An official website of the United States government
This paper presents an approach to enhanced endoscopic tool segmentation combining separate pathways utilizing input images in two different coordinate representations. The proposed method examines U-Net convolutional neural networks with input endoscopic images represented via (1) the original rectangular coordinate format alongside (2) a morphological polar coordinate transformation. To maximize information and the breadth of the endoscope frustrum, imaging sensors are oftentimes larger than the image circle. This results in unused border regions. Ideally, the region of interest is proximal to the image center. The above two observations formed the basis for the morphological polar transformation pathway as an augmentation to typical rectangular input image representations. Results indicate that neither of the two investigated coordinate representations consistently yielded better segmentation performance as compared to the other. Improved segmentation can be achieved with a hybrid approach that carefully selects which of the two pathways to be used for individual input images. Towards that end, two binary classifiers were trained to identify, given an input endoscopic image, which of the two coordinate representation segmentation pathways (rectangular or polar), would result in better segmentation performance. Results are promising and suggest marked improvements using a hybrid pathway selection approach compared to either alone. The experiment used to evaluate the proposed hybrid method utilized a dataset consisting of 8360 endoscopic images from real surgery and evaluated segmentation performance with Dice coefficient and Intersection over Union. The results suggest that on-the-fly polar transformation for tool segmentation is useful when paired with the proposed hybrid tool-segmentation approach.
more »
« less
Improving Echocardiography Segmentation by Polar Transformation
Segmentation of echocardiograms plays an essential role in the quantitative analysis of the heart and helps diagnose cardiac diseases. In the recent decade, deep learning-based approaches have significantly improved the performance of echocardiogram segmentation. Most deep learning-based methods assume that the image to be processed is rectangular in shape. However, typically echocardiogram images are formed within a sector of a circle, with a significant region in the overall rectangular image where there is no data, a result of the ultrasound imaging methodology. This large non-imaging region can influence the training of deep neural networks. In this paper, we propose to use polar transformation to help train deep learning algorithms. Using the r-θ transformation, a significant portion of the non-imaging background is removed, allowing the neural network to focus on the heart image. The segmentation model is trained on both x-y and r-θ images. During inference, the predictions from the x-y and r-θ images are combined using max-voting. We verify the efficacy of our method on the CAMUS dataset with a variety of segmentation networks, encoder networks, and loss functions. The experimental results demonstrate the effectiveness and versatility of our proposed method for improving the segmentation results.
more »
« less
- Award ID(s):
- 1633295
- PAR ID:
- 10572722
- Publisher / Repository:
- Statistical Atlas and Computational Models of the Heart
- Date Published:
- Format(s):
- Medium: X
- Location:
- Singapore
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
This paper presents a tool-pose-informed variable center morphological polar transform to enhance segmentation of endoscopic images. The representation, while not loss-less, transforms rigid tool shapes into morphologies consistently more rectangular that may be more amenable to image segmentation networks. The proposed method was evaluated using the U-Net convolutional neural network, and the input images from endoscopy were represented in one of the four different coordinate formats (1) the original rectangular image representation, (2) the morphological polar coordinate transform, (3) the proposed variable center transform about the tool-tip pixel and (4) the proposed variable center transform about the tool vanishing point pixel. Previous work relied on the observations that endoscopic images typically exhibit unused border regions with content in the shape of a circle (since the image sensor is designed to be larger than the image circle to maximize available visual information in the constrained environment) and that the region of interest (ROI) was most ideally near the endoscopic image center. That work sought an intelligent method for, given an input image, carefully selecting between methods (1) and (2) for best image segmentation prediction. In this extension, the image center reference constraint for polar transformation in method (2) is relaxed via the development of a variable center morphological transformation. Transform center selection leads to different spatial distributions of image loss, and the transform-center location can be informed by robot kinematic model and endoscopic image data. In particular, this work is examined using the tool-tip and tool vanishing point on the image plane as candidate centers. The experiments were conducted for each of the four image representations using a data set of 8360 endoscopic images from real sinus surgery. The segmentation performance was evaluated with standard metrics, and some insight about loss and tool location effects on performance are provided. Overall, the results are promising, showing that selecting a transform center based on tool shape features using the proposed method can improve segmentation performance.more » « less
-
null (Ed.)X-ray CT imaging provides a 3D view of a sample and is a powerful tool for investigating the internal features of porous rock. Reliable phase segmentation in these images is highly necessary but, like any other digital rock imaging technique, is time-consuming, labor-intensive, and subjective. Combining 3D X-ray CT imaging with machine learning methods that can simultaneously consider several extracted features in addition to color attenuation, is a promising and powerful method for reliable phase segmentation. Machine learning-based phase segmentation of X-ray CT images enables faster data collection and interpretation than traditional methods. This study investigates the performance of several filtering techniques with three machine learning methods and a deep learning method to assess the potential for reliable feature extraction and pixel-level phase segmentation of X-ray CT images. Features were first extracted from images using well-known filters and from the second convolutional layer of the pre-trained VGG16 architecture. Then, K-means clustering, Random Forest, and Feed Forward Artificial Neural Network methods, as well as the modified U-Net model, were applied to the extracted input features. The models’ performances were then compared and contrasted to determine the influence of the machine learning method and input features on reliable phase segmentation. The results showed considering more dimensionality has promising results and all classification algorithms result in high accuracy ranging from 0.87 to 0.94. Feature-based Random Forest demonstrated the best performance among the machine learning models, with an accuracy of 0.88 for Mancos and 0.94 for Marcellus. The U-Net model with the linear combination of focal and dice loss also performed well with an accuracy of 0.91 and 0.93 for Mancos and Marcellus, respectively. In general, considering more features provided promising and reliable segmentation results that are valuable for analyzing the composition of dense samples, such as shales, which are significant unconventional reservoirs in oil recovery.more » « less
-
Abstract Computational fluid dynamics (CFD) modeling of left ventricle (LV) flow combined with patient medical imaging data has shown great potential in obtaining patient-specific hemodynamics information for functional assessment of the heart. A typical model construction pipeline usually starts with segmentation of the LV by manual delineation followed by mesh generation and registration techniques using separate software tools. However, such approaches usually require significant time and human efforts in the model generation process, limiting large-scale analysis. In this study, we propose an approach toward fully automating the model generation process for CFD simulation of LV flow to significantly reduce LV CFD model generation time. Our modeling framework leverages a novel combination of techniques including deep-learning based segmentation, geometry processing, and image registration to reliably reconstruct CFD-suitable LV models with little-to-no user intervention.1 We utilized an ensemble of two-dimensional (2D) convolutional neural networks (CNNs) for automatic segmentation of cardiac structures from three-dimensional (3D) patient images and our segmentation approach outperformed recent state-of-the-art segmentation techniques when evaluated on benchmark data containing both magnetic resonance (MR) and computed tomography(CT) cardiac scans. We demonstrate that through a combination of segmentation and geometry processing, we were able to robustly create CFD-suitable LV meshes from segmentations for 78 out of 80 test cases. Although the focus on this study is on image-to-mesh generation, we demonstrate the feasibility of this framework in supporting LV hemodynamics modeling by performing CFD simulations from two representative time-resolved patient-specific image datasets.more » « less
-
Traditionally, a high-performance microscope with a large numerical aperture is required to acquire high-resolution images. However, the images’ size is typically tremendous. Therefore, they are not conveniently managed and transferred across a computer network or stored in a limited computer storage system. As a result, image compression is commonly used to reduce image size resulting in poor image resolution. Here, we demonstrate custom convolution neural networks (CNNs) for both super-resolution image enhancement from low-resolution images and characterization of both cells and nuclei from hematoxylin and eosin (H&E) stained breast cancer histopathological images by using a combination of generator and discriminator networks so-called super-resolution generative adversarial network-based on aggregated residual transformation (SRGAN-ResNeXt) to facilitate cancer diagnosis in low resource settings. The results provide high enhancement in image quality where the peak signal-to-noise ratio and structural similarity of our network results are over 30 dB and 0.93, respectively. The derived performance is superior to the results obtained from both the bicubic interpolation and the well-known SRGAN deep-learning methods. In addition, another custom CNN is used to perform image segmentation from the generated high-resolution breast cancer images derived with our model with an average Intersection over Union of 0.869 and an average dice similarity coefficient of 0.893 for the H&E image segmentation results. Finally, we propose the jointly trained SRGAN-ResNeXt and Inception U-net Models, which applied the weights from the individually trained SRGAN-ResNeXt and inception U-net models as the pre-trained weights for transfer learning. The jointly trained model’s results are progressively improved and promising. We anticipate these custom CNNs can help resolve the inaccessibility of advanced microscopes or whole slide imaging (WSI) systems to acquire high-resolution images from low-performance microscopes located in remote-constraint settings.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
