Search

Primary care is the largest healthcare delivery platform in the US. Facing the Artificial Intelligence and Machine Learning technology (AI/ML) revolution, the primary care community would benefit from a roadmap revealing priority areas and opportunities for developing and integrating AI/ML-driven clinical tools. This article presents a framework that identifies five domains for AI/ML integration in primary care to support care delivery transformation and achieve the Quintuple Aims of the healthcare system. We concluded that primary care plays a critical role in developing, introducing, implementing, and monitoring AI/ML tools in healthcare and must not be overlooked as AI/ML transforms healthcare.

Monocular 3D face reconstruction from a single image has been an active research topic due to its wide applications. It has been demonstrated that the 3D face can be reconstructed efficiently using a PCA-based subspace model for facial shape representation and facial landmarks for model parameter estimation. However, due to the limited expressiveness of the subspace model and the inaccuracy of landmark detection, most existing methods are not robust to pose and illumination variation. To overcome this limitation, this work proposes a coupled-dictionary model for parametric facial shape representation and a two-stage framework for 3D face reconstruction from a single 2D image by using facial landmarks. Motivated by image super-resolution, the proposed coupled-model consists of two dictionaries for the sparse and the dense 3D facial shapes, respectively. In the first stage, the sparse 3D face is estimated from facial landmarks by using partial least-squares regression. In the second stage, the dense 3D face is reconstructed by 3D super-resolution on the estimated sparse 3D face. Comprehensive experimental evaluations demonstrate that the proposed coupled-dictionary model outperforms the PCA-based subspace model in 3D face modeling accuracy and that the proposed framework achieves much lower reconstruction error on facial images with pose and illumination variations compared to state-of-the-art algorithms. Moreover, qualitative analysis demonstrates that the proposed method is generalizable to different types of data, including facial images, portraits, and facial sketches.

Visual attributes, from simple objects (e.g., backpacks, hats) to soft-biometrics (e.g., gender, height, clothing) have proven to be a powerful representational approach for many applications such as image description and human identification. In this paper, we introduce a novel method to combine the advantages of both multi-task and curriculum learning in a visual attribute classification framework. Individual tasks are grouped after performing hierarchical clustering based on their correlation. The clusters of tasks are learned in a curriculum learning setup by transferring knowledge between clusters. The learning process within each cluster is performed in a multi-task classification setup. By leveraging the acquired knowledge, we speed-up the process and improve performance. We demonstrate the effectiveness of our method via ablation studies and a detailed analysis of the covariates, on a variety of publicly available datasets of humans standing with their full-body visible. Extensive experimentation has proven that the proposed approach boosts the performance by 4% to 10%., Comment: Published in Pattern Recognition

Facial landmark localization is a fundamental module for pose-invariant face recognition. The most common approach for facial landmark detection is cascaded regression, which is composed of two steps: feature extraction and facial shape regression. Recent methods employ deep convolutional networks to extract robust features for each step, while the whole system could be regarded as a deep cascaded regression architecture. In this work, instead of employing a deep regression network, a Globally Optimized Dual-Pathway (GoDP) deep architecture is proposed to identify the target pixels through solving a cascaded pixel labeling problem without resorting to high-level inference models or complex stacked architecture. The proposed end-to-end system relies on distance-aware soft-max functions and dual-pathway proposal-refinement architecture. Results show that it outperforms the state-of-the-art cascaded regression-based methods on multiple in-the-wild face alignment databases. The model achieves 1.84 normalized mean error (NME) on the AFLW database [1], which outperforms 3DDFA [2] by 61.8%. Experiments on face identification demonstrate that GoDP, coupled with DPM-headhunter [3], is able to improve rank-1 identification rate by 44.2% compare to Dlib [4] toolbox on a challenging database.

Developing a local hierarchical ensemble framework for Hierarchical Multi-label Classification (HMC), in which all the structural relationships in the class hierarchy are used to obtain global prediction.Introducing empirical loss minimization into HMC, so that the learned model can capture the most useful information from historical data.Proposing sparse, kernel, and binary constraint HMC models. Traditional flat classification methods (e.g., binary or multi-class classification) neglect the structural information between different classes. In contrast, Hierarchical Multi-label Classification (HMC) considers the structural information embedded in the class hierarchy, and uses it to improve classification performance. In this paper, we propose a local hierarchical ensemble framework for HMC, Fully Associative Ensemble Learning (FAEL). We model the relationship between each class nodes global prediction and the local predictions of all the class nodes as a multi-variable regression problem with Frobenius norm or l1 norm regularization. It can be extended using the kernel trick, which explores the complex correlation between global and local prediction. In addition, we introduce a binary constraint model to restrict the optimal weight matrix learning. The proposed models have been applied to image annotation and gene function prediction datasets with tree structured class hierarchy and large scale visual recognition dataset with Direct Acyclic Graph (DAG) structured class hierarchy. The experimental results indicate that our models achieve better performance when compared with other baseline methods.

Objective Study the impact of local policies on near-future hospitalization and mortality rates. Materials and Methods We introduce a novel risk-stratified SIR-HCD model that introduces new variables to model the dynamics of low-contact (e.g., work from home) and high-contact (e.g., work on-site) subpopulations while sharing parameters to control their respective R 0 ( t ) over time. We test our model on data of daily reported hospitalizations and cumulative mortality of COVID-19 in Harris County, Texas, from May 1, 2020, until October 4, 2020, collected from multiple sources (USA FACTS, U.S. Bureau of Labor Statistics, Southeast Texas Regional Advisory Council COVID-19 report, TMC daily news, and Johns Hopkins University county-level mortality reporting). Results We evaluated our model’s forecasting accuracy in Harris County, TX (the most populated county in the Greater Houston area) during Phase-I and Phase-II reopening. Not only does our model outperform other competing models, but it also supports counterfactual analysis to simulate the impact of future policies in a local setting, which is unique among existing approaches. Discussion Mortality and hospitalization rates are significantly impacted by local quarantine and reopening policies. Existing models do not directly account for the effect of these policies on infection, hospitalization, and death rates in an explicit and explainable manner. Our work is an attempt to improve prediction of these trends by incorporating this information into the model, thus supporting decision-making. Conclusion Our work is a timely effort to attempt to model the dynamics of pandemics under the influence of local policies.

In this paper, we address the problem of image set classification, where each set contains a different number of images acquired from the same subject. In most of the existing literature, each image set is modeled using all its available samples. As a result, the corresponding time and storage costs are high. To address this problem, we propose a joint prototype and metric learning approach. The prototypes are learned to represent each gallery image set using fewer samples without affecting the recognition performance. A Mahalanobis metric is learned simultaneously to measure the similarity between sets more accurately. In particular, each gallery set is represented as a regularized affine hull spanned by the learned prototypes. The set-to-set distance is optimized via updating the prototypes and the Mahalanobis metric in an alternating manner. To highlight the importance of representing image sets using fewer samples, we analyzed the corresponding test time complexity with respect to the number of images used per set. Experimental results using YouTube Celebrity, YouTube Faces, and ETH-80 datasets illustrate the efficiency on the task of video face recognition, and object categorization. A Set-based Prototype and Metric Learning framework is proposed for image set classification.It can represent gallery image set using fewer but more discriminative prototypes.It is robust to the changes of feature length.It works better with fewer prototypes.It works better with k-means initialization than the random initialization.

Describes a conceptual framework for 3D-2D (or 2D-3D) face recognition.Proposes a novel 3D-2D system for 2D image face recognition from 3D datasets.Proposes a method to build subject-specific 3D gallery models, using 3D+2D data, and a method for model-based, texture representation and relighting.3D-2D recognition surpasses 2D-2D on challenging 2D+3D data with pose and illumination variations, and can approximate 3D-3D, shape-based similarity methods.Representation and normalization using 3D models can compensate for non-frontal poses or different lighting conditions. In this paper, we propose a 3D-2D framework for face recognition that is more practical than 3D-3D, yet more accurate than 2D-2D. For 3D-2D face recognition, the gallery data comprises of 3D shape and 2D texture data and the probes are arbitrary 2D images. A 3D-2D system (UR2D) is presented that is based on a 3D deformable face model that allows registration of 3D and 2D data, face alignment, and normalization of pose and illumination. During enrollment, subject-specific 3D models are constructed using 3D+2D data. For recognition, 2D images are represented in a normalized image space using the gallery 3D models and landmark-based 3D-2D projection estimation. A method for bidirectional relighting is applied for non-linear, local illumination normalization between probe and gallery textures, and a global orientation-based correlation metric is used for pairwise similarity scoring. The generated, personalized, pose- and light- normalized signatures can be used for one-to-one verification or one-to-many identification. Results for 3D-2D face recognition on the UHDB11 3D-2D database with 2D images under large illumination and pose variations support our hypothesis that, in challenging datasets, 3D-2D outperforms 2D-2D and decreases the performance gap against 3D-3D face recognition. Evaluations on FRGC v2.0 3D-2D data with frontal facial images, demonstrate that the method can generalize to databases with different and diverse illumination conditions.

Review of the recent literature in 3D human pose estimation from RGB images and videos.Release of a challenging, publicly available, 3D pose estimation synthetic dataset.Extensive experimental evaluation of some representative state-of-the-art methods. Estimating the pose of a human in 3D given an image or a video has recently received significant attention from the scientific community. The main reasons for this trend are the ever increasing new range of applications (e.g., human-robot interaction, gaming, sports performance analysis) which are driven by current technological advances. Although recent approaches have dealt with several challenges and have reported remarkable results, 3D pose estimation remains a largely unsolved problem because real-life applications impose several challenges which are not fully addressed by existing methods. For example, estimating the 3D pose of multiple people in an outdoor environment remains a largely unsolved problem. In this paper, we review the recent advances in 3D human pose estimation from RGB images or image sequences. We propose a taxonomy of the approaches based on the input (e.g., single image or video, monocular or multi-view) and in each case we categorize the methods according to their key characteristics. To provide an overview of the current capabilities, we conducted an extensive experimental evaluation of state-of-the-art approaches in a synthetic dataset created specifically for this task, which along with its ground truth is made publicly available for research purposes. Finally, we provide an in-depth discussion of the insights obtained from reviewing the literature and the results of our experiments. Future directions and challenges are identified.

Environmental factors affecting nutrient availability during development can have long-term effects including predisposition to diseases later in life. In order to identify chemicals in the environment capable of altering nutrient mobilization, we analyzed yolk malabsorption in the zebrafish embryo, which relies on maternally-derived yolk for nutrition during the first week of life. Embryos of the transgenic zebrafish line HGn50D, which fluoresce in the yolk syncytial layer, the heart, and the eyes of the zebrafish, were exposed from two to five days post fertilization to nine chemicals, including pesticides (prochloraz, imazalil, and butralin), pharmaceuticals (clofibrate and gemfibrozil), flame retardants (tetrabromobisphenol A, and tetrachlorobisphenol A), a surfactant (perfluorooctanoic acid) and a biocide (tributyltin). We developed a software package named ZebRA to automatically and accurately segment and quantify the area of the fluorescing yolk in images captured at the end of the treatment period. Based on this quantification, we found that prochloraz decreased yolk absorption, while butralin, tetrabromobisphenol A, tetrachlorobisphenol A and tributyltin, at the higher concentrations tested, increased yolk absorption. Given the number and variety of industrial chemicals in commerce today, development of automated image processing allowing for high-speed analysis and accurate quantification of biological effects is an important step for enabling high throughput screening to identify chemicals adversely impacting nutrient absorption.

Facial landmark detection is a crucial first step in facial analysis for biometrics and numerous other applications. However, it has proved to be a very challenging task due to the numerous sources of variation in 2D and 3D facial data. Although landmark detection based on descriptors of the 2D and 3D appearance of the face has been extensively studied, the fusion of such feature descriptors is a relatively under-studied issue. In this paper, a novel generalized framework for combining facial feature descriptors is presented, and several feature fusion schemes are proposed and evaluated. The proposed framework maps each feature into a similarity score and combines the individual similarity scores into a resultant score, used to select the optimal solution for a queried landmark. The evaluation of the proposed fusion schemes for facial landmark detection clearly indicates that a quadratic distance to similarity mapping in conjunction with a root mean square rule for similarity fusion achieves the best performance in accuracy, efficiency, robustness and monotonicity.

We propose a graph based algorithm with social cues to discover interacting groups.We propose a novel descriptor to recognize group activity in top-down approach.The descriptor capturing the motion and interaction of people in discovered group.The advantage of social cue in better understanding of activities in crowded scene. This paper presents a novel and efficient framework for group activity analysis. People in a scene can be intuitively represented by an undirected graph where vertices are people and the edges between two people are weighted by how much they are interacting. Social signaling cues are used to describe the degree of interaction between people. We propose a graph-based clustering algorithm to discover interacting groups in crowded scenes. Two social signaling cues are presented and compared for group discovery. The grouping of people in the scene serves to isolate the groups engaged in the dominant activity, effectively eliminating dataset contamination. Using discovered interacting groups, we create a descriptor capturing the motion and interaction of people within it. A bag-of-words approach is used to represent group activity and a SVM classifier is used for activity recognition. The proposed framework is evaluated in its ability to discover interacting groups and perform group activity recognition using two public datasets. The overall recognition system is compared to a baseline top-down model to understand the impact of social cues for activity recognition. The results of both the steps show that our method outperforms state-of-the-art methods for group discovery and achieves recognition rates comparable to state-of-the-art methods for group activity recognition.

Human interaction dynamics are known to play an important role in the development of robust pedestrian trackers that are needed for a variety of applications in video surveillance. Traditional approaches to pedestrian tracking assume that each pedestrian walks independently and the tracker predicts the location based on an underlying motion model, such as a constant velocity or autoregressive model. Recent approaches have begun to leverage interaction, especially by modeling the repulsion forces among pedestrians to improve motion predictions. However, human interaction is more complex and is influenced by multiple social effects. This motivates the use of a more complex human interaction model for pedestrian tracking. In this paper, we propose a novel human tracking method by modeling complex social interactions. We present an algorithm that decomposes social interactions into multiple potential interaction modes. We integrate these multiple social interaction modes into an interactive Markov Chain Monte Carlo tracker and demonstrate how the developed method translates into a more informed motion prediction, resulting in robust tracking performance. We test our method on videos from unconstrained outdoor environments and evaluate it against common multi-object trackers.

Contains fulltext : 136858.pdf (Publisher’s version ) (Open Access) This paper describes an evaluation framework that allows a standardized and quantitative comparison of IVUS lumen and media segmentation algorithms. This framework has been introduced at the MICCAI 2011 Computing and Visualization for (Intra)Vascular Imaging (CVII) workshop, comparing the results of eight teams that participated. We describe the available data-base comprising of multi-center, multi-vendor and multi-frequency IVUS datasets, their acquisition, the creation of the reference standard and the evaluation measures. The approaches address segmentation of the lumen, the media, or both borders; semi- or fully-automatic operation; and 2-D vs. 3-D methodology. Three performance measures for quantitative analysis have been proposed. The results of the evaluation indicate that segmentation of the vessel lumen and media is possible with an accuracy that is comparable to manual annotation when semi-automatic methods are used, as well as encouraging results can be obtained also in case of fully-automatic segmentation. The analysis performed in this paper also highlights the challenges in IVUS segmentation that remains to be solved.

A learning-based framework for action representation and recognition relying on the description of an action by time series of optical flow motion features is presented. In the learning step, the motion curves representing each action are clustered using Gaussian mixture modeling (GMM). In the recognition step, the optical flow curves of a probe sequence are also clustered using a GMM, then each probe sequence is projected onto the training space and the probe curves are matched to the learned curves using a non-metric similarity function based on the longest common subsequence, which is robust to noise and provides an intuitive notion of similarity between curves. Alignment between the mean curves is performed using canonical time warping. Finally, the probe sequence is categorized to the learned action with the maximum similarity using a nearest neighbor classification scheme. We also present a variant of the method where the length of the time series is reduced by dimensionality reduction in both training and test phases, in order to smooth out the outliers, which are common in these type of sequences. Experimental results on KTH, UCF Sports and UCF YouTube action databases demonstrate the effectiveness of the proposed method.

Facial expression analysis has interested many researchers in the past decade due to its potential applications in various fields such as human-computer interaction, psychological studies, and facial animation. Three-dimensional facial data has been proven to be insensitive to illumination condition and head pose, and has hence gathered attention in recent years. In this paper, we focus on discrete expression classification using 3D data from the human face. The paper is divided in two parts. In the first part, we present improvement to the fitting of the Annotated Face Model (AFM) so that a dense point correspondence can be found in terms of both position and semantics among static 3D face scans or frames in 3D face sequences. Then, an expression recognition framework on static 3D images is presented. It is based on a Point Distribution Model (PDM) which can be built on different features. In the second part of this article, a systematic pipeline that operates on dynamic 3D sequences (4D datasets or 3D videos) is proposed and alternative modules are investigated as a comparative study. We evaluated both 3D and 4D Facial Expression Recognition pipelines on two publicly available facial expression databases and obtained promising results.

This paper presents a novel and efficient framework for human action recognition based on modeling the motion of human body-parts. Intuitively, a collective understanding of human body-part movements can lead to better understanding and representation of any human action. In this paper, we propose a generative representation of the motion of human body-parts to learn and classify human actions. The proposed representation combines the advantages of both local and global representations, encoding the relevant motion information as well as being robust to local appearance changes. Our work is motivated by the pictorial structures model and the framework of sparse representations for recognition. Human body-part movements are represented efficiently through quantization in the polar space. The key discrimination within each action is efficiently encoded by sparse representation for classification. The proposed framework is evaluated on both the KTH and the UCF Sport action datasets and results compared against several state-of-the-art methods.

This paper presents a framework for automatic face recognition based on a silhouetted face profile (URxD-PV). Previous research has demonstrated the high discriminative potential of this biometric. Compared to traditional approaches in profile-based recognition, our approach is not limited to only standard side-view faces. We propose to explore the feature space of profiles under various rotations with the aid of a 3D face model. In the enrollment mode, 3D data of subjects are acquired and used to create profiles under different rotations. The features extracted from these profiles are used to train a classifier. In the identification mode, the profile is extracted from the side-view image and its metadata is matched with the gallery metadata. We validate the accuracy of URxD-PV using data from publicly available databases.

As biomedical images and volumes are being collected at an increasing speed, there is a growing demand for efficient means to organize spatial information for comparative analysis. In many scenarios, such as determining gene expression patterns by in situ hybridization, the images are collected from multiple subjects over a common anatomical region, such as the brain. A fundamental challenge in comparing spatial data from different images is how to account for the shape variations among subjects, which make direct image-to-image comparisons meaningless. In this paper, we describe subdivision meshes as a geometric means to efficiently organize 2D images and 3D volumes collected from different subjects for comparison. The key advantages of a subdivision mesh for this purpose are its light-weight geometric structure and its explicit modeling of anatomical boundaries, which enable efficient and accurate registration. The multi-resolution structure of a subdivision mesh also allows development of fast comparison algorithms among registered images and volumes.

As the accuracy of biometrics improves, it is getting increasingly hard to push the limits using a single modality. In this paper, a unified approach that fuses three-dimensional facial and ear data is presented. An annotated deformable model is fitted to the data and a geometry image is extracted. Wavelet coefficients are computed from the geometry image and used as a biometric signature. The method is evaluated using the largest publicly available database and achieves 99.7% rank-one recognition rate. The state-of-the-art accuracy of the multimodal fusion is attributed to the low correlation between the individual differentiability of the two modalities.

Sectioning tissues for optical microscopy often introduces upon the resulting sections distortions that make 3D reconstruction difficult. Here we present an automatic method for producing a smooth 3D volume from distorted 2D sections in the absence of any undistorted references. The method is based on pairwise elastic image warps between successive tissue sections, which can be computed by 2D image registration. Using a Gaussian filter, an average warp is computed for each section from the pairwise warps in a group of its neighboring sections. The average warps deform each section to match its neighboring sections, thus creating a smooth volume where corresponding features on successive sections lie close to each other. The proposed method can be used with any existing 2D image registration method for 3D reconstruction. In particular, we present a novel image warping algorithm based on dynamic programming that extends Dynamic Time Warping in 1D speech recognition to compute pairwise warps between high-resolution 2D images. The warping algorithm efficiently computes a restricted class of 2D local deformations that are characteristic between successive tissue sections. Finally, a validation framework is proposed and applied to evaluate the quality of reconstruction using both real sections and a synthetic volume.

In this paper, we give an estimation for the smoothness of a new family of symmetric orthonormal wavelets arising from Hermite Distributed Approximating Functionals (HDAFs). Their corresponding associated low pass filters are symmetric with respect to the origin.

In this paper, we report our ongoing research into multi-sensory investigation of geoscientific data. Our geoscientific data investigation system (GDIS) integrates three-dimensional, interactive computer graphics, touch (haptics) and real-time sonification into a multi-sensory virtual environment. GDIS has been used to investigate geological structures derived from high-resolution bathymetry-maps of the Mid-Atlantic Ridge. It helped us obtain improved models of the mega-mullions, which promise to grant insight into the deeper structures of the ocean floor. Haptic force feedback was used to precisely digitize line-features on three-dimensional morphology and to feel surface properties via varying friction settings; additional, overlapping data can be perceived via sound (sonification). We also report on the results of a psycho-acoustic study about the absolute recognition of sound signals, and on the actual feedback that we have received from a number of geoscientists during a recent major geoscience conference.

In this paper, we present a four-step technique for simultaneously estimating a human's anthropometric measurements (up to a scale parameter) and pose from a single uncalibrated image. The user initially selects a set of image points that constitute the projection of selected landmarks. Using this information, along with a priori statistical information about the human body, a set of plausible segment length estimates is produced. In the third step, a set of plausible poses is inferred using a geometric method based on joint limit constraints. In the fourth step, pose and anthropometric measurements are obtained by minimizing an appropriate cost function subject to the associated constraints. The novelty of our approach is the use of anthropometric statistics to constrain the estimation process that allows the simultaneous estimation of both anthropometry and pose. We demonstrate the accuracy, advantages, and limitations of our method for various classes of both synthetic and real input data.

We describe a case where a mild carotid atherosclerotic plaque was assessed by contrast enhanced harmonic ultrasonography and image analysis. Quantitative indices of plaque echogenicity were determined prior and after the injection of microbubbles. Changes in plaque echogenicity were detected possibly due to the flow of microbubbles through vasa vasorum within the plaque and at the plaque base at the adventitial level. Future histological studies remain to be done to link the presence and the extent of plaque and adventitial neovascularization with the visual and quantitative findings derived by contrast enhanced harmonic ultrasound and image analysis.