The effects of 3D region-based compression on the performance of an automatic lung nodule detection system

Benjamin L. Odry, Karthik Krishnan, Mariappan Nadar, David P. Naidich, Carol L. Novak, Michael W Marcellin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

There is growing interest in computer aided diagnosis applications including automatic detection of lung nodules from multislice computed tomography (CT). However the increase in the number and size of CT datasets introduces high costs for data storage and transmission, and becomes an obstacle to routine clinical exam as well as hindering widespread utilization of computerized applications. We investigated the effects of 3D lossy region-based JPEG2000 standard compression on the results of an automatic lung nodule detection system. As the algorithm detects the lungs within the datasets, we used this lung segmentation to define a region of interest (ROI) where the compression should be of higher fidelity. We tested 4 methods of 3D compression: 1) default compression of the whole image. 2) default compression of segmented lungs with masking out all non-lung regions, 3) ROI-based compression as specified in the JPEG2000 standard and 4) compression where voxels in the ROI are weighted to be given emphasis in the encoding. We tested 7 compression ratios per method: 1, 4, 6, 8, 10, 20, and 30 to 1. We then evaluated our experimental CAD algorithm on 10 patients with 67 documented nodules initially identified on the decompressed data Sensitivities and false positive rates were compared for the various compression methods and ratios. We found that region-based compression generally performs better than default compression. The sensitivity with default compression decreased from 85% at no compression to 61% at 30:1 compression, a decrease of 25%, whereas the masked compression method saw a decreased in sensitivity on only 13.5% at maximum compression. At compression levels up to 10:1, all 3 region-based compression methods had decreases in sensitivity of 7.5% or less. Detection of small nodules (< 4mm in diameter) was more affected by compression than detection of large nodules: sensitivity to calcified nodules was less affected by compression than to non-calcified nodules.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
EditorsJ.M. Fitzpatrick, J.M. Reinhardt
Pages883-894
Number of pages12
Volume5747
EditionII
DOIs
StatePublished - 2005
EventMedical Imaging 2005 - Image Processing - San Diego, CA, United States
Duration: Feb 13 2005Feb 17 2005

Other

OtherMedical Imaging 2005 - Image Processing
CountryUnited States
CitySan Diego, CA
Period2/13/052/17/05

Fingerprint

Tomography
Computer aided diagnosis
Computer aided design
Data storage equipment
Costs

Keywords

  • CAD
  • Computed Tomography
  • JPEG2000
  • Lung Nodule
  • Region-based compression

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Odry, B. L., Krishnan, K., Nadar, M., Naidich, D. P., Novak, C. L., & Marcellin, M. W. (2005). The effects of 3D region-based compression on the performance of an automatic lung nodule detection system. In J. M. Fitzpatrick, & J. M. Reinhardt (Eds.), Progress in Biomedical Optics and Imaging - Proceedings of SPIE (II ed., Vol. 5747, pp. 883-894). [92] https://doi.org/10.1117/12.595475

The effects of 3D region-based compression on the performance of an automatic lung nodule detection system. / Odry, Benjamin L.; Krishnan, Karthik; Nadar, Mariappan; Naidich, David P.; Novak, Carol L.; Marcellin, Michael W.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. ed. / J.M. Fitzpatrick; J.M. Reinhardt. Vol. 5747 II. ed. 2005. p. 883-894 92.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Odry, BL, Krishnan, K, Nadar, M, Naidich, DP, Novak, CL & Marcellin, MW 2005, The effects of 3D region-based compression on the performance of an automatic lung nodule detection system. in JM Fitzpatrick & JM Reinhardt (eds), Progress in Biomedical Optics and Imaging - Proceedings of SPIE. II edn, vol. 5747, 92, pp. 883-894, Medical Imaging 2005 - Image Processing, San Diego, CA, United States, 2/13/05. https://doi.org/10.1117/12.595475
Odry BL, Krishnan K, Nadar M, Naidich DP, Novak CL, Marcellin MW. The effects of 3D region-based compression on the performance of an automatic lung nodule detection system. In Fitzpatrick JM, Reinhardt JM, editors, Progress in Biomedical Optics and Imaging - Proceedings of SPIE. II ed. Vol. 5747. 2005. p. 883-894. 92 https://doi.org/10.1117/12.595475
Odry, Benjamin L. ; Krishnan, Karthik ; Nadar, Mariappan ; Naidich, David P. ; Novak, Carol L. ; Marcellin, Michael W. / The effects of 3D region-based compression on the performance of an automatic lung nodule detection system. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. editor / J.M. Fitzpatrick ; J.M. Reinhardt. Vol. 5747 II. ed. 2005. pp. 883-894
@inproceedings{a91915881412423681d082dab66a1e31,
title = "The effects of 3D region-based compression on the performance of an automatic lung nodule detection system",
abstract = "There is growing interest in computer aided diagnosis applications including automatic detection of lung nodules from multislice computed tomography (CT). However the increase in the number and size of CT datasets introduces high costs for data storage and transmission, and becomes an obstacle to routine clinical exam as well as hindering widespread utilization of computerized applications. We investigated the effects of 3D lossy region-based JPEG2000 standard compression on the results of an automatic lung nodule detection system. As the algorithm detects the lungs within the datasets, we used this lung segmentation to define a region of interest (ROI) where the compression should be of higher fidelity. We tested 4 methods of 3D compression: 1) default compression of the whole image. 2) default compression of segmented lungs with masking out all non-lung regions, 3) ROI-based compression as specified in the JPEG2000 standard and 4) compression where voxels in the ROI are weighted to be given emphasis in the encoding. We tested 7 compression ratios per method: 1, 4, 6, 8, 10, 20, and 30 to 1. We then evaluated our experimental CAD algorithm on 10 patients with 67 documented nodules initially identified on the decompressed data Sensitivities and false positive rates were compared for the various compression methods and ratios. We found that region-based compression generally performs better than default compression. The sensitivity with default compression decreased from 85{\%} at no compression to 61{\%} at 30:1 compression, a decrease of 25{\%}, whereas the masked compression method saw a decreased in sensitivity on only 13.5{\%} at maximum compression. At compression levels up to 10:1, all 3 region-based compression methods had decreases in sensitivity of 7.5{\%} or less. Detection of small nodules (< 4mm in diameter) was more affected by compression than detection of large nodules: sensitivity to calcified nodules was less affected by compression than to non-calcified nodules.",
keywords = "CAD, Computed Tomography, JPEG2000, Lung Nodule, Region-based compression",
author = "Odry, {Benjamin L.} and Karthik Krishnan and Mariappan Nadar and Naidich, {David P.} and Novak, {Carol L.} and Marcellin, {Michael W}",
year = "2005",
doi = "10.1117/12.595475",
language = "English (US)",
volume = "5747",
pages = "883--894",
editor = "J.M. Fitzpatrick and J.M. Reinhardt",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
edition = "II",

}

TY - GEN

T1 - The effects of 3D region-based compression on the performance of an automatic lung nodule detection system

AU - Odry, Benjamin L.

AU - Krishnan, Karthik

AU - Nadar, Mariappan

AU - Naidich, David P.

AU - Novak, Carol L.

AU - Marcellin, Michael W

PY - 2005

Y1 - 2005

N2 - There is growing interest in computer aided diagnosis applications including automatic detection of lung nodules from multislice computed tomography (CT). However the increase in the number and size of CT datasets introduces high costs for data storage and transmission, and becomes an obstacle to routine clinical exam as well as hindering widespread utilization of computerized applications. We investigated the effects of 3D lossy region-based JPEG2000 standard compression on the results of an automatic lung nodule detection system. As the algorithm detects the lungs within the datasets, we used this lung segmentation to define a region of interest (ROI) where the compression should be of higher fidelity. We tested 4 methods of 3D compression: 1) default compression of the whole image. 2) default compression of segmented lungs with masking out all non-lung regions, 3) ROI-based compression as specified in the JPEG2000 standard and 4) compression where voxels in the ROI are weighted to be given emphasis in the encoding. We tested 7 compression ratios per method: 1, 4, 6, 8, 10, 20, and 30 to 1. We then evaluated our experimental CAD algorithm on 10 patients with 67 documented nodules initially identified on the decompressed data Sensitivities and false positive rates were compared for the various compression methods and ratios. We found that region-based compression generally performs better than default compression. The sensitivity with default compression decreased from 85% at no compression to 61% at 30:1 compression, a decrease of 25%, whereas the masked compression method saw a decreased in sensitivity on only 13.5% at maximum compression. At compression levels up to 10:1, all 3 region-based compression methods had decreases in sensitivity of 7.5% or less. Detection of small nodules (< 4mm in diameter) was more affected by compression than detection of large nodules: sensitivity to calcified nodules was less affected by compression than to non-calcified nodules.

AB - There is growing interest in computer aided diagnosis applications including automatic detection of lung nodules from multislice computed tomography (CT). However the increase in the number and size of CT datasets introduces high costs for data storage and transmission, and becomes an obstacle to routine clinical exam as well as hindering widespread utilization of computerized applications. We investigated the effects of 3D lossy region-based JPEG2000 standard compression on the results of an automatic lung nodule detection system. As the algorithm detects the lungs within the datasets, we used this lung segmentation to define a region of interest (ROI) where the compression should be of higher fidelity. We tested 4 methods of 3D compression: 1) default compression of the whole image. 2) default compression of segmented lungs with masking out all non-lung regions, 3) ROI-based compression as specified in the JPEG2000 standard and 4) compression where voxels in the ROI are weighted to be given emphasis in the encoding. We tested 7 compression ratios per method: 1, 4, 6, 8, 10, 20, and 30 to 1. We then evaluated our experimental CAD algorithm on 10 patients with 67 documented nodules initially identified on the decompressed data Sensitivities and false positive rates were compared for the various compression methods and ratios. We found that region-based compression generally performs better than default compression. The sensitivity with default compression decreased from 85% at no compression to 61% at 30:1 compression, a decrease of 25%, whereas the masked compression method saw a decreased in sensitivity on only 13.5% at maximum compression. At compression levels up to 10:1, all 3 region-based compression methods had decreases in sensitivity of 7.5% or less. Detection of small nodules (< 4mm in diameter) was more affected by compression than detection of large nodules: sensitivity to calcified nodules was less affected by compression than to non-calcified nodules.

KW - CAD

KW - Computed Tomography

KW - JPEG2000

KW - Lung Nodule

KW - Region-based compression

UR - http://www.scopus.com/inward/record.url?scp=23844519369&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=23844519369&partnerID=8YFLogxK

U2 - 10.1117/12.595475

DO - 10.1117/12.595475

M3 - Conference contribution

AN - SCOPUS:23844519369

VL - 5747

SP - 883

EP - 894

BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

A2 - Fitzpatrick, J.M.

A2 - Reinhardt, J.M.

ER -