Active relearning for robust supervised classification of pulmonary emphysema

Sushravya Raghunath, Srinivasan Rajagopalan, Ronald A. Karwoski, Brian Jack Bartholmai, Richard A. Robb

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

1 Citation (Scopus)

Abstract

Radiologists are adept at recognizing the appearance of lung parenchymal abnormalities in CT scans. However, the inconsistent differential diagnosis, due to subjective aggregation, mandates supervised classification. Towards optimizing Emphysema classification, we introduce a physician-in-the-loop feedback approach in order to minimize uncertainty in the selected training samples. Using multi-view inductive learning with the training samples, an ensemble of Support Vector Machine (SVM) models, each based on a specific pair-wise dissimilarity metric, was constructed in less than six seconds. In the active relearning phase, the ensemble-expert label conflicts were resolved by an expert. This just-in-time feedback with unoptimized SVMs yielded 15% increase in classification accuracy and 25% reduction in the number of support vectors. The generality of relearning was assessed in the optimized parameter space of six different classifiers across seven dissimilarity metrics. The resultant average accuracy improved to 21%. The co-operative feedback method proposed here could enhance both diagnostic and staging throughput efficiency in chest radiology practice.

Original languageEnglish (US)
Title of host publicationProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume8315
DOIs
StatePublished - 2012
EventMedical Imaging 2012: Computer-Aided Diagnosis - San Diego, CA, United States
Duration: Feb 7 2012Feb 9 2012

Other

OtherMedical Imaging 2012: Computer-Aided Diagnosis
CountryUnited States
CitySan Diego, CA
Period2/7/122/9/12

Fingerprint

emphysema
Pulmonary Emphysema
Feedback
education
Radiology
physicians
Computerized tomography
radiology
chest
Emphysema
abnormalities
classifiers
lungs
learning
Uncertainty
Support vector machines
Labels
Differential Diagnosis
Classifiers
Thorax

Keywords

  • Active relearning
  • Emphysema
  • HRCT
  • Supervised classification
  • SVM

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Raghunath, S., Rajagopalan, S., Karwoski, R. A., Bartholmai, B. J., & Robb, R. A. (2012). Active relearning for robust supervised classification of pulmonary emphysema. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 8315). [83152Q] https://doi.org/10.1117/12.911648

Active relearning for robust supervised classification of pulmonary emphysema. / Raghunath, Sushravya; Rajagopalan, Srinivasan; Karwoski, Ronald A.; Bartholmai, Brian Jack; Robb, Richard A.

Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8315 2012. 83152Q.

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

Raghunath, S, Rajagopalan, S, Karwoski, RA, Bartholmai, BJ & Robb, RA 2012, Active relearning for robust supervised classification of pulmonary emphysema. in Progress in Biomedical Optics and Imaging - Proceedings of SPIE. vol. 8315, 83152Q, Medical Imaging 2012: Computer-Aided Diagnosis, San Diego, CA, United States, 2/7/12. https://doi.org/10.1117/12.911648
Raghunath S, Rajagopalan S, Karwoski RA, Bartholmai BJ, Robb RA. Active relearning for robust supervised classification of pulmonary emphysema. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8315. 2012. 83152Q https://doi.org/10.1117/12.911648
Raghunath, Sushravya ; Rajagopalan, Srinivasan ; Karwoski, Ronald A. ; Bartholmai, Brian Jack ; Robb, Richard A. / Active relearning for robust supervised classification of pulmonary emphysema. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 8315 2012.
@inproceedings{9556ba067c844b4e942fc37579dab45d,
title = "Active relearning for robust supervised classification of pulmonary emphysema",
abstract = "Radiologists are adept at recognizing the appearance of lung parenchymal abnormalities in CT scans. However, the inconsistent differential diagnosis, due to subjective aggregation, mandates supervised classification. Towards optimizing Emphysema classification, we introduce a physician-in-the-loop feedback approach in order to minimize uncertainty in the selected training samples. Using multi-view inductive learning with the training samples, an ensemble of Support Vector Machine (SVM) models, each based on a specific pair-wise dissimilarity metric, was constructed in less than six seconds. In the active relearning phase, the ensemble-expert label conflicts were resolved by an expert. This just-in-time feedback with unoptimized SVMs yielded 15{\%} increase in classification accuracy and 25{\%} reduction in the number of support vectors. The generality of relearning was assessed in the optimized parameter space of six different classifiers across seven dissimilarity metrics. The resultant average accuracy improved to 21{\%}. The co-operative feedback method proposed here could enhance both diagnostic and staging throughput efficiency in chest radiology practice.",
keywords = "Active relearning, Emphysema, HRCT, Supervised classification, SVM",
author = "Sushravya Raghunath and Srinivasan Rajagopalan and Karwoski, {Ronald A.} and Bartholmai, {Brian Jack} and Robb, {Richard A.}",
year = "2012",
doi = "10.1117/12.911648",
language = "English (US)",
isbn = "9780819489647",
volume = "8315",
booktitle = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",

}

TY - GEN

T1 - Active relearning for robust supervised classification of pulmonary emphysema

AU - Raghunath, Sushravya

AU - Rajagopalan, Srinivasan

AU - Karwoski, Ronald A.

AU - Bartholmai, Brian Jack

AU - Robb, Richard A.

PY - 2012

Y1 - 2012

N2 - Radiologists are adept at recognizing the appearance of lung parenchymal abnormalities in CT scans. However, the inconsistent differential diagnosis, due to subjective aggregation, mandates supervised classification. Towards optimizing Emphysema classification, we introduce a physician-in-the-loop feedback approach in order to minimize uncertainty in the selected training samples. Using multi-view inductive learning with the training samples, an ensemble of Support Vector Machine (SVM) models, each based on a specific pair-wise dissimilarity metric, was constructed in less than six seconds. In the active relearning phase, the ensemble-expert label conflicts were resolved by an expert. This just-in-time feedback with unoptimized SVMs yielded 15% increase in classification accuracy and 25% reduction in the number of support vectors. The generality of relearning was assessed in the optimized parameter space of six different classifiers across seven dissimilarity metrics. The resultant average accuracy improved to 21%. The co-operative feedback method proposed here could enhance both diagnostic and staging throughput efficiency in chest radiology practice.

AB - Radiologists are adept at recognizing the appearance of lung parenchymal abnormalities in CT scans. However, the inconsistent differential diagnosis, due to subjective aggregation, mandates supervised classification. Towards optimizing Emphysema classification, we introduce a physician-in-the-loop feedback approach in order to minimize uncertainty in the selected training samples. Using multi-view inductive learning with the training samples, an ensemble of Support Vector Machine (SVM) models, each based on a specific pair-wise dissimilarity metric, was constructed in less than six seconds. In the active relearning phase, the ensemble-expert label conflicts were resolved by an expert. This just-in-time feedback with unoptimized SVMs yielded 15% increase in classification accuracy and 25% reduction in the number of support vectors. The generality of relearning was assessed in the optimized parameter space of six different classifiers across seven dissimilarity metrics. The resultant average accuracy improved to 21%. The co-operative feedback method proposed here could enhance both diagnostic and staging throughput efficiency in chest radiology practice.

KW - Active relearning

KW - Emphysema

KW - HRCT

KW - Supervised classification

KW - SVM

UR - http://www.scopus.com/inward/record.url?scp=84874837997&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874837997&partnerID=8YFLogxK

U2 - 10.1117/12.911648

DO - 10.1117/12.911648

M3 - Conference contribution

SN - 9780819489647

VL - 8315

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

ER -