Automated tuberculosis screening using image processing tools

B. Castaneda; N. G. Aguilar; J. Ticona; D. Kanashiro; R. Lavarello; L. Huaroto

doi:10.1109/PAHCE.2010.5474590

Automated tuberculosis screening using image processing tools

B. Castaneda, N. G. Aguilar, J. Ticona, D. Kanashiro, R. Lavarello, L. Huaroto

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

We present a preliminary work as a proof of concept on how image processing algorithms can be applied to detect and diagnose Tuberculosis in microscopic images of sputum samples stained with the Ziehl-Neelsen method. 300 images were acquired at the Hospital Nacional Dos de Mayo and processed using edge detection and mathematical morphology to extract objects of interest. Bacilli are discriminated from these objects applying a classifier based on the Mahalanobis distance and using shape characteristics as features. Results show a specificity value over 90% which is close to previously reported attempts on samples processed with Auramine.

Original language	English
Title of host publication	Pan American Health Care Exchanges, PAHCE 2010
Pages	111
Number of pages	1
DOIs	https://doi.org/10.1109/PAHCE.2010.5474590
State	Published - 2010
Event	Pan American Health Care Exchanges, PAHCE 2010 - Lima, Peru Duration: 15 Mar 2010 → 19 Mar 2010

Publication series

Name	Pan American Health Care Exchanges, PAHCE 2010

Conference

Conference	Pan American Health Care Exchanges, PAHCE 2010
Country/Territory	Peru
City	Lima
Period	15/03/10 → 19/03/10

Keywords

Diagnosis
Image processing
Pattern recognition
Tuberculosis

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/PAHCE.2010.5474590

Cite this

@inproceedings{9766c94e719248c594e96767f2a88f01,

title = "Automated tuberculosis screening using image processing tools",

abstract = "We present a preliminary work as a proof of concept on how image processing algorithms can be applied to detect and diagnose Tuberculosis in microscopic images of sputum samples stained with the Ziehl-Neelsen method. 300 images were acquired at the Hospital Nacional Dos de Mayo and processed using edge detection and mathematical morphology to extract objects of interest. Bacilli are discriminated from these objects applying a classifier based on the Mahalanobis distance and using shape characteristics as features. Results show a specificity value over 90% which is close to previously reported attempts on samples processed with Auramine.",

keywords = "Diagnosis, Image processing, Pattern recognition, Tuberculosis",

author = "B. Castaneda and Aguilar, {N. G.} and J. Ticona and D. Kanashiro and R. Lavarello and L. Huaroto",

year = "2010",

doi = "10.1109/PAHCE.2010.5474590",

language = "English",

isbn = "9781424462926",

series = "Pan American Health Care Exchanges, PAHCE 2010",

pages = "111",

booktitle = "Pan American Health Care Exchanges, PAHCE 2010",

note = "Pan American Health Care Exchanges, PAHCE 2010 ; Conference date: 15-03-2010 Through 19-03-2010",

}

Castaneda, B, Aguilar, NG, Ticona, J, Kanashiro, D, Lavarello, R & Huaroto, L 2010, Automated tuberculosis screening using image processing tools. in Pan American Health Care Exchanges, PAHCE 2010., 5474590, Pan American Health Care Exchanges, PAHCE 2010, pp. 111, Pan American Health Care Exchanges, PAHCE 2010, Lima, Peru, 15/03/10. https://doi.org/10.1109/PAHCE.2010.5474590

TY - GEN

T1 - Automated tuberculosis screening using image processing tools

AU - Castaneda, B.

AU - Aguilar, N. G.

AU - Ticona, J.

AU - Kanashiro, D.

AU - Lavarello, R.

AU - Huaroto, L.

PY - 2010

Y1 - 2010

N2 - We present a preliminary work as a proof of concept on how image processing algorithms can be applied to detect and diagnose Tuberculosis in microscopic images of sputum samples stained with the Ziehl-Neelsen method. 300 images were acquired at the Hospital Nacional Dos de Mayo and processed using edge detection and mathematical morphology to extract objects of interest. Bacilli are discriminated from these objects applying a classifier based on the Mahalanobis distance and using shape characteristics as features. Results show a specificity value over 90% which is close to previously reported attempts on samples processed with Auramine.

AB - We present a preliminary work as a proof of concept on how image processing algorithms can be applied to detect and diagnose Tuberculosis in microscopic images of sputum samples stained with the Ziehl-Neelsen method. 300 images were acquired at the Hospital Nacional Dos de Mayo and processed using edge detection and mathematical morphology to extract objects of interest. Bacilli are discriminated from these objects applying a classifier based on the Mahalanobis distance and using shape characteristics as features. Results show a specificity value over 90% which is close to previously reported attempts on samples processed with Auramine.

KW - Diagnosis

KW - Image processing

KW - Pattern recognition

KW - Tuberculosis

UR - http://www.scopus.com/inward/record.url?scp=77954247751&partnerID=8YFLogxK

U2 - 10.1109/PAHCE.2010.5474590

DO - 10.1109/PAHCE.2010.5474590

M3 - Conference contribution

AN - SCOPUS:77954247751

SN - 9781424462926

T3 - Pan American Health Care Exchanges, PAHCE 2010

SP - 111

BT - Pan American Health Care Exchanges, PAHCE 2010

T2 - Pan American Health Care Exchanges, PAHCE 2010

Y2 - 15 March 2010 through 19 March 2010

ER -

Automated tuberculosis screening using image processing tools

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this