Computer-aided diagnosis for interstitial infiltrates in chest radiographs: Optical-density dependence of texture measures

Junji Morishita, Kunio Doi, Shigehiko Katsuragawa, Laurence Monnier-Cholley, Heber MarMahon

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

We have been developing a computerized scheme for automated detection and characterization of interstitial infiltrates based on the Fourier transform of lung texture. To improve the performance of the scheme, which was developed using digitized screen-film radiographs, optical-density dependence of both the gradient of the film used and the system noise associated with the laser scanner were investigated. Two hundred chest radiographs, including 100 abnormal cases with interstitial infiltrates, were digitized using a laser scanner. The root-mean-square (RMS) variations and the first moments of the power spectra, which correspond to the magnitude and coarseness of lung texture, were determined by Fourier transform of lung textures in numerous regions of interest (ROIs). The RMS variation was dependent upon the average optical density in the ROI, though no obvious trend existed for the first moment of the power spectrum. Dependence of the RMS variations on optical density was corrected for using the gradient curve of the film. Also, system noise associated with the laser scanner was corrected. Results indicated that the specificity was improved from 81% (without correction) to 89% (with corrections), without any loss of sensitivity (90%). Thus, the correspondence between the computer output and consensus interpretation of radiologists was improved with the new scheme compared to the previous one. This improved computerized scheme may be useful to radiologists in detecting interstitial infiltrates in chest radiographs.

Original languageEnglish
Pages (from-to)1515-1522
Number of pages8
JournalMedical Physics
Volume22
Issue number9
DOIs
Publication statusPublished - Jan 1 1995
Externally publishedYes

Fingerprint

Lasers
Thorax
Fourier Analysis
Lung
Radiologists

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

Computer-aided diagnosis for interstitial infiltrates in chest radiographs : Optical-density dependence of texture measures. / Morishita, Junji; Doi, Kunio; Katsuragawa, Shigehiko; Monnier-Cholley, Laurence; MarMahon, Heber.

In: Medical Physics, Vol. 22, No. 9, 01.01.1995, p. 1515-1522.

Research output: Contribution to journalArticle

Morishita, Junji ; Doi, Kunio ; Katsuragawa, Shigehiko ; Monnier-Cholley, Laurence ; MarMahon, Heber. / Computer-aided diagnosis for interstitial infiltrates in chest radiographs : Optical-density dependence of texture measures. In: Medical Physics. 1995 ; Vol. 22, No. 9. pp. 1515-1522.
@article{674554d46a524ebeba9a36c414088e3a,
title = "Computer-aided diagnosis for interstitial infiltrates in chest radiographs: Optical-density dependence of texture measures",
abstract = "We have been developing a computerized scheme for automated detection and characterization of interstitial infiltrates based on the Fourier transform of lung texture. To improve the performance of the scheme, which was developed using digitized screen-film radiographs, optical-density dependence of both the gradient of the film used and the system noise associated with the laser scanner were investigated. Two hundred chest radiographs, including 100 abnormal cases with interstitial infiltrates, were digitized using a laser scanner. The root-mean-square (RMS) variations and the first moments of the power spectra, which correspond to the magnitude and coarseness of lung texture, were determined by Fourier transform of lung textures in numerous regions of interest (ROIs). The RMS variation was dependent upon the average optical density in the ROI, though no obvious trend existed for the first moment of the power spectrum. Dependence of the RMS variations on optical density was corrected for using the gradient curve of the film. Also, system noise associated with the laser scanner was corrected. Results indicated that the specificity was improved from 81{\%} (without correction) to 89{\%} (with corrections), without any loss of sensitivity (90{\%}). Thus, the correspondence between the computer output and consensus interpretation of radiologists was improved with the new scheme compared to the previous one. This improved computerized scheme may be useful to radiologists in detecting interstitial infiltrates in chest radiographs.",
author = "Junji Morishita and Kunio Doi and Shigehiko Katsuragawa and Laurence Monnier-Cholley and Heber MarMahon",
year = "1995",
month = "1",
day = "1",
doi = "10.1118/1.597419",
language = "English",
volume = "22",
pages = "1515--1522",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "9",

}

TY - JOUR

T1 - Computer-aided diagnosis for interstitial infiltrates in chest radiographs

T2 - Optical-density dependence of texture measures

AU - Morishita, Junji

AU - Doi, Kunio

AU - Katsuragawa, Shigehiko

AU - Monnier-Cholley, Laurence

AU - MarMahon, Heber

PY - 1995/1/1

Y1 - 1995/1/1

N2 - We have been developing a computerized scheme for automated detection and characterization of interstitial infiltrates based on the Fourier transform of lung texture. To improve the performance of the scheme, which was developed using digitized screen-film radiographs, optical-density dependence of both the gradient of the film used and the system noise associated with the laser scanner were investigated. Two hundred chest radiographs, including 100 abnormal cases with interstitial infiltrates, were digitized using a laser scanner. The root-mean-square (RMS) variations and the first moments of the power spectra, which correspond to the magnitude and coarseness of lung texture, were determined by Fourier transform of lung textures in numerous regions of interest (ROIs). The RMS variation was dependent upon the average optical density in the ROI, though no obvious trend existed for the first moment of the power spectrum. Dependence of the RMS variations on optical density was corrected for using the gradient curve of the film. Also, system noise associated with the laser scanner was corrected. Results indicated that the specificity was improved from 81% (without correction) to 89% (with corrections), without any loss of sensitivity (90%). Thus, the correspondence between the computer output and consensus interpretation of radiologists was improved with the new scheme compared to the previous one. This improved computerized scheme may be useful to radiologists in detecting interstitial infiltrates in chest radiographs.

AB - We have been developing a computerized scheme for automated detection and characterization of interstitial infiltrates based on the Fourier transform of lung texture. To improve the performance of the scheme, which was developed using digitized screen-film radiographs, optical-density dependence of both the gradient of the film used and the system noise associated with the laser scanner were investigated. Two hundred chest radiographs, including 100 abnormal cases with interstitial infiltrates, were digitized using a laser scanner. The root-mean-square (RMS) variations and the first moments of the power spectra, which correspond to the magnitude and coarseness of lung texture, were determined by Fourier transform of lung textures in numerous regions of interest (ROIs). The RMS variation was dependent upon the average optical density in the ROI, though no obvious trend existed for the first moment of the power spectrum. Dependence of the RMS variations on optical density was corrected for using the gradient curve of the film. Also, system noise associated with the laser scanner was corrected. Results indicated that the specificity was improved from 81% (without correction) to 89% (with corrections), without any loss of sensitivity (90%). Thus, the correspondence between the computer output and consensus interpretation of radiologists was improved with the new scheme compared to the previous one. This improved computerized scheme may be useful to radiologists in detecting interstitial infiltrates in chest radiographs.

UR - http://www.scopus.com/inward/record.url?scp=0029125788&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029125788&partnerID=8YFLogxK

U2 - 10.1118/1.597419

DO - 10.1118/1.597419

M3 - Article

C2 - 8531883

AN - SCOPUS:0029125788

VL - 22

SP - 1515

EP - 1522

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 9

ER -