Optical coherence tomography is a crucial imaging technique for the detection and analysis of retinal diseases. Precise classification of optical coherence tomography images helps ophthalmologists and healthcare providers design personalized treatment plans in clinical practice. In this paper, we focus on optical coherence tomography image classification for seven types of retinal diseases and normal retina. Although existing deep neural networks could be applied to optical coherence tomography images for the classification, the features were extracted within same hyperspace, causing "feature congestion". Moreover, the class of normal retina was regarded as a "type of retinal disease", impeding the extraction of true imaging structures for retinal diseases. To deal with the two issues, we innovate a deep neural network module to enhance imaging features so that the enhanced features are more distinct for classification. Consistent to medical findings, we propose two domains of retinal diseases and anchor imaging features onto cross-domains. We tested and evaluated our model on two datasets for eight-classes and four-classes classification, respectively. Our experimental results demonstrated that the proposed module outperforms state-of-the-art methods. We also conducted ablation studies and sensitivity tests for comprehensive evaluation of our method.

We present a role-aware virtual agent navigational interaction that generates consistent, role-aligned movement behaviors. Our approach leverages Multimodal Large Language Models (MLLMs) to interpret multimodal inputs including scene information, user state, and high-level language role instruction, producing discrete navigation decisions and stylized planning path. Our approach enables virtual agents to behave consistently with narrative roles and respond to dynamic actions, such as playing a hide-and-seek taking into account the agent's role and the user's possible intention. Our approach demonstrates how MLLMs can go beyond language-based interaction to support embodied, spatial, and role-aware agent behaviors in immersive environments such as augmented reality.

3D garment simulation is a critical component for producing cloth-based graphics. Recent advancements in graph neural networks (GNNs) offer a promising approach for efficient garment simulation. However, GNNs require extensive message-passing to propagate information such as physical forces and maintain contact awareness across the entire garment mesh, which becomes computationally inefficient at higher resolutions. To address this, we devise a novel GNN-based mesh learning framework with two key components to extend the message-passing range with minimal overhead, namely the Laplacian-Smoothed Dual Message-Passing (LSDMP) and the Geodesic Self-Attention (GSA) modules. LSDMP enhances message-passing with a Laplacian features smoothing process, which efficiently propagates the impact of each vertex to nearby vertices. Concurrently, GSA introduces geodesic distance embeddings to represent the spatial relationship between vertices and utilises attention mechanisms to capture global mesh information. The two modules operate in parallel to ensure both short- and long-range mesh modelling. Extensive experiments demonstrate the state-of-the-art performance of our method, requiring fewer layers and lower inference latency.1

In this paper, we investigate the use of multimodal large language models (MLLMs) for generating virtual activities, leveraging the integration of vision-language modalities to enable the interpretation of virtual environments. Our approach recognizes and abstracts key scene elements including scene layouts, semantic contexts, and object identities with MLLMs' multimodal reasoning capabilities. By correlating these abstractions with massive knowledge about human activities, MLLMs are capable of generating adaptive and contextually relevant virtual activities. We propose a structured framework to articulate abstract activity descriptions, emphasizing detailed multi-character interactions within virtual spaces. Utilizing the derived high-level contexts, our approach accurately positions virtual characters and ensures that their interactions and behaviors are realistically and contextually appropriate through strategic optimization. Experiment results demonstrate the effectiveness of our approach, providing a novel direction for enhancing the realism and context-awareness in simulated virtual environments.

Radar imaging provides non-contact, privacy-preserving, and environmentally robust monitoring for continuous human motion recognition (HMR) by leveraging diverse information embedded in various radar signal domains. However, current research has not effectively integrated multi-radar and multi-domain imaging to fully exploit the benefits of distributed radar systems. To bridge this gap, we propose a multi-radar, multi-domain parallel cross-attention model with four key components: intra-domain cross-radar weight sharing encoders specific to each domain for consistent feature extraction and parameter reduction, domain-level parallel cross-attention (DLPCAN) modules to fuse domain-specific features and enhance feature representation robustness in each radar, a source-level attention fusion (SLAF) module to highlight significant features from multiple radar inputs, and two bi-directional gated recurrent unit (BiGRU) modules to capture temporal information. The model is trained using connectionist temporal classification (CTC) loss for effective sequence prediction. By integrating data from multiple radar nodes and domains, our approach significantly improves continuous HMR performance compared to single radar systems and single domain data. Comparative evaluations demonstrate that our model outperforms state-of-the-art radar imaging-based HMR solutions.

Radio frequency (RF) signals have been proved to be flexible for human silhouette segmentation (HSS) under complex environments. Existing studies are mainly based on a one-shot approach, which lacks a coherent projection ability from the RF domain. Additionally, the spatio-temporal patterns have not been fully explored for human motion dynamics in HSS. Therefore, we propose a two-stage Sequential Diffusion Model (SDM) to progressively synthesize high-quality segmentation jointly with the considerations on motion dynamics. Cross-view transformation blocks are devised to guide the diffusion model in a multi-scale manner for comprehensively characterizing human related patterns in an individual frame such as directional projection from signal planes. Moreover, spatio-temporal blocks are devised to fine-tune the frame-level model to incorporate spatio-temporal contexts and motion dynamics, enhancing the consistency of the segmentation maps. Comprehensive experiments on a public benchmark - HIBER demonstrate the state-of-the-art performance of our method with an IoU 0.732. Our code is available at https://github.com/ph-w2000/SDM.

Deep neural networks (DNNs) have been increasingly used in recent years to achieve higher inference accuracy; however, implementing deeper networks in edge-computing environments can be challenging. Current methods for accelerating CNN inference focus on finding a trade-off between accuracy and latency under an assumed uniform distribution, ignoring the impact of real-world data distributions. To address this, we propose the Coarse-to-Fine (C2F) framework, which includes a C2F model and a corresponding C2F inference architecture to better exploit distributional differences in the edge environment. The C2F model is derived from various adaptations of Convolutional Neural Networks (CNNs). By deconstructing the original CNNs into multiple smaller models, the C2F model increases memory consumption within an acceptable range to improve inference speed without sacrificing accuracy. The C2F architecture deploys C2F models more logically in complex edge environments, reducing inference costs and memory consumption. We conduct experiments on the CIFAR dataset with different backbone networks and show that our C2F framework can simultaneously reduce latency and improve accuracy in complex edge environments.

Universal lesion detection (ULD) has great value in clinical practice as it can detect various lesions across multiple organs. Deep learning-based detectors have great potential but require high-quality annotated training data. In practice, due to cost, expertise requirements, and the diverse nature of lesions, incomplete annotations are encountered. Directly training ULD detectors under this condition can yield suboptimal results. Leading pseudo-label methods rely on a dynamic lesion-mining mechanism operating at the mini-batch level to address this issue. However, the quality of mined lesions is inconsistent across different iterations, potentially limiting performance enhancement. Inspired by the observation that deep models learn concepts with increasing complexity, we propose an exploratory-training-based ULD (ET-ULD) method to assess the reliability of mined lesions over time. Our approach uses a teacher-student detection model where the teacher mines suspicious lesions, which are then combined with incomplete annotations to train the student. On top of that, we design a bounding-box bank to record the mining timestamps. Each image is trained in several rounds, allowing us to get a sequence of timestamps for the mined lesions. If a mined lesion consistently appears, it is likely to be a true lesion, otherwise, it may just be a noise. This serves as a crucial criterion for selecting reliable mined lesions for retraining. Experimental results show that ET-ULD surpass existing state-of-the-art methods on two distinct lesion image datasets. Notably, on the DeepLesion dataset, ET-ULD achieved a 5.4% improvement in Average Precision (AP) over the previous methods, demonstrating its superior performance.

The recent popularity of augmented reality (AR) devices has enabled players to participate in interactive narratives through virtual events and characters populated in a real-world environment, where different actions may lead to different story branches. In this paper, we propose a novel approach to adapt narratives to real spaces for AR experiences. Our optimization-based approach automatically assigns contextually compatible locations to story events, synthesizing a navigation graph to guide players through different story branches while considering their walking experiences. We validated the effectiveness of our approach for adapting AR narratives to different scenes through experiments and user studies.

The recent popularity of consumer-grade virtual reality devices has enabled users to experience immersive shopping in virtual environments. As in a real-world store, the placement of products in a virtual store should appeal to shoppers, which could be time-consuming, tedious, and non-trivial to create manually. Thus, this work introduces a novel approach for automatically optimizing product placement in virtual stores. Our approach considers product exposure and spatial constraints, applying an optimizer to search for optimal product placement solutions. We conducted qualitative scene rationality and quantitative product exposure experiments to validate our approach with users. The results show that the proposed approach can synthesize reasonable product placements and increase product exposures for different virtual stores.

We present an approach to generate virtual activity snippets, which comprise sequenced keyframes of multi-character, multi-object interaction scenarios in 3D environments, by learning from recordings of human-scene interactions. The generation consists of two stages. First, we use a sequential deep graph generative model with a temporal module to iteratively generate keyframe descriptions, which represent abstract interactions using graphs, while preserving spatial-temporal relations through the activities. Second, we devise an optimization framework to instantiate the activity snippets in virtual 3D environments guided by the generated keyframe descriptions. Our approach optimizes the poses of character and object instances encoded by the graph nodes to satisfy the relations and constraints encoded by the graph edges. The instantiation process includes a coarse 2D optimization followed by a fine 3D optimization to effectively explore the complex solution space for placing and posing the instances. Through experiments and a perceptual study, we applied our approach to generate plausible activity snippets under different settings.

We present a novel interactive augmented reality (AR) storytelling approach guided by indoor scene semantics. Our approach automatically populates virtual contents in real-world environments to deliver AR stories, which match both the story plots and scene semantics. During the storytelling process, a player can participate as a character in the story. Meanwhile, the behaviors of the virtual characters and the placement of the virtual items adapt to the player's actions. An input raw story is represented as a sequence of events, which contain high-level descriptions of the characters' states, and is converted into a graph representation with automatically supplemented low-level spatial details. Our hierarchical story sampling approach samples realistic character behaviors that fit the story contexts through optimizations; and an animator, which estimates and prioritizes the player's actions, animates the virtual characters to tell the story in AR. Through experiments and a user study, we validated the effectiveness of our approach for AR storytelling in different environments.

We present a novel approach for synthesizing scene-aware virtual reality teleport graphs, which facilitate navigation in indoor virtual environments by suggesting desirable teleport positions. Our approach analyzes panoramic views at candidate teleport positions by extracting scene perception graphs, which encode scene perception relationships between the observer and the surrounding objects, and predict how desirable the views at these positions are. We train a graph convolutional model to predict the scene perception scores of different teleport positions. Based on such predictions, we apply an optimization approach to sample a set of desirable teleport positions while considering other navigation properties such as coverage and connectivity to synthesize a teleport graph. Using teleport graphs, users can navigate virtual environments efficaciously. We demonstrate our approach for synthesizing teleport graphs for common indoor scenes. By conducting a user study, we validate the efficacy and desirability of navigating virtual environments via the synthesized teleport graphs. We also extend our approach to cope with different constraints, user preferences, and practical scenarios.

We propose Bayesian Inverse Reinforcement Learning with Failure (BIRLF), which makes use of failed demonstrations that were often ignored or filtered in previous methods due to the difficulties to incorporate them in addition to the successful ones. Specifically, we leverage halfspaces derived from policy optimality conditions to incorporate failed demonstrations under Bayesian Inverse Reinforcement Learning (BIRL) framework. Under the continuous control setting, the reward function and policy are learned in an alternative manner, both of which are estimated by function approximators to guarantee the learning ability. Our approach is formulated as a model-free Inverse Reinforcement Learning (IRL) method that naturally accommodates more complex environments with continuous state and action spaces. In experiments, we demonstrate the proposed method in a virtual grasping task, achieving a significant performance boost compared to existing methods.

Conventional low-level feature-based saliency detection methods tend to use nonrobust prior knowledge and do not perform well in complex or low-contrast images. In this paper, to address these issues in existing methods, we propose a novel deep neural network (DNN)-based dense and sparse labeling (DSL) framework for saliency detection. DSL consists of three major steps, namely, dense labeling (DL), sparse labeling (SL), and deep convolutional (DC) network. The DL and SL steps conduct initial saliency estimations with macro object contours and low-level image features, respectively, which effectively approximate the location of the salient object and generate accurate guidance channels for the DC step; the DC step, on the other hand, takes in the results of DL and SL, establishes a six-channeled input data structure (including local superpixel information), and conducts accurate final saliency classification. Our DSL framework exploits the saliency estimation guidance from both macro object contours and local low-level features, as well as utilizing the DNN for high-level saliency feature extraction. Extensive experiments are conducted on six well-recognized public data sets against 16 state-of-the-art saliency detection methods, including ten conventional feature-based methods and six learning-based methods. The results demonstrate the superior performance of DSL on various challenging cases in terms of both accuracy and robustness.

In recent saliency detection research, many graph-based algorithms have applied boundary priors as background queries, which may generate completely "reversed" saliency maps if the salient objects are on the image boundaries. Moreover, these algorithms usually depend heavily on pre-processed superpixel segmentation, which may lead to notable degradation in image detail features. In this paper, a novel saliency detection method is proposed to overcome the above issues. First, we propose a saliency reversion correction process, which locates and removes the boundary-adjacent foreground superpixels, and thereby increases the accuracy and robustness of the boundary prior-based saliency estimations. Second, we propose a regularized random walk ranking model, which introduces prior saliency estimation to every pixel in the image by taking both region and pixel image features into account, thus leading to pixel-detailed and superpixel-independent saliency maps. Experiments are conducted on four well-recognized data sets; the results indicate the superiority of our proposed method against 14 state-of-the-art methods, and demonstrate its general extensibility as a saliency optimization algorithm. We further evaluate our method on a new data set comprised of images that we define as boundary adjacent object saliency, on which our method performs better than the comparison methods.

The segmentation of skin lesions in dermoscopic images is a fundamental step in automated computer-aided diagnosis of melanoma. Conventional segmentation methods, however, have difficulties when the lesion borders are indistinct and when contrast between the lesion and the surrounding skin is low. They also perform poorly when there is a heterogeneous background or a lesion that touches the image boundaries; this then results in under- and oversegmentation of the skin lesion. We suggest that saliency detection using the reconstruction errors derived from a sparse representation model coupled with a novel background detection can more accurately discriminate the lesion from surrounding regions. We further propose a Bayesian framework that better delineates the shape and boundaries of the lesion. We also evaluated our approach on two public datasets comprising 1100 dermoscopic images and compared it to other conventional and state-of-the-art unsupervised (i.e., no training required) lesion segmentation methods, as well as the state-of-the-art unsupervised saliency detection methods. Our results show that our approach is more accurate and robust in segmenting lesions compared to other methods. We also discuss the general extension of our framework as a saliency optimization algorithm for lesion segmentation.

We derived an automated algorithm for accurately measuring the thalamic diameter from 2-D fetal ultrasound (US) brain images. The algorithm overcomes the inherent limitations of the US image modality: nonuniform density; missing boundaries; and strong speckle noise. We introduced a "guitar" structure that represents the negative space surrounding the thalamic regions. The guitar acts as a landmark for deriving the widest points of the thalamus even when its boundaries are not identifiable. We augmented a generalized level-set framework with a shape prior and constraints derived from statistical shape models of the guitars; this framework was used to segment US images and measure the thalamic diameter. Our segmentation method achieved a higher mean Dice similarity coefficient, Hausdorff distance, specificity, and reduced contour leakage when compared to other well-established methods. The automatic thalamic diameter measurement had an interobserver variability of -0.56 ± 2.29 mm compared to manual measurement by an expert sonographer. Our method was capable of automatically estimating the thalamic diameter, with the measurement accuracy on par with clinical assessment. Our method can be used as part of computer-assisted screening tools that automatically measure the biometrics of the fetal thalamus; these biometrics are linked to neurodevelopmental outcomes.

Recent popularity of consumer-grade virtual reality devices, such as the Oculus Rift and the HTC Vive, has enabled household users to experience highly immersive virtual environments. We take advantage of the commercial availability of these devices to provide an immersive and novel virtual reality training approach, designed to teach individuals how to survive earthquakes, in common indoor environments. Our approach makes use of virtual environments realistically populated with furniture objects for training. During a training, a virtual earthquake is simulated. The user navigates in, and manipulates with, the virtual environments to avoid getting hurt, while learning the observation and self-protection skills to survive an earthquake. We demonstrated our approach for common scene types such as offices, living rooms and dining rooms. To test the effectiveness of our approach, we conducted an evaluation by asking users to train in several rooms of a given scene type and then test in a new room of the same type. Evaluation results show that our virtual reality training approach is effective, with the participants who are trained by our approach performing better, on average, than those trained by alternative approaches in terms of the capabilities to avoid physical damage and to detect potentially dangerous objects.

Unsupervised segmentation and contour detection remains a challenging task. In graph-based unsupervised segmentation, the formulation of the affinity graph is pivotal to segmentation performance. Conventional graph-based approaches often only define pixels as graph nodes, and may overlook important regional information. In this paper, we propose a novel scheme for affinity graph construction, where the affinity weight matrix unifies the association across pixel-wise nodes and multilevel region-wise nodes of different scales. Integrating the multilevel regional information, which is formulated using superpixels, into the affinity graph contributes to better capture of image intensity and color cues. Experimental evaluation of our approach on the BSDS500 dataset showed that our proposed method achieved the second best performance compared to other nine unsupervised state-of-art methods commonly used for comparison.

In saliency object detection, inappropriate boundary-background priors is known to degrade performance in challenging image datasets, and even may lead to `inverse' results when saliency regions are attached to the image boundaries. This is an active field where many works have proposed various techniques to lessen such degradation by inappropriate boundary-background priors. Although the use of boundary-background priors has shown to be capable of improving the detection, inherently, these techniques confront serious challenges in background suppression. To overcome this limitation, we propose an adaptive background extractor to search background seeds without the need of boundary-background priors. With the adaptive background seeds, the saliency objects can be then extracted via our proposed hierarchical foreground estimation model. We evaluate our adaptive Background Search and Foreground Estimation (BSFE) algorithm in comparison with six state-of-the-art methods on four well-recognized public datasets. The experimental results demonstrate that our BSFE algorithm outperforms compared methods in majority of the datasets and in particular achieves double-winners in terms of F-measure and mean absolute error on two challenging datasets.

BACKGROUND: With the developments of DNA sequencing technology, large amounts of sequencing data have become available in recent years and provide unprecedented opportunities for advanced association studies between somatic point mutations and cancer types/subtypes, which may contribute to more accurate somatic point mutation based cancer classification (SMCC). However in existing SMCC methods, issues like high data sparsity, small volume of sample size, and the application of simple linear classifiers, are major obstacles in improving the classification performance. RESULTS: To address the obstacles in existing SMCC studies, we propose DeepGene, an advanced deep neural network (DNN) based classifier, that consists of three steps: firstly, the clustered gene filtering (CGF) concentrates the gene data by mutation occurrence frequency, filtering out the majority of irrelevant genes; secondly, the indexed sparsity reduction (ISR) converts the gene data into indexes of its non-zero elements, thereby significantly suppressing the impact of data sparsity; finally, the data after CGF and ISR is fed into a DNN classifier, which extracts high-level features for accurate classification. Experimental results on our curated TCGA-DeepGene dataset, which is a reformulated subset of the TCGA dataset containing 12 selected types of cancer, show that CGF, ISR and DNN all contribute in improving the overall classification performance. We further compare DeepGene with three widely adopted classifiers and demonstrate that DeepGene has at least 24% performance improvement in terms of testing accuracy. CONCLUSIONS: Based on deep learning and somatic point mutation data, we devise DeepGene, an advanced cancer type classifier, which addresses the obstacles in existing SMCC studies. Experiments indicate that DeepGene outperforms three widely adopted existing classifiers, which is mainly attributed to its deep learning module that is able to extract the high level features between combinatorial somatic point mutations and cancer types.

In the field of saliency detection, many graph-based algorithms heavily depend on the accuracy of the pre-processed superpixel segmentation, which leads to significant sacrifice of detail information from the input image. In this paper, we propose a novel bottom-up saliency detection approach that takes advantage of both region-based features and image details. To provide more accurate saliency estimations, we first optimize the image boundary selection by the proposed erroneous boundary removal. By taking the image details and region-based estimations into account, we then propose the regularized random walks ranking to formulate pixel-wised saliency maps from the superpixel-based background and foreground saliency estimations. Experiment results on two public datasets indicate the significantly improved accuracy and robustness of the proposed algorithm in comparison with 12 state-of-the-art saliency detection approaches.

Accurate liver segmentation from computed tomography (CT) images is problematic due to non-uniform density, weak boundaries and because there may be multiple liver tumors that have heterogeneous intensities in region(s) of interest (ROIs). So we propose a generalized energy framework that harnesses the statistical intensity approximation with image data on graphs. Our statistical energy term takes advantage of the mixture-of-mixtures Gaussian model to approximate the probability density distribution of the liver and background to better differentiate between the two. The probability density estimation can be combined with the spatial cohesion of the graph-based Dirichlet integral by using graph calculus. Matrix decomposition and differentiation are used to minimize our proposed energy functional. We tested our approach on 20 public high-contrast CT images with single and multiple liver tumors. Our method had an average dice similarity coefficient (DSC) of 93.75±1.29%, an average false positive (FP) rate of 9.43±3.52% and an average false negative (FN) rate of 3.48±1.48%. Our method outperformed the benchmark graph-based Random Walker algorithm (average DSC=81.97±4.09%, average FP rate 34.10±10.53%, and average FN rate 7.10±4.35%).

Accurate estimation of noise level in images plays an important role in different image processing applications. The current algorithms can precisely estimate noise with smooth images, but it is still the challenge to approximate noise level from richly textured images. In this paper, we proposed a new adaptive superpixel classification algorithm for noise estimation in complicated textured images. Firstly, our new superpixel algorithm adapts the finite Gaussian clustering approach, which can better approximate homogeneous patches in noisy images. Then noise information is obtained locally from each superpixel patch. Finally, the best estimation of noise level is calculated with a statistical approach. Experimental results with various kinds of images demonstrate that our method is more accurate and robust compared to the five existing common used algorithms.

Intensity inhomogeneities and different types/levels of image noise are the two major obstacles to accurate image segmentation by region-based level set models. To provide a more general solution to these challenges, we propose a novel segmentation model that considers global and local image statistics to eliminate the influence of image noise and to compensate for intensity inhomogeneities. In our model, the global energy derived from a Gaussian model estimates the intensity distribution of the target object and background; the local energy derived from the mutual influences of neighboring pixels can eliminate the impact of image noise and intensity inhomogeneities. The robustness of our method is validated on segmenting synthetic images with/without intensity inhomogeneities, and with different types/levels of noise, including Gaussian noise, speckle noise, and salt and pepper noise, as well as images from different medical imaging modalities. Quantitative experimental comparisons demonstrate that our method is more robust and more accurate in segmenting the images with intensity inhomogeneities than the local binary fitting technique and its more recent systematic model. Our technique also outperformed the region-based Chan–Vese model when dealing with images without intensity inhomogeneities and produce better segmentation results than the graph-based algorithms including graph-cuts and random walker when segmenting noisy images.

Segmentation of the target object(s) from images that have multiple complicated regions, mixture intensity distributions or are corrupted by noise poses a challenge for the level set models. In addition, the conventional piecewise smooth level set models normally require prior knowledge about the number of image segments. To address these problems, we propose a novel segmentation energy function with two distribution descriptors to model the background and the target. The single background descriptor models the heterogeneous background with multiple regions. Then, the target descriptor takes into account the intensity distribution and incorporates local spatial constraint. Our descriptors, which have more complete distribution information, construct the unique energy function to differentiate the target from the background and are more tolerant of image noise. We compare our approach to three other level set models: 1) the Chan-Vese; 2) the multiphase level set; and 3) the geodesic level set. This comparison using 260 synthetic images with varying levels and types of image noise and medical images with more complicated backgrounds showed that our method outperforms these models for accuracy and immunity to noise. On an additional set of 300 synthetic images, our model is also less sensitive to the contour initialization as well as to different types and levels of noise.

Equation (16) in the above paper (ibid., vol. 22, no. 8, pp. 3296-3309, Aug. 2013) contained an error in the numerator. Equation (17) in the same paper contained errors in both the numerator and the denominator. The corrected versions of both equations are presented here.

We propose a novel joint probabilistic model that correlates a new probabilistic shape model with the corresponding global intensity distribution to segment multiple abdominal organs simultaneously. Our probabilistic shape model estimates the probability of an individual voxel belonging to the estimated shape of the object. The probability density of the estimated shape is derived from a combination of the shape variations of target class and the observed shape information. To better capture the shape variations, we used probabilistic principle component analysis optimized by expectation maximization to capture the shape variations and reduce computational complexity. The maximum a posteriori estimation was optimized by the iterated conditional mode-expectation maximization. We used 72 training datasets including low- and high-contrast CT images to construct the shape models for the liver, spleen and both kidneys. We evaluated our algorithm on 40 test datasets that were grouped into normal (34 normal cases) and pathologic (6 datasets) classes. The testing datasets were from different databases and manual segmentation was performed by different clinicians. We measured the volumetric overlap percentage error, relative volume difference, average square symmetric surface distance, false positive rate and false negative rate and our method achieved accurate and robust segmentation for multiple abdominal organs simultaneously.

Automated liver segmentation is problematic due to variations in liver shape / size and because the liver has a similar density distribution to surrounding structures. We propose a method that: 1) utilizes iteratively constructed probabilistic liver and rib cage atlases, 2) conducts the Gaussian distribution analysis to avoid incorrectly classifying the irrelevant surrounding tissues as `liver region' in the conventional probabilistic atlas based method, and maps the intensity range of the input candidate liver region onto the liver atlas, 3) retrieves the `missing parts' of the liver by deformable registration. Our approach is automated and able to segment the liver from high-contrast and low-contrast CT volumes. Forty clinical CT studies were used for atlas construction and validation. Our method outperformed two other probabilistic atlas-based liver segmentation methods.