scholarly journals A reformulation of pLSA for uncertainty estimation and hypothesis testing in bio-imaging

2020 ◽  
Vol 36 (13) ◽  
pp. 4080-4087
Author(s):  
P D Tar ◽  
N A Thacker ◽  
S Deepaisarn ◽  
J P B O’Connor ◽  
A W McMahon
Keyword(s):  
Hypothesis Testing ◽  
Degrees Of Freedom ◽  
Semantic Analysis ◽  
High Sensitivity ◽  
Supplementary Information ◽  
Probabilistic Latent Semantic Analysis ◽  
Source Image ◽  
Model Parameters ◽  
Bio Imaging ◽  
Magnetic Resonance Imaging Mri

Abstract Motivation Probabilistic latent semantic analysis (pLSA) is commonly applied to describe mass spectra (MS) images. However, the method does not provide certain outputs necessary for the quantitative scientific interpretation of data. In particular, it lacks assessment of statistical uncertainty and the ability to perform hypothesis testing. We show how linear Poisson modelling advances pLSA, giving covariances on model parameters and supporting χ2 testing for the presence/absence of MS signal components. As an example, this is useful for the identification of pathology in MALDI biological samples. We also show potential wider applicability, beyond MS, using magnetic resonance imaging (MRI) data from colorectal xenograft models. Results Simulations and MALDI spectra of a stroke-damaged rat brain show MS signals from pathological tissue can be quantified. MRI diffusion data of control and radiotherapy-treated tumours further show high sensitivity hypothesis testing for treatment effects. Successful χ2 and degrees-of-freedom are computed, allowing null-hypothesis thresholding at high levels of confidence. Availability and implementation Open-source image analysis software available from TINA Vision, www.tina-vision.net. Supplementary information Supplementary data are available at Bioinformatics online.

scholarly journals Renormalization Analysis of Topic Models

Entropy ◽  
2020 ◽  
Vol 22 (5) ◽  
pp. 556
Author(s):  
Sergei Koltcov ◽  
Vera Ignatenko
Keyword(s):  
Latent Dirichlet Allocation ◽  
Semantic Analysis ◽  
Optimal Number ◽  
Probabilistic Latent Semantic Analysis ◽  
Model Parameters ◽  
Grid Search ◽  
Renormalization Procedure ◽  
Allocation Model ◽  
Latent Dirichlet Allocation Model ◽  
Dirichlet Allocation

In practice, to build a machine learning model of big data, one needs to tune model parameters. The process of parameter tuning involves extremely time-consuming and computationally expensive grid search. However, the theory of statistical physics provides techniques allowing us to optimize this process. The paper shows that a function of the output of topic modeling demonstrates self-similar behavior under variation of the number of clusters. Such behavior allows using a renormalization technique. A combination of renormalization procedure with the Renyi entropy approach allows for quick searching of the optimal number of topics. In this paper, the renormalization procedure is developed for the probabilistic Latent Semantic Analysis (pLSA), and the Latent Dirichlet Allocation model with variational Expectation–Maximization algorithm (VLDA) and the Latent Dirichlet Allocation model with granulated Gibbs sampling procedure (GLDA). The experiments were conducted on two test datasets with a known number of topics in two different languages and on one unlabeled test dataset with an unknown number of topics. The paper shows that the renormalization procedure allows for finding an approximation of the optimal number of topics at least 30 times faster than the grid search without significant loss of quality.

scholarly journals Estimating Topic Modeling Performance with Sharma–Mittal Entropy

Entropy ◽  
2019 ◽  
Vol 21 (7) ◽  
pp. 660 ◽  
Author(s):  
Sergei Koltcov ◽  
Vera Ignatenko ◽  
Olessia Koltsova
Keyword(s):  
Latent Semantic Analysis ◽  
Topic Modeling ◽  
Statistical Physics ◽  
Latent Dirichlet Allocation ◽  
Semantic Analysis ◽  
Probabilistic Latent Semantic Analysis ◽  
Model Parameters ◽  
Theoretical Ground ◽  
Text Documents ◽  
Optimizing Model

Topic modeling is a popular approach for clustering text documents. However, current tools have a number of unsolved problems such as instability and a lack of criteria for selecting the values of model parameters. In this work, we propose a method to solve partially the problems of optimizing model parameters, simultaneously accounting for semantic stability. Our method is inspired by the concepts from statistical physics and is based on Sharma–Mittal entropy. We test our approach on two models: probabilistic Latent Semantic Analysis (pLSA) and Latent Dirichlet Allocation (LDA) with Gibbs sampling, and on two datasets in different languages. We compare our approach against a number of standard metrics, each of which is able to account for just one of the parameters of our interest. We demonstrate that Sharma–Mittal entropy is a convenient tool for selecting both the number of topics and the values of hyper-parameters, simultaneously controlling for semantic stability, which none of the existing metrics can do. Furthermore, we show that concepts from statistical physics can be used to contribute to theory construction for machine learning, a rapidly-developing sphere that currently lacks a consistent theoretical ground.

scholarly journals Model forB1Imaging in MRI Using the Rotating RF Field

2014 ◽  
Vol 2014 ◽  
pp. 1-11 ◽  
Author(s):  
Adnan Trakic ◽  
Jin Jin ◽  
Ewald Weber ◽  
Stuart Crozier
Keyword(s):  
Degrees Of Freedom ◽  
Acoustic Noise ◽  
Bloch Equation ◽  
Gradient Coil ◽  
Imaging Method ◽  
Nonuniform Field ◽  
Rf Coil ◽  
Image Encoding ◽  
Magnetic Resonance Imaging Mri ◽  
Space Requirements

Conventionally, magnetic resonance imaging (MRI) is performed by pulsing gradient coils, which invariably leads to strong acoustic noise, patient safety concerns due to induced currents, and costly power/space requirements. This modeling study investigates a new silent, gradient coil-free MR imaging method, in which a radiofrequency (RF) coil and its nonuniform field (B1+) are mechanically rotated about the patient. The advantage of the rotatingB1+field is that, for the first time, it provides a large number of degrees of freedom to aid a successfulB1+image encoding process. The mathematical modeling was performed using flip angle modulation as part of a finite-difference-based Bloch equation solver. Preliminary results suggest that representative MR images with intensity deviations of <5% from the original image can be obtained using rotating RF field approach. This method may open up new avenues towards anatomical and functional imaging in medicine.

scholarly journals Neuroimaging in normal pressure hydrocephalus

2015 ◽  
Vol 9 (4) ◽  
pp. 350-355 ◽  
Author(s):  
Benito Pereira Damasceno
Keyword(s):  
Normal Pressure Hydrocephalus ◽  
Cerebral Atrophy ◽  
High Sensitivity ◽  
Normal Pressure ◽  
Evans Index ◽  
Flow Void ◽  
High Positive Predictive Value ◽  
Subarachnoid Spaces ◽  
Magnetic Resonance Imaging Mri ◽  
Improved Accuracy

ABSTRACT Normal pressure hydrocephalus (NPH) is a syndrome characterized by the triad of gait disturbance, mental deterioration and urinary incontinence, associated with ventriculomegaly and normal cerebrospinal fluid (CSF) pressure. The clinical presentation (triad) may be atypical or incomplete, or mimicked by other diseases, hence the need for supplementary tests, particularly to predict postsurgical outcome, such as CSF tap-tests and computed tomography (CT) or magnetic resonance imaging (MRI). The CSF tap-test, especially the 3 to 5 days continuous external lumbar drainage of at least 150 ml/day, is the only procedure that simulates the effect of definitive shunt surgery, with high sensitivity (50-100%) and high positive predictive value (80-100%). According to international guidelines, the following are CT or MRI signs decisive for NPH diagnosis and selection of shunt-responsive patients: ventricular enlargement disproportionate to cerebral atrophy (Evans index >0.3), and associated ballooning of frontal horns; periventricular hyperintensities; corpus callosum thinning and elevation, with callosal angle between 40º and 90º; widening of temporal horns not fully explained by hippocampal atrophy; and aqueductal or fourth ventricular flow void; enlarged Sylvian fissures and basal cistern, and narrowing of sulci and subarachnoid spaces over the high convexity and midline surface of the brain. On the other hand, other imaging methods such as radionuclide cisternography, SPECT, PET, and also DTI or resting-state functional MRI, although suitable for NPH diagnosis, do not yet provide improved accuracy for identifying shunt-responsive cases.

ZnO coated CoFe2O4 nanoparticles for multimodal bio-imaging

RSC Advances ◽  
2016 ◽  
Vol 6 (23) ◽  
pp. 18843-18851 ◽  
Author(s):  
N. Venkatesha ◽  
Yasrib Qurishi ◽  
Hanudatta S. Atreya ◽  
Chandan Srivastava
Keyword(s):  
Magnetic Resonance Imaging ◽  
Magnetic Resonance ◽  
Core Shell ◽  
Resonance Imaging ◽  
Cofe2o4 Nanoparticles ◽  
Shell Nanoparticles ◽  
Fluorescence Optical Imaging ◽  
Bio Imaging ◽  
Magnetic Resonance Imaging Mri ◽  
Core Shell Nanoparticles

The potential of CoFe2O4–ZnO core–shell nanoparticles for fluorescence optical imaging and as a contrast agent for magnetic resonance imaging (MRI) is demonstrated.

scholarly journals Visualization of Lesions of Metaphyses and Epiphyses of Bones in Newborns and Young Children

2021 ◽  
pp. 82-90
Author(s):  
N. A. Sholokhova
Keyword(s):  
Young Children ◽  
False Negative ◽  
Age Groups ◽  
High Sensitivity ◽  
Negative Results ◽  
False Negative Results ◽  
Inflammatory Processes ◽  
Group 5 ◽  
Magnetic Resonance Imaging Mri ◽  
Diagnostic Capabilities

The aim of this study was to determine the diagnostic capabilities of various methods of radiological diagnostics for lesions of the metaphyses and epiphyses of bones in newborns and young children.The study involved 108 children in the age group 5 days – 12 months with pathological changes in the pineal gland and bone metaphysis. The possibilities and advantages of standard radiography (СR), ultrasound examination (US) and magnetic resonance imaging (MRI) in the early and differential diagnosis of the osteomyelitis process and epiphyseolysis have been determined. High sensitivity (98 %), specificity (99 %) and accuracy (98 %) for ultrasound and sensitivity (94 %), specificity (89 %) and accuracy (95 %) of MRI in diagnosing osteomyelitis in patients of this age groups. At the same time, the possibilities of standard radiography at the stages of early diagnosis of inflammatory processes in the distal parts of the bones were limited due to a number of factors. The use of diagnostic algorithms greatly facilitates the work of a radiologist and reduces the number of false negative results during the initial treatment of patients.

Sign in / Sign up

Export Citation Format

Share Document