Automatic Segmentation Framework of Building Anatomical Mouse Model for Bioluminescence Tomography

Accurate segmentation and classification of different anatomical structures of teeth from medical images plays an essential role in many clinical applications. Usually, the anatomical structures of teeth are manually labelled by experienced clinical doctors, which is time consuming. However, automatic segmentation and classification is a challenging task because the anatomical structures and surroundings of the tooth in medical images are rather complex. Therefore, in this paper, we propose an effective framework which is designed to segment the tooth with a Selective Binary and Gaussian Filtering Regularized Level Set (GFRLS) method improved by fully utilizing three dimensional (3D) information, and classify the tooth by employing unsupervised learning Pulse Coupled Neural Networks (PCNN) model. In order to evaluate the proposed method, the experiments are conducted on the different datasets of mandibular molars and the experimental results show that our method can achieve better accuracy and robustness compared to other four state of the art clustering methods.

BackgroundThis study investigates the variation in segmentation of several pelvic anatomical structures on computed tomography (CT) between multiple observers and a commercial automatic segmentation method, in the context of quality assurance and evaluation during a multicentre clinical trial.MethodsCT scans of two prostate cancer patients (‘benchmarking cases’), one high risk (HR) and one intermediate risk (IR), were sent to multiple radiotherapy centres for segmentation of prostate, rectum and bladder structures according to the TROG 03.04 “RADAR” trial protocol definitions. The same structures were automatically segmented using iPlan software for the same two patients, allowing structures defined by automatic segmentation to be quantitatively compared with those defined by multiple observers. A sample of twenty trial patient datasets were also used to automatically generate anatomical structures for quantitative comparison with structures defined by individual observers for the same datasets.ResultsThere was considerable agreement amongst all observers and automatic segmentation of the benchmarking cases for bladder (mean spatial variations < 0.4 cm across the majority of image slices). Although there was some variation in interpretation of the superior-inferior (cranio-caudal) extent of rectum, human-observer contours were typically within a mean 0.6 cm of automatically-defined contours. Prostate structures were more consistent for the HR case than the IR case with all human observers segmenting a prostate with considerably more volume (mean +113.3%) than that automatically segmented. Similar results were seen across the twenty sample datasets, with disagreement between iPlan and observers dominant at the prostatic apex and superior part of the rectum, which is consistent with observations made during quality assurance reviews during the trial.ConclusionsThis study has demonstrated quantitative analysis for comparison of multi-observer segmentation studies. For automatic segmentation algorithms based on image-registration as in iPlan, it is apparent that agreement between observer and automatic segmentation will be a function of patient-specific image characteristics, particularly for anatomy with poor contrast definition. For this reason, it is suggested that automatic registration based on transformation of a single reference dataset adds a significant systematic bias to the resulting volumes and their use in the context of a multicentre trial should be carefully considered.

Objective:Precise segmentation of human organs and anatomic structures (especially organs at risk, OARs) is the basis and prerequisite for the treatment planning of radiation therapy. In order to ensure rapid and accurate design of radiotherapy treatment planning, an automatic organ segmentation technique was investigated based on deep learning convolutional neural network.MethodA deep learning convolutional neural network (CNN) algorithm called BCDU-Net has been modified and developed further by us. Twenty two thousand CT images and the corresponding organ contours of 17 types delineated manually by experienced physicians from 329 patients were used to train and validate the algorithm. The CT images randomly selected were employed to test the modified BCDU-Net algorithm. The weight parameters of the algorithm model were acquired from the training of the convolutional neural network.ResultThe average Dice similarity coefficient (DSC) of the automatic segmentation and manual segmentation of the human organs of 17 types reached 0.8376, and the best coefficient reached up to 0.9676. It took 1.5–2 s and about 1 h to automatically segment the contours of an organ in an image of the CT dataset for a patient and the 17 organs for the CT dataset with the method developed by us, respectively.ConclusionThe modified deep neural network algorithm could be used to automatically segment human organs of 17 types quickly and accurately. The accuracy and speed of the method meet the requirements of its application in radiotherapy.

Accurate segmentation of anatomical brain structures is crucial for many neuroimaging applications, e.g., early brain development studies and the study of imaging biomarkers of neurodegenerative diseases. Although multi-atlas segmentation (MAS) has achieved many successes in the medical imaging area, this approach encounters limitations in segmenting anatomical structures associated with poor image contrast. To address this issue, we propose a new MAS method that uses a hypergraph learning framework to model the complex subject-within and subject-to-atlas image voxel relationships and propagate the label on the atlas image to the target subject image. To alleviate the low-image contrast issue, we propose two strategies equipped with our hypergraph learning framework. First, we use a hierarchical strategy that exploits high-level context features for hypergraph construction. Because the context features are computed on the tentatively estimated probability maps, we can ultimately turn the hypergraph learning into a hierarchical model. Second, instead of only propagating the labels from the atlas images to the target subject image, we use a dynamic label propagation strategy that can gradually use increasing reliably identified labels from the subject image to aid in predicting the labels on the difficult-to-label subject image voxels. Compared with the state-of-the-art label fusion methods, our results show that the hierarchical hypergraph learning framework can substantially improve the robustness and accuracy in the segmentation of anatomical brain structures with low image contrast from magnetic resonance (MR) images.

Semantic segmentation of anatomical structures in laparoscopic videos is a crucial task to enable the development of new computer-assisted systems that can assist surgeons during surgery. However, this is a difficult task due to artifacts and similar visual characteristics of anatomical structures on the laparoscopic videos. Recently, deep learning algorithms have been showed promising results on the segmentation of laparoscopic instruments. However, due to the lack of large public datasets for semantic segmentation of anatomical structures, there are only a few studies on this task. In this work, we evaluate the performance of five networks, namely U-Net, U-Net++, DynUNet, UNETR and DeepLabV3+, for segmentation of laparoscopic cholecystectomy images from the recently released CholecSeg8k dataset. To the best of our knowledge, this is the first benchmark performed on this dataset. Training was performed with dice loss. The networks were evaluated on segmentation of 8 anatomical structures and instruments, performance was quantified through the dice coefficient, intersection over union, recall, and precision. Apart from the U-Net, all networks obtained scores similar to each other, with the U-Net++ being the network with the best overall score with a mean Dice value of 0.62. Overall, the results show that there is still room for improvement in the segmentation of anatomical structures from laparoscopic videos. Clinical Relevance- The results of this study show the potential of deep neural networks for the segmentation of anatomical structures in laparoscopic images which can later be incorporated into computer-aided systems for surgeons.

The visualization of medical images with advanced techniques, such as augmented reality and virtual reality, represent a breakthrough for medical professionals. In contrast to more traditional visualization tools lacking 3D capabilities, these systems use the three available dimensions. To visualize medical images in 3D, the anatomical areas of interest must be segmented. Currently, manual segmentation, which is the most commonly used technique, and semi-automatic approaches can be time consuming because a doctor is required, making segmentation for each individual case unfeasible. Using new technologies, such as computer vision and artificial intelligence for segmentation algorithms and augmented and virtual reality for visualization techniques implementation, we designed a complete platform to solve this problem and allow medical professionals to work more frequently with anatomical 3D models obtained from medical imaging. As a result, the Nextmed project, due to the different implemented software applications, permits the importation of digital imaging and communication on medicine (dicom) images on a secure cloud platform and the automatic segmentation of certain anatomical structures with new algorithms that improve upon the current research results. A 3D mesh of the segmented structure is then automatically generated that can be printed in 3D or visualized using both augmented and virtual reality, with the designed software systems. The Nextmed project is unique, as it covers the whole process from uploading dicom images to automatic segmentation, 3D reconstruction, 3D visualization, and manipulation using augmented and virtual reality. There are many researches about application of augmented and virtual reality for medical image 3D visualization; however, they are not automated platforms. Although some other anatomical structures can be studied, we focused on one case: a lung study. Analyzing the application of the platform to more than 1000 dicom images and studying the results with medical specialists, we concluded that the installation of this system in hospitals would provide a considerable improvement as a tool for medical image visualization.

Traditional Chinese Medicine (TCM) has a long history and has been recognized as a popular alternative medicine in western countries. Tongue diagnosis is a significant procedure in computer-aided TCM, where tongue image analysis plays a dominant role. In this paper, we proposed a fully automatic tongue detection and tongue segmentation framework, which is an essential step in computer-aided tongue image analysis. Comparing with other existing methods, our method is fully automatic without any need of adjusting parameters for different images and do not need any initialization.

Hierarchies of artificial neural networks(ANN's) were trained to segment regularly-shaped and constantly-located anatomical structures in x-ray computed tomography (CT) images. These neural networks learned to associate a point in an image with the anatomical structure containing the point using the image pixel intensity values located in a pattern around the point. The single layer ANN and the bilayer and multi-layer hierarchies of neural networks were developed and evaluated. The hierarchical Artificial Neural Networks(HANN's) consisted of a high-level ANN that identified large-scale anatomical structures (e.g., the head or chest), whose result was passed to a group of neural networks that identified smaller structures (e.g., the brain, sinus, soft tissue, skull, bone, or lung) within the large-scale structures. The ANN's were trained to segment and classify images based on different numbers of training images, numbers of sampling points per image, pixel intensity sampling patterns, hidden layer configuration. The experimental results indicate that multi-layer hierarchy of ANN's trained with data collected from multiple image series accurately classified anatomical structures in unknown chest and head CT images.

Hippocampal subfields play important and divergent roles in both memory formation and early diagnosis of many neurological diseases, but automatic subfield segmentation is less explored due to its small size and poor image contrast. In this paper, we propose an automatic learning-based hippocampal subfields segmentation framework using multi-modality 3TMR images, including T1 MRI and resting-state fMRI (rs-fMRI). To do this, we first acquire both 3T and 7T T1 MRIs for each training subject, and then the 7T T1 MRI are linearly registered onto the 3T T1 MRI. Six hippocampal subfields are manually labeled on the aligned 7T T1 MRI, which has the 7T image contrast but sits in the 3T T1 space. Next, corresponding appearance and relationship features from both 3T T1 MRI and rs-fMRI are extracted to train a structured random forest as a multi-label classifier to conduct the segmentation. Finally, the subfield segmentation is further refined iteratively by additional context features and updated relationship features. To our knowledge, this is the first work that addresses the challenging automatic hippocampal subfields segmentation using 3T routine T1 MRI and rs-fMRI. The quantitative comparison between our results and manual ground truth demonstrates the effectiveness of our method. Besides, we also find that (a) multi-modality features significantly improved subfield segmentation performance due to the complementary information among modalities; (b) automatic segmentation results using 3T multimodality images are partially comparable to those on 7T T1 MRI.

Accurate and efficient automated tumor segmentation in breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is highly desirable for computer-aided tumor diagnosis. We propose a novel automatic segmentation framework which incorporates mean-shift smoothing, superpixel-wise classification, pixel-wise graph-cuts partitioning, and morphological refinement. A set of 15 breast DCE-MR images, obtained from the American College of Radiology Imaging Network (ACRIN) 6657 I-SPY trial, were manually segmented to generate tumor masks (as ground truth) and breast masks (as regions of interest). Four state-of-the-art segmentation approaches based on diverse models were also utilized for comparison. Based on five standard evaluation metrics for segmentation, the proposed framework consistently outperformed all other approaches. The performance of the proposed framework was: 1) 0.83 for Dice similarity coefficient, 2) 0.96 for pixel-wise accuracy, 3) 0.72 for VOC score, 4) 0.79 mm for mean absolute difference, and 5) 11.71 mm for maximum Hausdorff distance, which surpassed the second best method (i.e., adaptive geodesic transformation), a semi-automatic algorithm depending on precise initialization. Our results suggest promising potential applications of our segmentation framework in assisting analysis of breast carcinomas.

Fast and Automatic Segmentation of Anatomical Chest Structures from CT Images Based on Image-Guided Lung Biopsy Surgery

Lens arrays enable us to easily achieve inexpensive light field acquisition in comparison with massive camera arrays. However, their micro-lenses often cause insufficient exposure to obtain high quality multi-view videos like camera arrays. In this paper, we propose a novel virtual camera array acquiring dynamic light fields as multi-view videos robustly and inexpensively, where mirror arrays consisting of tangent planes to a paraboloid are effectively combined with a single video camera. First, we demonstrate multi-view imaging of not an object but ordinary scenes with sufficient exposure by using a simple horizontal mirror array implemented as a prototype. Next, we theoretically reveal that there exists smooth extension of our mirror array to the vertical direction, that can be implemented easily by 3D printing, CNC machines, and so on. Finally, our design of light field acquisition is improved for effectively utilizing the whole of its mirror array by introducing an one-way mirror to prevent the camera itself from appearing on multi-view images.

A framework for automatic segmentation in three dimensions of microstructural tomography data

In order to perform preoperative surgical planning, accurate segmentation of anatomical structures in cone-beam computed tomography (CBCT) images is required. However, this image segmentation is often impeded by metal artifacts, and it takes a lot of time due to morphological variability in patients. In this paper, we proposed a deep learning based automatic multi-class segmentation method for anatomical structures in CBCT images containing metal artifacts. Four U-Net based deep learning models were used for anatomical structure segmentation. Each deep learning model was constructed by changing the encoder of U-Net architecture to the backbones (DenseNet121, VGGNet16, ResNet101, and EfficienNetB4). For training and testing our method, we used 20744 CBCT images containing metal artifacts from 30 patient datasets. Experimental results show that the segmentation performances of the mandible, midfacial bone, mandibular canal, and maxillary sinus were achieved F1 scores of 0.912±0.070, 0.880±0.080, 0.687±0.265, and 0.954±0.063 using DenseNet121 with Tversky loss, respectively. Furthermore, our method was able to perform robust and accurate segmentation of anatomical structures in CBCT images containing metal artifacts.

Rapid advances in imaging modalities have increased the importance of image segmentation techniques. These techniques automatically extract data for the anatomical structure of interest and facilitate their quantitative analysis. Here we present a framework for a semi-automatic segmentation method that incorporates a local classifier derived from a neighboring image. Using the local classifier we were able to consider otherwise challenging cases of segmentation merely as two-class classification without any complicated parameters. Our method is simple to implement and easy to operate. We successfully tested our method on computed tomography images.