Search

In this work, doubly stochastic Poisson (Cox) processes and convolutional neural net (CNN) classifiers are used to estimate the number of instances of an object in an image. Poisson processes are well suited to model events that occur randomly in space, such as the location of objects in an image or the enumeration of objects in a scene. The proposed algorithm selects a subset of bounding boxes in the image domain, then queries them for the presence of the object of interest by running a pre-trained CNN classifier. The resulting observations are then aggregated, and a posterior distribution over the intensity of a Cox process is computed. This intensity function is summed up, providing an estimator of the number of instances of the object over the entire image. Despite the flexibility and versatility of Cox processes, their application to large datasets is limited as their computational complexity and storage requirements do not easily scale with image size, typically requiring O(n3) computation time and O(n2) storage, where n is the number of observations. To mitigate this problem, we employ the Kronecker algebra, which takes advantage of direct product structures. As the likelihood is non-Gaussian, the Laplace approximation is used for inference, employing the conjugate gradient and Newton's method. Our approach has then close to linear performance, requiring only O(n3/2) computation time and O(n) memory. Results are presented on simulated data and on images from the publicly available MS COCO dataset. We compare our counting results with the state-of-the-art detection method, Faster RCNN, and demonstrate superior performance. [ABSTRACT FROM AUTHOR]

Purpose: Disrupted-in-schizophrenia-1 (DISC1) is a promising genetic susceptibility factor for major psychiatric conditions, such as schizophrenia. We hypothesized that the mutant DISC1 alters the homeostasis of multi-receptor interactions between dopaminergic [dopamine 2/3 (DR)], glutamatergic [metabotropic glutamate 5 (mGluR)], cannabinoid 1 (CBR), and nicotinic acetylcholine (α4β2-nAChR) receptors in the brains of mice with inducible forebrain neuronal expression of dominant-negative mutant DISC1. Procedures: The quantitative in vitro autoradiography was performed with positron emission tomography (PET) ligands using [C]raclopride (DR), [C]ABP688 (mGluR), [C]OMAR (CBR), and [F]AZAN (nAChR). Total binding (pmol/cc) from standard and binding index, defined as [(region of interest − reference) / reference], was analyzed in the parasagittal sections. The cerebellum was used as a reference for DR, mGluR, and α4β2-nAChR, while the midbrain was the reference tissue for CBR, because of the high density of CBR in the cerebellum. Results: We observed a significant positive correlation between mGluR and DR in the nucleus accumbens (NAc) in mutant DISC1 (rho = 0.6, p = 0.04; y = 0.02 x + 6.7) and a trend of negative correlation between those receptors in the dorsal striatum (DS) in control animals (rho = −0.5, p = 0.09; y = −0.03 x + 23), suggesting a co-release of dopamine (DA) and glutamate (Glu) in the NAc, but not in the DS. There were trends of an inverse relationship between striatal CBR and DR (rho = −0.7, p = 0.07) as well as between dorsal thalamic nAChR and striatal DR (rho = −0.5, p = 0.08). There was no statistically significant difference of the individual receptor density in the majority of brain regions. Conclusions: The mutant DISC1 altered the homeostasis of multi-receptor interactions of coincident signaling of DA and Glu in the NAc, but not in the DS, and mutually negative control of striatal CBR and DR. Multi-receptor mapping with PET ligands in relevant animal models could be a valuable translational approach for psychiatric drug development. [ABSTRACT FROM AUTHOR]

Finding correspondences between images by template matching is a common problem in image understanding. Although a variety of solutions have been proposed, most of them rely on the arbitrary choice of a template and a matching function. Often, different cost functions lead to different results, and the choice of a good cost for a specific application remains an art. Statistical models on the other hand, allow us to derive optimal learning and matching algorithms from modeling assumptions using likelihood maximization principles. The key contribution of this paper is the development of a statistical framework for learning what function to optimize from training examples. We present a family of statistical models for grayscale images, which allow us to derive optimal template-matching algorithms. The intensity at each pixel is described by a random variable whose distribution is encoded by a deformable template. Firstly, we assume the intensity distribution to be Gaussian and derive an intensity-matching algorithm, which is a generalization of the classical sum-of-squared differences. Then, we introduce a hidden segmentation variable in the probabilistic model and derive a segmentation-matching algorithm that can handle photometric variations. Both models are exemplified on the automatic detection of anatomical landmarks in brain Magnetic Resonance Images. [ABSTRACT FROM AUTHOR]