Can a 3D task transfer function accurately represent the signal transfer properties of low-contrast lesions in non-linear CT systems?

2018 ◽  
Author(s):  
Marthony Robins ◽  
Justin B. Solomon ◽  
Ehsan Samei
Keyword(s):  
Transfer Function ◽  
Signal Transfer ◽  
Low Contrast ◽  
Task Transfer ◽  
Non Linear ◽  
Transfer Properties

Correction of the contrast transfer function to determine the radial density distribution of frozen-hydrated tobacco mosaic virus

1992 ◽  
Vol 50 (1) ◽  
pp. 536-537
Author(s):  
Michael F. Smith ◽  
John P. Langmore
Keyword(s):  
Transfer Function ◽  
Phase Contrast ◽  
Heavy Atom ◽  
Biological Molecules ◽  
Frequency Compensation ◽  
Contrast Transfer Function ◽  
High Background ◽  
Low Contrast ◽  
Radial Density Distribution ◽  
Excellent Preservation

The purpose of image reconstruction is to determine the mass densities within molecules by analysis of the intensities within images. Cryo-EM offers this possibility by virtue of the excellent preservation of internal structure without heavy atom staining. Cryo-EM images, however, have low contrast because of the similarity between the density of biological material and the density of vitreous ice. The images also contain a high background of inelastic scattering. To overcome the low signal and high background, cryo-images are typically recorded 1-3 μm underfocus to maximize phase contrast. Under those conditions the image intensities bear little resemblance to the object, due to the dependence of the contrast transfer function (CTF) upon spatial frequency. Compensation (i.e., correction) for the CTF is theoretically possible, but implementation has been rare. Despite numerous studies of molecules in ice, there has never been a quantitative evaluation of compensated images of biological molecules of known structure.

Fourier Series-Based Neural Networks for Vehicle System Identification

1999 ◽  
Author(s):  
Imtiaz Haque ◽  
Juergen Schuller
Keyword(s):  
Neural Networks ◽  
Fourier Series ◽  
System Identification ◽  
Transfer Function ◽  
Parameter Identification ◽  
Frequency Domain ◽  
Vehicle System ◽  
Non Linear ◽  
Function Identification ◽  
Non Linear System

Abstract The use of neural networks in system identification is an emerging field. Neural networks have become popular in recent years as a means to identify linear and non-linear systems whose characteristics are unknown. The success of sigmoidal networks in parameter identification has been limited. However, harmonic activation-based neural networks, recent arrivals in the field of neural networks, have shown excellent promise in linear and non-linear system parameter identification. They have been shown to have excellent generalization capability, computational parallelism, absence of local minima, and good convergence properties. They can be used in the time and frequency domain. This paper presents the application of a special class of such networks, namely Fourier Series neural networks (FSNN) to vehicle system identification. In this paper, the applications of the FSNNs are limited to the frequency domain. Two examples are presented. The results of the identification are based on simulation data. The first example demonstrates the transfer function identification of a two-degree-of freedom lateral dynamics model of an automobile. The second example involves transfer function identification for a quarter car model. The network set-up for such identification is described. The results of the network identification are compared with theory. The results indicate excellent prediction properties of such networks.

A Compartmental Model of Linear Resonance and Signal Transfer in Dendrites

2012 ◽  
Vol 24 (12) ◽  
pp. 3126-3144 ◽  
Author(s):  
Alan Schoen ◽  
Ali Salehiomran ◽  
Matthew E. Larkum ◽  
Erik P. Cook
Keyword(s):  
Continuous Time ◽  
Compartmental Model ◽  
Signal Transfer ◽  
Experimental Conditions ◽  
Impedance Model ◽  
Membrane Impedance ◽  
Dendritic Membrane ◽  
Linear Resonance ◽  
Impedance Functions ◽  
Transfer Properties

Dendrites carry signals between synapses and the soma and play a central role in neural computation. Although they contain many nonlinear ion channels, their signal-transfer properties are linear under some experimental conditions. In experiments with continuous-time inputs, a resonant linear two-port model has been shown to provide a near-perfect fit to the dendrite-to-soma input-output relationship. In this study, we focused on this linear aspect of signal transfer using impedance functions that replace biophysical channel models in order to describe the electrical properties of the dendritic membrane. The membrane impedance model of dendrites preserves the accuracy of the two-port model with minimal computational complexity. Using this approach, we demonstrate two membrane impedance profiles of dendrites that reproduced the experimentally observed two-port results. These impedance profiles demonstrate that the two-port results are compatible with different computational schemes. In addition, our model highlights how dendritic resonance can minimize the location-dependent attenuation of signals at the resonant frequency. Thus, in this model, dendrites function as linear-resonant filters that carry signals between nonlinear computational units.

Sign in / Sign up

Export Citation Format

Share Document