Showing papers by "Xiaohui Xie published in 2019"

AnatomyNet: Deep learning for fast and fully automated whole-volume segmentation of head and neck anatomy

Abstract: Purpose Radiation therapy (RT) is a common treatment option for head and neck (HaN) cancer. An important step involved in RT planning is the delineation of organs-at-risks (OARs) based on HaN computed tomography (CT). However, manually delineating OARs is time-consuming as each slice of CT images needs to be individually examined and a typical CT consists of hundreds of slices. Automating OARs segmentation has the benefit of both reducing the time and improving the quality of RT planning. Existing anatomy autosegmentation algorithms use primarily atlas-based methods, which require sophisticated atlas creation and cannot adequately account for anatomy variations among patients. In this work, we propose an end-to-end, atlas-free three-dimensional (3D) convolutional deep learning framework for fast and fully automated whole-volume HaN anatomy segmentation. Methods Our deep learning model, called AnatomyNet, segments OARs from head and neck CT images in an end-to-end fashion, receiving whole-volume HaN CT images as input and generating masks of all OARs of interest in one shot. AnatomyNet is built upon the popular 3D U-net architecture, but extends it in three important ways: (a) a new encoding scheme to allow autosegmentation on whole-volume CT images instead of local patches or subsets of slices, (b) incorporating 3D squeeze-and-excitation residual blocks in encoding layers for better feature representation, and (c) a new loss function combining Dice scores and focal loss to facilitate the training of the neural model. These features are designed to address two main challenges in deep learning-based HaN segmentation: (a) segmenting small anatomies (i.e., optic chiasm and optic nerves) occupying only a few slices, and (b) training with inconsistent data annotations with missing ground truth for some anatomical structures. Results We collected 261 HaN CT images to train AnatomyNet and used MICCAI Head and Neck Auto Segmentation Challenge 2015 as a benchmark dataset to evaluate the performance of AnatomyNet. The objective is to segment nine anatomies: brain stem, chiasm, mandible, optic nerve left, optic nerve right, parotid gland left, parotid gland right, submandibular gland left, and submandibular gland right. Compared to previous state-of-the-art results from the MICCAI 2015 competition, AnatomyNet increases Dice similarity coefficient by 3.3% on average. AnatomyNet takes about 0.12 s to fully segment a head and neck CT image of dimension 178 × 302 × 225, significantly faster than previous methods. In addition, the model is able to process whole-volume CT images and delineate all OARs in one pass, requiring little pre- or postprocessing. Conclusion Deep learning models offer a feasible solution to the problem of delineating OARs from CT images. We demonstrate that our proposed model can improve segmentation accuracy and simplify the autosegmentation pipeline. With this method, it is possible to delineate OARs of a head and neck CT within a fraction of a second.

VTNFP: An Image-Based Virtual Try-On Network With Body and Clothing Feature Preservation

Abstract: Image-based virtual try-on systems with the goal of transferring a desired clothing item onto the corresponding region of a person have made great strides recently, but challenges remain in generating realistic looking images that preserve both body and clothing details. Here we present a new virtual try-on network, called VTNFP, to synthesize photo-realistic images given the images of a clothed person and a target clothing item. In order to better preserve clothing and body features, VTNFP follows a three-stage design strategy. First, it transforms the target clothing into a warped form compatible with the pose of the given person. Next, it predicts a body segmentation map of the person wearing the target clothing, delineating body parts as well as clothing regions. Finally, the warped clothing, body segmentation map and given person image are fused together for fine-scale image synthesis. A key innovation of VTNFP is the body segmentation map prediction module, which provides critical information to guide image synthesis in regions where body parts and clothing intersects, and is very beneficial for preventing blurry pictures and preserving clothing and body part details. Experiments on a fashion dataset demonstrate that VTNFP generates substantially better results than state-of-the-art methods.

Clinically applicable deep learning framework for organs at risk delineation in CT images

Abstract: Radiation therapy is one of the most widely used therapies for cancer treatment. A critical step in radiation therapy planning is to accurately delineate all organs at risk (OARs) to minimize potential adverse effects to healthy surrounding organs. However, manually delineating OARs based on computed tomography images is time-consuming and error-prone. Here, we present a deep learning model to automatically delineate OARs in head and neck, trained on a dataset of 215 computed tomography scans with 28 OARs manually delineated by experienced radiation oncologists. On a hold-out dataset of 100 computed tomography scans, our model achieves an average Dice similarity coefficient of 78.34% across the 28 OARs, significantly outperforming human experts and the previous state-of-the-art method by 10.05% and 5.18%, respectively. Our model takes only a few seconds to delineate an entire scan, compared to over half an hour by human experts. These findings demonstrate the potential for deep learning to improve the quality and reduce the treatment planning time of radiation therapy. To keep radiation therapy from damaging healthy tissue, expert radiologists have to segment CT scans into individual organs. A new deep learning-based method for delineating organs in the area of head and neck performs faster and more accurately than human experts.

NoduleNet: Decoupled False Positive Reduction for Pulmonary Nodule Detection and Segmentation

Abstract: Pulmonary nodule detection, false positive reduction and segmentation represent three of the most common tasks in the computer aided analysis of chest CT images. Methods have been proposed for each task with deep learning based methods heavily favored recently. However training deep learning models to solve each task separately may be sub-optimal - resource intensive and without the benefit of feature sharing. Here, we propose a new end-to-end 3D deep convolutional neural net (DCNN), called NoduleNet, to solve nodule detection, false positive reduction and nodule segmentation jointly in a multi-task fashion. To avoid friction between different tasks and encourage feature diversification, we incorporate two major design tricks: (1) decoupled feature maps for nodule detection and false positive reduction, and (2) a segmentation refinement subnet for increasing the precision of nodule segmentation. Extensive experiments on the large-scale LIDC dataset demonstrate that the multi-task training is highly beneficial, improving the nodule detection accuracy by 10.27%, compared to the baseline model trained to only solve the nodule detection task. We also carry out systematic ablation studies to highlight contributions from each of the added components. Code is available at https://github.com/uci-cbcl/NoduleNet.

Automated pulmonary nodule detection using 3D deep convolutional neural networks.

Abstract: Early detection of pulmonary nodules in computed tomography (CT) images is essential for successful outcomes among lung cancer patients. Much attention has been given to deep convolutional neural network (DCNN)-based approaches to this task, but models have relied at least partly on 2D or 2.5D components for inherently 3D data. In this paper, we introduce a novel DCNN approach, consisting of two stages, that is fully three-dimensional end-to-end and utilizes the state-of-the-art in object detection. First, nodule candidates are identified with a U-Net-inspired 3D Faster R-CNN trained using online hard negative mining. Second, false positive reduction is performed by 3D DCNN classifiers trained on difficult examples produced during candidate screening. Finally, we introduce a method to ensemble models from both stages via consensus to give the final predictions. By using this framework, we ranked first of 2887 teams in Season One of Alibaba's 2017 TianChi AI Competition for Healthcare.

NoduleNet: Decoupled False Positive Reductionfor Pulmonary Nodule Detection and Segmentation

Abstract: Pulmonary nodule detection, false positive reduction and segmentation represent three of the most common tasks in the computeraided analysis of chest CT images. Methods have been proposed for eachtask with deep learning based methods heavily favored recently. However training deep learning models to solve each task separately may be sub-optimal - resource intensive and without the benefit of feature sharing. Here, we propose a new end-to-end 3D deep convolutional neural net (DCNN), called NoduleNet, to solve nodule detection, false positive reduction and nodule segmentation jointly in a multi-task fashion. To avoid friction between different tasks and encourage feature diversification, we incorporate two major design tricks: 1) decoupled feature maps for nodule detection and false positive reduction, and 2) a segmentation refinement subnet for increasing the precision of nodule segmentation. Extensive experiments on the large-scale LIDC dataset demonstrate that the multi-task training is highly beneficial, improving the nodule detection accuracy by 10.27%, compared to the baseline model trained to only solve the nodule detection task. We also carry out systematic ablation studies to highlight contributions from each of the added components. Code is available at this https URL.

Automatic Pulmonary Lobe Segmentation Using Deep Learning

Abstract: Pulmonary lobe segmentation is an important task for pulmonary disease related Computer Aided Diagnosis systems (CADs). Classical methods for lobe segmentation rely on successful detection of fissures and other anatomical information such as the location of blood vessels and airways. With the success of deep learning in recent years, Deep Convolutional Neural Network (DCNN) has been widely applied to analyze medical images like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), which, however, requires a large number of ground truth annotations. In this work, we release our manually labeled 50 CT scans which are randomly chosen from the LUNA16 dataset and explore the use of deep learning on this task. We propose pre-processing CT image by cropping region that is covered by the convex hull of the lungs in order to mitigate the influence of noise from outside the lungs. Moreover, we use a hybrid loss function with dice loss to tackle extreme class imbalance issue and focal loss to force model to focus on voxels that are hard to be discriminated. To validate the robustness and performance of our proposed framework trained with a small number of training examples, we further tested our model on CT scans from an independent dataset. Experimental results show the robustness of the proposed approach, which consistently improves performance across different datasets by a maximum of 5.87% as compared to a baseline model. The annotations are public available https://github.com/deep-voxel/automatic_pulmonary_lobe_segmentation_using_deep_learning/ and are for non-commercial use only.

An End-to-End Framework for Integrated Pulmonary Nodule Detection and False Positive Reduction

Abstract: Pulmonary nodule detection using low-dose Computed Tomography (CT) is often the first step in lung disease screening and diagnosis. Recently, algorithms based on deep convolutional neural nets have shown great promise for automated nodule detection. Most of the existing deep learning nodule detection systems are constructed in two steps: a) nodule candidates screening and b) false positive reduction, using two different models trained separately. Although it is commonly adopted, the two-step approach not only imposes significant resource overhead on training two independent deep learning models, but also is sub-optimal because it prevents crosstalk between the two. In this work, we present an end-to-end framework for nodule detection, integrating nodule candidate screening and false positive reduction into one model, trained jointly. We demonstrate that the end-to-end system improves the performance by 3.88% over the two-step approach, while at the same time reducing model complexity by one third and cutting inference time by 3.6 fold. Code will be made publicly available.

Grid-based Evaluation Metrics for Web Image Search

Abstract: Compared to general web search engines, web image search engines display results in a different way. In web image search, results are typically placed in a grid-based manner rather than a sequential result list. In this scenario, users can view results not only in a vertical direction but also in a horizontal direction. Moreover, pagination is usually not (explicitly) supported on image search search engine result pages (SERPs), and users can view results by scrolling down without having to click a “next page” button. These differences lead to different interaction mechanisms and user behavior patterns, which, in turn, create challenges to evaluation metrics that have originally been developed for general web search. While considerable effort has been invested in developing evaluation metrics for general web search, there has been relatively little effort to construct grid-based evaluation metrics. To inform the development of grid-based evaluation metrics for web image search, we conduct a comprehensive analysis of user behavior so as to uncover how users allocate their attention in a grid-based web image search result interface. We obtain three findings: (1) “Middle bias”: Confirming previous studies, we find that image results in the horizontal middle positions may receive more attention from users than those in the leftmost or rightmost positions. (2) “Slower decay”: Unlike web search, users' attention does not decrease monotonically or dramatically with the rank position in image search, especially within a row. (3) “Row skipping”: Users may ignore particular rows and directly jump to results at some distance. Motivated by these observations, we propose corresponding user behavior assumptions to capture users' search interaction processes and evaluate their search performance. We show how to derive new metrics from these assumptions and demonstrate that they can be adopted to revise traditional list-based metrics like Discounted Cumulative Gain (DCG) and Rank-Biased Precision (RBP). To show the effectiveness of the proposed grid-based metrics, we compare them against a number of list-based metrics in terms of their correlation with user satisfaction. Our experimental results show that the proposed grid-based evaluation metrics better reflect user satisfaction in web image search.

Adaptive Graphical Model Network for 2D Handpose Estimation.

Abstract: In this paper, we propose a new architecture called Adaptive Graphical Model Network (AGMN) to tackle the task of 2D hand pose estimation from a monocular RGB image. The AGMN consists of two branches of deep convolutional neural networks for calculating unary and pairwise potential functions, followed by a graphical model inference module for integrating unary and pairwise potentials. Unlike existing architectures proposed to combine DCNNs with graphical models, our AGMN is novel in that the parameters of its graphical model are conditioned on and fully adaptive to individual input images. Experiments show that our approach outperforms the state-of-the-art method used in 2D hand keypoints estimation by a notable margin on two public datasets.

Improving Web Image Search with Contextual Information

Abstract: In web image search, items users search for are images instead of Web pages or online services. Web image search constitutes a very important part of web search. Re-ranking is a trusted technique to improve retrieval effectiveness in web search. Previous work on re-ranking web image search results mainly focuses on intra-query information (e.g., human interactions with the initial list of the current query). Contextual information such as the query sequence and implicit user feedback provided during a search session prior to the current query is known to improve the performance of general web search but has so far not been used in web image search. The differences in result placement and interaction mechanisms of image search make the search process rather different from general Web search engines. Because of these differences, context-aware re-ranking models that have originally been developed for general web search cannot simply be applied to web image search. We propose CARM, a context-aware re-ranking model, a neural network-based framework to re-rank web image search results for a query based on previous interaction behavior in the search session in which the query was submitted. Specifically, we explore a hybrid encoder with an attention mechanism to model intra-query and inter-query user preferences for image results in a two-stage structure. We train context-aware re-ranking model (CARM) to jointly learn query and image representations so as to be able to deal with the multimodal characteristics of web image search. Extensive experiments are carried out on a commercial web image search dataset. The results show that CARM outperforms state-of-the-art baseline models in terms of personalized evaluation metrics. Also, CARM combines the original ranking can improve the original ranking on personalized ranking and relevance estimation. We make the implementation of CARM and relevant datasets publicly available to facilitate future studies.

Single-cell transcriptome profiling reveals neutrophil heterogeneity and orchestrated maturation during homeostasis and bacterial infection

Abstract: Summary The full neutrophil heterogeneity and differentiation landscape remains incompletely characterized. Here we profiled >25,000 differentiating and mature mouse neutrophils using single-cell RNA sequencing to provide a comprehensive transcriptional landscape of neutrophil maturation, function, and fate decision in their steady state and during bacterial infection. Eight neutrophil populations were defined by distinct molecular signatures. The three mature peripheral blood neutrophil subsets arise from distinct maturing bone marrow neutrophil subsets. Driven by both known and uncharacterized transcription factors, neutrophils gradually acquire microbicidal capability as they traverse the transcriptional landscape, representing an evolved mechanism for fine-tuned regulation of an effective but balanced neutrophil response. Bacterial infection reprograms the genetic architecture of neutrophil populations, alters dynamic transition between each subpopulation, and primes neutrophils for augmented functionality without affecting overall heterogeneity. In summary, these data establish a reference model and general framework for studying neutrophil-related disease mechanisms, biomarkers, and therapeutic targets at single-cell resolution. Graphical Abstract Highlights A comprehensive single-cell resolution transcriptional landscape of mouse neutrophil maturation and fate decision under steady-state and bacterial infection conditions. The pathogen clearance machinery in neutrophils is continuously and gradually built during neutrophil differentiation, maturation, and aging, driven by both known and uncharacterized transcription factors. The three mature neutrophil subsets in peripheral blood, including a novel ISG-expressing subset, are derived from distinct bone marrow neutrophil precursors. Bacterial infection reprograms the genetic architecture of neutrophil populations, alters dynamic transition between each subpopulation, and primes neutrophils for augmented functionality without affecting overall neutrophil heterogeneity. Bacterial infection-induced emergency granulopoiesis is mediated by augmented proliferation of early stage neutrophil progenitors and accelerated post-mitotic maturation.

Adaptive Graphical Model Network for 2D Handpose Estimation

Abstract: In this paper, we propose a new architecture called Adaptive Graphical Model Network (AGMN) to tackle the task of 2D hand pose estimation from a monocular RGB image. The AGMN consists of two branches of deep convolutional neural networks for calculating unary and pairwise potential functions, followed by a graphical model inference module for integrating unary and pairwise potentials. Unlike existing architectures proposed to combine DCNNs with graphical models, our AGMN is novel in that the parameters of its graphical model are conditioned on and fully adaptive to individual input images. Experiments show that our approach outperforms the state-of-the-art method used in 2D hand keypoints estimation by a notable margin on two public datasets.

Parallel clustering of single cell transcriptomic data with split-merge sampling on Dirichlet process mixtures.

Abstract: Motivation With the development of droplet based systems, massive single cell transcriptome data has become available, which enables analysis of cellular and molecular processes at single cell resolution and is instrumental to understanding many biological processes. While state-of-the-art clustering methods have been applied to the data, they face challenges in the following aspects: (i) the clustering quality still needs to be improved; (ii) most models need prior knowledge on number of clusters, which is not always available; (iii) there is a demand for faster computational speed. Results We propose to tackle these challenges with Parallelized Split Merge Sampling on Dirichlet Process Mixture Model (the Para-DPMM model). Unlike classic DPMM methods that perform sampling on each single data point, the split merge mechanism samples on the cluster level, which significantly improves convergence and optimality of the result. The model is highly parallelized and can utilize the computing power of high performance computing (HPC) clusters, enabling massive inference on huge datasets. Experiment results show the model outperforms current widely used models in both clustering quality and computational speed. Availability and implementation Source code is publicly available on https://github.com/tiehangd/Para_DPMM/tree/master/Para_DPMM_package. Supplementary information Supplementary data are available at Bioinformatics online.

Neural Multi-Scale Self-Supervised Registration for Echocardiogram Dense Tracking

Abstract: Echocardiography has become routinely used in the diagnosis of cardiomyopathy and abnormal cardiac blood flow. However, manually measuring myocardial motion and cardiac blood flow from echocardiogram is time-consuming and error-prone. Computer algorithms that can automatically track and quantify myocardial motion and cardiac blood flow are highly sought after, but have not been very successful due to noise and high variability of echocardiography. In this work, we propose a neural multi-scale self-supervised registration (NMSR) method for automated myocardial and cardiac blood flow dense tracking. NMSR incorporates two novel components: 1) utilizing a deep neural net to parameterize the velocity field between two image frames, and 2) optimizing the parameters of the neural net in a sequential multi-scale fashion to account for large variations within the velocity field. Experiments demonstrate that NMSR yields significantly better registration accuracy than state-of-the-art methods, such as advanced normalization tools (ANTs) and VoxelMorph, for both myocardial and cardiac blood flow dense tracking. Our approach promises to provide a fully automated method for fast and accurate analyses of echocardiograms.

DNA writing at a single genomic site enables lineage tracing and analog recording in mammalian cells

Abstract: Summary The study of intricate cellular and developmental processes in the context of complex multicellular organisms is difficult because it can require the non-destructive observation of thousands, millions, or even billions of cells deep within an animal. To address this difficulty, several groups have recently reported CRISPR-based DNA recorders that convert transient cellular experiences and processes into changes in the genome, which can then be read by sequencing in high-throughput. However, existing DNA recorders act primarily by erasing DNA: they use the random accumulation of CRISPR-induced deletions to record information. This is problematic because in the limit of progressive deletion, no record remains. Here, we present a new type of DNA recorder that acts primarily by writing new DNA. Our system, called CHYRON (Cell HistorY Recording by Ordered iNsertion), inserts random nucleotides at a single locus in temporal order in vivo and can be applied as an evolving lineage tracer as well as a recorder of user-selected cellular stimuli. As a lineage tracer, CHYRON allowed us to perfectly reconstruct the population lineage relationships among 16 groups of human cells descended from four starting groups that were subject to a series of splitting steps. In this experiment, CHYRON progressively wrote and retained base insertions in 20% percent of cells where the average amount written was 8.4 bp (~14.5 bits), reflecting high information content and density. As a stimulus recorder, we showed that when the CHYRON machinery was placed under the control of a stress-responsive promoter, the frequency and length of writing reflected the dose and duration of the stress. We believe CHYRON represents a conceptual advance in DNA recording technologies where writing rather than erasing becomes the primary mode of information accumulation. With further engineering of CHYRON’s components to increase writing efficiency, CHYRON should lead to single-cell-resolution recording of lineage and other information through long periods of time in complex animals or tumors, advancing the pursuit of a full picture of mammalian development.

An End-to-end Framework For Integrated Pulmonary Nodule Detection and False Positive Reduction

Abstract: Pulmonary nodule detection using low-dose Computed Tomography (CT) is often the first step in lung disease screening and diagnosis. Recently, algorithms based on deep convolutional neural nets have shown great promise for automated nodule detection. Most of the existing deep learning nodule detection systems are constructed in two steps: a) nodule candidates screening and b) false positive reduction, using two different models trained separately. Although it is commonly adopted, the two-step approach not only imposes significant resource overhead on training two independent deep learning models, but also is sub-optimal because it prevents cross-talk between the two. In this work, we present an end-to-end framework for nodule detection, integrating nodule candidate screening and false positive reduction into one model, trained jointly. We demonstrate that the end-to-end system improves the performance by 3.88\% over the two-step approach, while at the same time reducing model complexity by one third and cutting inference time by 3.6 fold. Code will be made publicly available.

Ordered insertional mutagenesis at a single genomic site enables lineage tracing and analog recording in mammalian cells

Abstract: Summary The study of intricate cellular and developmental processes in the context of complex multicellular organisms is difficult because it may require the non-destructive observation of thousands, millions, or even billions of cells deep within an animal. To overcome this difficulty, several groups have recently reported CRISPR-based DNA recorders that convert transient cellular experiences or processes into durable genomic mutations, which can then be read by next-generation sequencing in high-throughput. However, existing DNA recorders rely primarily on the accumulation of CRISPR-induced deletion mutations, which can be problematic because in the limit of progressive deletion mutations, no record remains. Here, we present a high-information DNA recorder that accumulates insertion mutations in temporal order at a single locus. Our recorder, called CHYRON (Cell HistorY Recording by Ordered iNsertion), can be applied as an evolving lineage tracer as well as a cellular stimulus recorder. As a lineage tracer, CHYRON allowed us to perfectly reconstruct the lineage relationships among 16 populations of human cells descended from four starting populations that were subject to a series of splitting steps. In this experiment, CHYRON progressively accumulated and retained insertions in 20% percent of cells such that the average length of insertion generated was 8.4 bp (~15 bits), reflecting high information content. As a stimulus recorder, we show that when the CHYRON machinery is placed under the control of a stress-responsive promoter, the frequency and lengths of insertions reflect the dose and duration of the stress. With further engineering of CHYRON’s components to increase encoding capabilities and reduce loss, CHYRON’s special ability to progressively accumulate insertion mutations should lead to single-cell-resolution recording of lineage and other information through long periods of time in complex animals or tumors, ultimately providing a full picture of mammalian development.

User Behavior Modeling for Web Image Search

Abstract: Web-based image search engines differ from Web search engines greatly. The intents or goals behind human interactions with image search engines are different. In image search, users mainly search images instead of Web pages or online services. It is essential to know why people search for images because user satisfaction may vary as intent varies. Furthermore, image search engines show results differently. For example, grid-based placement is used in image search instead of the linear result list, so that users can browse result list both vertically and horizontally. Different user intents and system UIs lead to different user behavior. Thus, it is hard to apply standard user behavior models developed for general Web search to image search. To better understand user intent and behavior in image search scenarios, we plan to conduct the lab-based user study, field study and commercial search log analysis. We then propose user behavior models based on the observation from data analysis to improve the performance of Web image search engines.

CTCF-mediated Chromatin Structures Dictate the Spatio-temporal Propagation of Replication Foci

Abstract: Mammalian DNA replication is initiated at numerous replication origins, which are clustered into thousands of replication domains (RDs) across the genome. However, it remains unclear whether the replication origins within each RD are activated stochastically. To understand how replication is regulated at the sub-RD level, we directly visualized the spatio-temporal organization, morphology, and in situ epigenetic signatures of individual replication foci (RFi) across S-phase using super-resolution stochastic optical reconstruction microscopy (STORM). Importantly, we revealed a hierarchical radial pattern of RFi propagation that reverses its directionality from early to late S-phase, and is diminished upon caffeine treatment or CTCF knockdown. Together with simulation and bioinformatic analyses, our findings point to a ‘CTCF-organized REplication Propagation’ (CoREP) model. The CoREP model suggests a non-random selection mechanism for replication activation mediated by CTCF at the sub-RD level, as well as the critical involvement of local chromatin environment in regulating replication in space and time.

A Combined Deep Learning-Gradient Boosting Machine Framework for Fluid Intelligence Prediction

Abstract: The ABCD Neurocognitive Prediction Challenge is a community driven competition asking competitors to develop algorithms to predict fluid intelligence score from T1-w MRIs. In this work, we propose a deep learning combined with gradient boosting machine framework to solve this task. We train a convolutional neural network to compress the high dimensional MRI data and learn meaningful image features by predicting the 123 continuous-valued derived data provided with each MRI. These extracted features are then used to train a gradient boosting machine that predicts the residualized fluid intelligence score. Our approach achieved mean square error (MSE) scores of 18.4374, 68.7868, and 96.1806 for the training, validation, and test set respectively.

Automatic Pulmonary Lobe Segmentation Using Deep Learning

Abstract: Pulmonary lobe segmentation is an important task for pulmonary disease related Computer Aided Diagnosis systems (CADs). Classical methods for lobe segmentation rely on successful detection of fissures and other anatomical information such as the location of blood vessels and airways. With the success of deep learning in recent years, Deep Convolutional Neural Network (DCNN) has been widely applied to analyze medical images like Computed Tomography (CT) and Magnetic Resonance Imaging (MRI), which, however, requires a large number of ground truth annotations. In this work, we release our manually labeled 50 CT scans which are randomly chosen from the LUNA16 dataset and explore the use of deep learning on this task. We propose pre-processing CT image by cropping region that is covered by the convex hull of the lungs in order to mitigate the influence of noise from outside the lungs. Moreover, we design a hybrid loss function with dice loss to tackle extreme class imbalance issue and focal loss to force model to focus on voxels that are hard to be discriminated. To validate the robustness and performance of our proposed framework trained with a small number of training examples, we further tested our model on CT scans from an independent dataset. Experimental results show the robustness of the proposed approach, which consistently improves performance across different datasets by a maximum of $5.87\%$ as compared to a baseline model.

Neural Multi-Scale Self-Supervised Registration for Echocardiogram Dense Tracking

Abstract: Echocardiography has become routinely used in the diagnosis of cardiomyopathy and abnormal cardiac blood flow. However, manually measuring myocardial motion and cardiac blood flow from echocar-diogram is time-consuming and error-prone. Computer algorithms that can automatically track and quantify myocardial motion and cardiac blood flow are highly sought after, but have not been very successful due to noise and high variability of echocardiography. In this work, we propose a neural multi-scale self-supervised registration (NMSR) method for automated myocardial and cardiac blood flow dense tracking. NMSR incorporates two novel components: 1) utilizing a deep neural net to parameterize the velocity field between two image frames, and 2) optimizing the parameters of the neural net in a sequential multi-scale fashion to account for large variations within the velocity field. Experiments demonstrate that NMSR yields significantly better registration accuracy than the state-of-the-art methods, such as advanced normalization tools (ANTs) and Voxel Morph, for both myocardial and cardiac blood flow dense tracking. Our approach promises to provide a fully automated method for fast and accurate analyses of echocardiograms.

Sequential Embedding Induced Text Clustering, a Non-parametric Bayesian Approach

Abstract: Current state-of-the-art nonparametric Bayesian text clustering methods model documents through multinomial distribution on bags of words. Although these methods can effectively utilize the word burstiness representation of documents and achieve decent performance, they do not explore the sequential information of text and relationships among synonyms. In this paper, the documents are modeled as the joint of bags of words, sequential features and word embeddings. We proposed Sequential Embedding induced Dirichlet Process Mixture Model (SiDPMM) to effectively exploit this joint document representation in text clustering. The sequential features are extracted by the encoder-decoder component. Word embeddings produced by the continuous-bag-of-words (CBOW) model are introduced to handle synonyms. Experimental results demonstrate the benefits of our model in two major aspects: (1) improved performance across multiple diverse text datasets in terms of the normalized mutual information (NMI); (2) more accurate inference of ground truth cluster numbers with regularization effect on tiny outlier clusters.

Network2Vec: Learning Node Representation Based on Space Mapping in Networks

Abstract: Complex networks represented as node adjacency matrices constrains the application of machine learning and parallel algorithms. To address this limitation, network embedding (i.e., graph representation) has been intensively studied to learn a fixed-length vector for each node in an embedding space, where the node properties in the original graph are preserved. Existing methods mainly focus on learning embedding vectors to preserve nodes proximity, i.e., nodes next to each other in the graph space should also be closed in the embedding space, but do not enforce algebraic statistical properties to be shared between the embedding space and graph space. In this work, we propose a lightweight model, entitled Network2Vec, to learn network embedding on the base of semantic distance mapping between the graph space and embedding space. The model builds a bridge between the two spaces leveraging the property of group homomorphism. Experiments on different learning tasks, including node classification, link prediction, and community visualization, demonstrate the effectiveness and efficiency of the new embedding method, which improves the state-of-the-art model by 19% in node classification and 7% in link prediction tasks at most. In addition, our method is significantly faster, consuming only a fraction of the time used by some famous methods.

A Combined Deep Learning-Gradient Boosting Machine Framework for Fluid Intelligence Prediction

Abstract: The ABCD Neurocognitive Prediction Challenge is a community driven competition asking competitors to develop algorithms to predict fluid intelligence score from T1-w MRIs. In this work, we propose a deep learning combined with gradient boosting machine framework to solve this task. We train a convolutional neural network to compress the high dimensional MRI data and learn meaningful image features by predicting the 123 continuous-valued derived data provided with each MRI. These extracted features are then used to train a gradient boosting machine that predicts the residualized fluid intelligence score. Our approach achieved mean square error (MSE) scores of 18.4374, 68.7868, and 96.1806 for the training, validation, and test set respectively.