor magnitude response, and the alignment image. The average overlap between the results of detection by image processing and manual delineation of the plateaus of L1-L4 in the 41 images tested was 0.917. The results are expected to be useful in the analysis of vertebral deformities and fractures.