No abstract

No abstract

No abstract

Quantitative Magnetic Resonance Imaging (qMRI) provides researchers insight into pathological and physiological alterations of living tissue, with the help of which, researchers hope to predict (local) therapeutic efficacy early and determine optimal treatment schedule. However, the analysis of qMRI has been limited to ad-hoc heuristic methods. Our research provides a powerful statistical framework for image analysis and sheds light on future localized adaptive treatment regimes tailored to the individual's response. We assume in an imperfect world we only observe a blurred and noisy version of the underlying pathological/physiological changes via qMRI, due to measurement errors or unpredictable influences. We use a hidden Markov Random Field to model the spatial dependence in the data and develop a maximum likelihood approach via the Expectation-Maximization algorithm with stochastic variation. An important improvement over previous work is the assessment of variability in parameter estimation, which is the valid basis for statistical inference. More importantly, we focus on the expected changes rather than image segmentation. Our research has shown that the approach is powerful in both simulation studies and on a real dataset, while quite robust in the presence of some model assumption violations.

No abstract

Skin color detection is an important subject in computer vision research. Color segmentation takes a great attention because color is an effective and robust visual cue for characterizing an object from the others. To aim at existing skin color algorithms considering the luminance information not enough, a reliable color modeling approach was proposed. It is based on the fact that color distribution of a single-colored object is not invariant with respect to luminance variations even in the Cb-Cr plane and does not ignore the influence on luminance Y component in YCbCr color space. Firstly, according to statistics of skin color pixels, we take the luminance Y by ascending order, divide the total range of Y into finite number of intervals, collect pixels whose luminance belongs to the same luminance interval, calculate the covariance and the mean value of Cb and Cr with respect to Y, and use the above data to train the BP neural network, then we get the self-adaptive skin color model and design a Gaussian model classifier. The experimental results have indicated that this algorithm can effectively fulfill the skin-color detection for images captured under different environmental condition and the performance of the skin color segmentation is significantly improved.

Extended exposure to sunlight may give rise to chemical and physical damages of human hairs. In this work, we report a novel method for non-destructive quantification of hair photodamage via multispectral imaging (MSI) technology. We show that the multispectral reflectance value in near-infrared region has a strong correlation with hair photodamage. More specifically, the hair segments with longer growing time and the same hair root segment after continuous ultraviolet (UV) irradiation displaying more severe photodamage observed via scanning electron microscopy (SEM) micrographs showed significantly higher multispectral reflectance value. Besides, the multispectral reflectance value of hair segments with different growing time was precisely reproduced by exposing the same hair root segment to specific durations of UV irradiation, suggesting that MSI can be adequately applied to determine the sunlight exposure time of the hair. The loss of cystine content of photodamaged hairs was identified to be the main factor that physiologically contributed to the morphological changes of hair surface fibers and hence the variation of their multispectral reflectance spectra. Considering the environmental information recording nature of hairs, we believe that MSI for non-destructive evaluation of hair photodamage would prove valuable for assessing sunlight exposure time of a subject in the biomedical fields.

Summary This work is motivated by a quantitative Magnetic Resonance Imaging study of the relative change in tumor vascular permeability during the course of radiation therapy. The differences in tumor and healthy brain tissue physiology and pathology constitute a notable feature of the image data—spatial heterogeneity with respect to its contrast uptake profile (a surrogate for permeability) and radiation induced changes in this profile. To account for these spatial aspects of the data, we employ a Gaussian hidden Markov random field (MRF) model. The model incorporates a latent set of discrete labels from the MRF governed by a spatial regularization parameter. We estimate the MRF regularization parameter and treat the number of MRF states as a random variable and estimate it via a reversible jump Markov chain Monte Carlo algorithm. We conduct simulation studies to examine the performance of the model and compare it with a recently proposed method using the Expectation-Maximization (EM) algorithm. Simulation results show that the Bayesian algorithm performs as well, if not slightly better than the EM based algorithm. Results on real data suggest that the tumor “core” vascular permeability increases relative to healthy tissue three weeks after starting radiotherapy, which may be an opportune time to initiate chemotherapy and warrants further investigation.