Audrey G. Chung

Salient Region Detection Using Self-Guided Statistical Non-Redundancy in Natural Images

We propose an effective framework for salient region detection in natural images based on the concept of self-guided statistical non-redundancy (SGNR). Salient regions are unique, because they have low information redundancy within a given image, while the rest of the scene may highly be redundant. We first analyze the structural characteristics of the image using structured image elements (samples) and classify them as being non-redundant or redundant based on textural compactness and overall non-redundancy. This guides saliency detection toward regions with low information redundancy by considering explicitly high information redundancy of samples potentially belonging to the background. We then compute the saliency map by determining the statistical non-redundancy of each sample using a conditional graph model. Experimental results based on publicly available data sets show that SGNR provides promising results when compared with existing saliency approaches.

Statistical Textural Distinctiveness in Multi-Parametric Prostate MRI for Suspicious Region Detection

Prostate cancer is the most diagnosed form of cancer, but survival rates are relatively high with sufficiently early diagnosis. Current computer-aided image-based cancer detection methods face notable challenges including noise in MRI images, variability between different MRI modalities, weak contrast, and non-homogeneous texture patterns, making it difficult for diagnosticians to identify tumour candidates. We propose a novel saliency-based method for identifying suspicious regions in multi-parametric MR prostate images based on statistical texture distinctiveness. In this approach, a sparse texture model is learned via expectation maximization from features derived from multi-parametric MR prostate images, and the statistical texture distinctiveness-based saliency based on this model is used to identify suspicious regions. The proposed method was evaluated using real clinical prostate MRI data, and results demonstrate a clear improvement in suspicious region detection relative to the state-of-art method.

High-resolution motion-compensated imaging photoplethysmography for remote heart rate monitoring

We present a novel non-contact photoplethysmographic (PPG) imaging system based on high-resolution video recordings of ambient reflectance of human bodies that compensates for body motion and takes advantage of skin erythema fluctuations to improve measurement reliability for the purpose of remote heart rate monitoring. A single measurement location for recording the ambient reflectance is automatically identified on an individual, and the motion for the location is determined over time via measurement location tracking. Based on the determined motion information motion-compensated reflectance measurements at different wavelengths for the measurement location can be acquired, thus providing more reliable measurements for the same location on the human over time. The reflectance measurement is used to determine skin erythema fluctuations over time, resulting in the capture of a PPG signal with a high signal-to-noise ratio. To test the efficacy of the proposed system, a set of experiments involving human motion in a front-facing position were performed under natural ambient light. The experimental results demonstrated that skin erythema fluctuations can achieve noticeably improved average accuracy in heart rate measurement when compared to previously proposed non-contact PPG imaging systems.

Discovery Radiomics for Pathologically-Proven Computed Tomography Lung Cancer Prediction

Lung cancer is the leading cause for cancer related deaths. As such, there is an urgent need for a streamlined process that can allow radiologists to provide diagnosis with greater efficiency and accuracy. A powerful tool to do this is radiomics: a high-dimension imaging feature set. In this study, we take the idea of radiomics one step further by introducing the concept of discovery radiomics for lung cancer prediction using CT imaging data. In this study, we realize these custom radiomic sequencers as deep convolutional sequencers using a deep convolutional neural network learning architecture. To illustrate the prognostic power and effectiveness of the radiomic sequences produced by the discovered sequencer, we perform cancer prediction between malignant and benign lesions from 97 patients using the pathologically-proven diagnostic data from the LIDC-IDRI dataset. Using the clinically provided pathologically-proven data as ground truth, the proposed framework provided an average accuracy of 77.52% via 10-fold cross-validation with a sensitivity of 79.06% and specificity of 76.11%, surpassing the state-of-the art method.

Improved fine structure modeling via guided stochastic clique formation in fully connected conditional random fields

Markov random fields (MRFs) and conditional random fields (CRFs) are influential tools in image modeling, particularly for applications such as image segmentation. Local MRFs and CRFs utilize local nodal interactions when modeling, leading to excessive smoothness on boundaries (i.e., the short-boundary bias problem). Recently, the concept of fully connected conditional random fields with stochastic cliques (SFCRF) was proposed to enable long-range nodal interactions while addressing the computational complexity associated with fully connected random fields. While SFCRF was shown to provide significant improvements in segmentation accuracy, there were still limitations with the preservation of fine structure boundaries. To address these limitations, we propose a new approach to stochastic clique formation for fully connected random fields (G-SFCRF) that is guided by the structural characteristics of different nodes within the random field. In particular, fine structures surrounding a node are modeled statistically by probability distributions, and stochastic cliques are formed by considering the statistical similarities between nodes within the random fields. Experimental results show that G-SFCRF outperforms existing fully connected CRF frameworks, SFCRF, and the principled deep random field framework for image segmentation.

Prostate Cancer Detection via a Quantitative Radiomics-Driven Conditional Random Field Framework

The use of high-volume quantitative radiomics features extracted from multi-parametric magnetic resonance imaging (MP-MRI) is gaining attraction for the autodetection of prostate tumors, since it provides a plethora of mineable data, which can be used for both detection and prognosis of prostate cancer. While current voxel-resolution radiomics-driven prostate tumor detection approaches utilize quantitative radiomics features associated with individual voxels on an independent basis, the incorporation of additional information regarding the spatial and radiomics feature relationships between voxels has significant potential for achieving a more reliable detection performance. Motivated by this, we present a novel approach for automatic prostate cancer detection using a radiomics-driven conditional random field (RD-CRF) framework. In addition to the high-throughput extraction and utilization of a comprehensive set of voxel-level quantitative radiomics features, the proposed RD-CRF framework leverages inter-voxel spatial and radiomics feature relationships to ensure that the autodetected tumor candidates exhibit interconnected tissue characteristics reflective of cancerous tumors. We evaluated the performance of the proposed framework using clinical prostate MP-MRI data of 20 patients, and the results of RD-CRF framework demonstrated a clear improvement with respect to the state-of-the-art in quantitative radiomics for automatic voxel-resolution prostate cancer detection.

TIGGER: A Texture-Illumination Guided Global Energy Response Model for Illumination Robust Object Saliency

Global saliency is an important aspect of many computerand robotic vision tasks, and with the increased interest infields such as autonomous navigation, a significant area ofresearch. A challenging aspect of modelling global saliencyin practical applications is the presence of varying or non-uniform illumination conditions. Many current models fail toaccurately detect salient regions in non-uniform illuminationconditions and often produce different saliency maps for thesame image under changing illumination. In this paper, wepropose a novel model for illumination robust global saliency. For a given input image, texture-illumination guided energyresponses (TIGERs) are computed at different scales usinga novel multi-scale extension of TIGER. To acquire theseresponses, image intensity is modelled as the summationof the low frequency illumination component and the highfrequency texture component. A captured image is disassociatedinto these components via Bayesian minimization, with the required posterior probability estimated through animportance-weighted Monte Carlo sampling approach. Thetexture-illumination guided global energy response (TIGGER) is computed as the aggregate sum of TIGERs acrossall scales. The global saliency map is obtained via a k-meansclustering-based region adjacency graph (RAG) model. Experimentalresults produce global saliency maps with improvedperformance in non-uniform lighting conditions andgreater consistency when compared to other state-of-the-artmethods.

Discovery radiomics via evolutionary deep radiomic sequencer discovery for pathologically proven lung cancer detection.

Abstract. While lung cancer is the second most diagnosed form of cancer in men and women, a sufficiently early diagnosis can be pivotal in patient survival rates. Imaging-based, or radiomics-driven, detection methods have been developed to aid diagnosticians, but largely rely on hand-crafted features that may not fully encapsulate the differences between cancerous and healthy tissue. Recently, the concept of discovery radiomics was introduced, where custom abstract features are discovered from readily available imaging data. We propose an evolutionary deep radiomic sequencer discovery approach based on evolutionary deep intelligence. Motivated by patient privacy concerns and the idea of operational artificial intelligence, the evolutionary deep radiomic sequencer discovery approach organically evolves increasingly more efficient deep radiomic sequencers that produce significantly more compact yet similarly descriptive radiomic sequences over multiple generations. As a result, this framework improves operational efficiency and enables diagnosis to be run locally at the radiologist’s computer while maintaining detection accuracy. We evaluated the evolved deep radiomic sequencer (EDRS) discovered via the proposed evolutionary deep radiomic sequencer discovery framework against state-of-the-art radiomics-driven and discovery radiomics methods using clinical lung CT data with pathologically proven diagnostic data from the LIDC-IDRI dataset. The EDRS shows improved sensitivity (93.42%), specificity (82.39%), and diagnostic accuracy (88.78%) relative to previous radiomics approaches.

Time-Frequency Domain Analysis via Pulselets for Non-contact Heart Rate Estimation from Remotely Acquired Photoplethysmograms

A novel method for remote heart rate estimation via analysis in the time-frequency domain is proposed. A photoplethysmogram (PPG) waveform is constructed via a Bayesian minimization approach with the required posterior probability obtained through an importance-weighted Monte Carlo sampling method. A pulselet (wavelet chosen for its similarities with a finger pulse oximiter PPG waveform), is used in the continuous wavelet transform to produce a map of the wavelet energy response in the time-frequency domain. This allows the heart rate frequency to be estimated at each time step, accounting for naturally occurring changes in heart rate over time which may cause error with frequency domain based methods. The frequency corresponding to the highest wavelet response at each time step is averaged across the entire time series to estimate the average heart rate. Experimental results against a data set of 30 videos show an improvement over current state-of-the-art methods.

High dynamic range map estimation via fully connected random fields with stochastic cliques

The reconstruction of high dynamic range (HDR) images via conventional camera systems and low dynamic range (LDR) images is a growing field of research in image acquisition. The radiance map associated with the HDR image of a scene is typically computed using multiple images of the same scene captured at different exposures (i.e., bracketed LDR imzages). This approach, though inexpensive, is sensitive to noise under high camera ISO. Each bracketed image is associated with a different level of noise due to the change in exposure time, and the noise is further amplified when tone-mapping the HDR image for display. A new framework is proposed to address the associated noise in the context of random fields. The estimation of the HDR image from a set of LDR images is formulated as a stochastically fully connected conditional random field where the spatial information is incorporated to compute the HDR value in combination with the LDR image values. Experimental results show that the proposed framework compensated the non-stationary ISO noise while preserving the boundaries in the estimated HDR images.