scholarly journals A Linear Least Squares Fit mapping method for information retrieval from natural language texts

1992 ◽  
Author(s):  
Yiming Yang ◽  
Christopher G. Chute
Keyword(s):  
Information Retrieval ◽  
Natural Language ◽  
Least Squares ◽  
Mapping Method ◽  
Linear Least Squares ◽  
Least Squares Fit

Detectability of simulated masses and calcifications in mammography

1998 ◽  
Vol 39 (5) ◽  
pp. 501-506 ◽  
Author(s):  
J. B. Olsen ◽  
A. Skretting
Keyword(s):  
Mathematical Model ◽  
Least Squares ◽  
Breast Tissue ◽  
Detection Threshold ◽  
Roc Curves ◽  
Receiver Operating Characteristics ◽  
Linear Least Squares ◽  
Operating Characteristics ◽  
Least Squares Fit ◽  
Novel Method

Purpose: to develop a method for receiver operating characteristics (ROC) studies in mammography Material and Methods: We developed a phantom based on excised breast tissue and overlay tiles that could be arranged in an arbitrary pattern across the surface of the breast tissue. Some of the tiles contained structures simulating calcifications or masses that produced image contrast near the experimentally determined detection threshold. Based on this phantom, a methodology for performing ROC studies in mammography was developed. the ROC curves were constructed from reporting schemes filled in by radiologists at five different laboratories. the curves were determined by a novel method: a non-linear least-squares fit of a mathematical model to the data Results: There were large differences among the areas under the ROC curves obtained from the five laboratories

Optimum Window Size for Quantitating Trace Elements Using Linear Least Squares Fit With Eels

1999 ◽  
Vol 5 (S2) ◽  
pp. 938-939
Author(s):  
Ruoya Ho ◽  
Jiang Lin Feng ◽  
Zhifeng Shao ◽  
Andrew P. Somlyo
Keyword(s):  
Least Squares ◽  
Window Size ◽  
Dry Weight ◽  
Elemental Distribution ◽  
Small Signal ◽  
Linear Least Squares ◽  
Experimental Conditions ◽  
Large Window ◽  
Small Window ◽  
Least Squares Fit

Trace element quantitation with EELS is a powerful tool for providing information about elemental distribution in biological and materials sciences. Linear least squares fit (LSF) to standards of the signal and the background is a widely used method for retrieving small signals superimposed on a large, slowly varying background. Theoretically, for most accurate background fitting, the fitting window should be as large as possible. However, when using a large window, there may be a mismatch between the fine structure of the standard and the measured background, due to instability of the instrument and the experimental conditions and radiation damage. This mismatch, distributed over the entire window, can severely affect the accuracy of the small signal extracted with LSF. To overcome this problem, a small window, slightly larger than the signal is often used. Fig. 1 shows the results of extracting a small Ca signal by using respectively, a large and a small window for fitting an EELS spectrum collected from a specimen containing 26.3 mmol/kg dry weight Ca.

Spreadsheet Approach to the Linear Least Squares Fit

1996 ◽  
Vol 1 (1) ◽  
pp. 1-13
Author(s):  
Marie L. Carman ◽  
Thomas G. Chasteen
Keyword(s):  
Least Squares ◽  
Linear Least Squares ◽  
Least Squares Fit
Sign in / Sign up

Export Citation Format

Share Document