An active learning approach for rapid characterization of endothelial cells in human tumors

Raghav K. Padmanabhan, Vinay H. Somasundar, Sandra D. Griffith, Jianliang Zhu, Drew Samoyedny, Kay See Tan, Jiahao Hu, Xuejun Liao, Lawrence Carin, Sam S. Yoon, Keith T. Flaherty, Robert S. DiPaola, Daniel F. Heitjan, Priti Lal, Michael D. Feldman, Badrinath Roysam, William M.F. Lee

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Currently, no available pathological or molecular measures of tumor angiogenesis predict response to antiangiogenic therapies used in clinical practice. Recognizing that tumor endothelial cells (EC) and EC activation and survival signaling are the direct targets of these therapies, we sought to develop an automated platform for quantifying activity of critical signaling pathways and other biological events in EC of patient tumors by histopathology. Computer image analysis of EC in highly heterogeneous human tumors by a statistical classifier trained using examples selected by human experts performed poorly due to subjectivity and selection bias. We hypothesized that the analysis can be optimized by a more active process to aid experts in identifying informative training examples. To test this hypothesis, we incorporated a novel active learning (AL) algorithm into FARSIGHT image analysis software that aids the expert by seeking out informative examples for the operator to label. The resulting FARSIGHT-AL system identified EC with specificity and sensitivity consistently greater than 0.9 and outperformed traditional supervised classification algorithms. The system modeled individual operator preferences and generated reproducible results. Using the results of EC classification, we also quantified proliferation (Ki67) and activity in important signal transduction pathways (MAP kinase, STAT3) in immunostained human clear cell renal cell carcinoma and other tumors. FARSIGHT-AL enables characterization of EC in conventionally preserved human tumors in a more automated process suitable for testing and validating in clinical trials. The results of our study support a unique opportunity for quantifying angiogenesis in a manner that can now be tested for its ability to identify novel predictive and response biomarkers.

Original languageEnglish (US)
Article numbere90495
JournalPLoS One
Volume9
Issue number3
DOIs
StatePublished - Mar 6 2014

Fingerprint

Problem-Based Learning
Endothelial cells
endothelial cells
Tumors
learning
Endothelial Cells
neoplasms
Neoplasms
angiogenesis
Image analysis
image analysis
Signal transduction
therapeutics
Aptitude
Critical Pathways
Selection Bias
Biomarkers
kidney cells
mitogen-activated protein kinase
Renal Cell Carcinoma

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Agricultural and Biological Sciences(all)

Cite this

Padmanabhan, R. K., Somasundar, V. H., Griffith, S. D., Zhu, J., Samoyedny, D., Tan, K. S., ... Lee, W. M. F. (2014). An active learning approach for rapid characterization of endothelial cells in human tumors. PLoS One, 9(3), [e90495]. https://doi.org/10.1371/journal.pone.0090495

An active learning approach for rapid characterization of endothelial cells in human tumors. / Padmanabhan, Raghav K.; Somasundar, Vinay H.; Griffith, Sandra D.; Zhu, Jianliang; Samoyedny, Drew; Tan, Kay See; Hu, Jiahao; Liao, Xuejun; Carin, Lawrence; Yoon, Sam S.; Flaherty, Keith T.; DiPaola, Robert S.; Heitjan, Daniel F.; Lal, Priti; Feldman, Michael D.; Roysam, Badrinath; Lee, William M.F.

In: PLoS One, Vol. 9, No. 3, e90495, 06.03.2014.

Research output: Contribution to journalArticle

Padmanabhan, RK, Somasundar, VH, Griffith, SD, Zhu, J, Samoyedny, D, Tan, KS, Hu, J, Liao, X, Carin, L, Yoon, SS, Flaherty, KT, DiPaola, RS, Heitjan, DF, Lal, P, Feldman, MD, Roysam, B & Lee, WMF 2014, 'An active learning approach for rapid characterization of endothelial cells in human tumors', PLoS One, vol. 9, no. 3, e90495. https://doi.org/10.1371/journal.pone.0090495
Padmanabhan RK, Somasundar VH, Griffith SD, Zhu J, Samoyedny D, Tan KS et al. An active learning approach for rapid characterization of endothelial cells in human tumors. PLoS One. 2014 Mar 6;9(3). e90495. https://doi.org/10.1371/journal.pone.0090495
Padmanabhan, Raghav K. ; Somasundar, Vinay H. ; Griffith, Sandra D. ; Zhu, Jianliang ; Samoyedny, Drew ; Tan, Kay See ; Hu, Jiahao ; Liao, Xuejun ; Carin, Lawrence ; Yoon, Sam S. ; Flaherty, Keith T. ; DiPaola, Robert S. ; Heitjan, Daniel F. ; Lal, Priti ; Feldman, Michael D. ; Roysam, Badrinath ; Lee, William M.F. / An active learning approach for rapid characterization of endothelial cells in human tumors. In: PLoS One. 2014 ; Vol. 9, No. 3.
@article{8868faa399584d1a97dd9474356ac9c0,
title = "An active learning approach for rapid characterization of endothelial cells in human tumors",
abstract = "Currently, no available pathological or molecular measures of tumor angiogenesis predict response to antiangiogenic therapies used in clinical practice. Recognizing that tumor endothelial cells (EC) and EC activation and survival signaling are the direct targets of these therapies, we sought to develop an automated platform for quantifying activity of critical signaling pathways and other biological events in EC of patient tumors by histopathology. Computer image analysis of EC in highly heterogeneous human tumors by a statistical classifier trained using examples selected by human experts performed poorly due to subjectivity and selection bias. We hypothesized that the analysis can be optimized by a more active process to aid experts in identifying informative training examples. To test this hypothesis, we incorporated a novel active learning (AL) algorithm into FARSIGHT image analysis software that aids the expert by seeking out informative examples for the operator to label. The resulting FARSIGHT-AL system identified EC with specificity and sensitivity consistently greater than 0.9 and outperformed traditional supervised classification algorithms. The system modeled individual operator preferences and generated reproducible results. Using the results of EC classification, we also quantified proliferation (Ki67) and activity in important signal transduction pathways (MAP kinase, STAT3) in immunostained human clear cell renal cell carcinoma and other tumors. FARSIGHT-AL enables characterization of EC in conventionally preserved human tumors in a more automated process suitable for testing and validating in clinical trials. The results of our study support a unique opportunity for quantifying angiogenesis in a manner that can now be tested for its ability to identify novel predictive and response biomarkers.",
author = "Padmanabhan, {Raghav K.} and Somasundar, {Vinay H.} and Griffith, {Sandra D.} and Jianliang Zhu and Drew Samoyedny and Tan, {Kay See} and Jiahao Hu and Xuejun Liao and Lawrence Carin and Yoon, {Sam S.} and Flaherty, {Keith T.} and DiPaola, {Robert S.} and Heitjan, {Daniel F.} and Priti Lal and Feldman, {Michael D.} and Badrinath Roysam and Lee, {William M.F.}",
year = "2014",
month = "3",
day = "6",
doi = "10.1371/journal.pone.0090495",
language = "English (US)",
volume = "9",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "3",

}

TY - JOUR

T1 - An active learning approach for rapid characterization of endothelial cells in human tumors

AU - Padmanabhan, Raghav K.

AU - Somasundar, Vinay H.

AU - Griffith, Sandra D.

AU - Zhu, Jianliang

AU - Samoyedny, Drew

AU - Tan, Kay See

AU - Hu, Jiahao

AU - Liao, Xuejun

AU - Carin, Lawrence

AU - Yoon, Sam S.

AU - Flaherty, Keith T.

AU - DiPaola, Robert S.

AU - Heitjan, Daniel F.

AU - Lal, Priti

AU - Feldman, Michael D.

AU - Roysam, Badrinath

AU - Lee, William M.F.

PY - 2014/3/6

Y1 - 2014/3/6

N2 - Currently, no available pathological or molecular measures of tumor angiogenesis predict response to antiangiogenic therapies used in clinical practice. Recognizing that tumor endothelial cells (EC) and EC activation and survival signaling are the direct targets of these therapies, we sought to develop an automated platform for quantifying activity of critical signaling pathways and other biological events in EC of patient tumors by histopathology. Computer image analysis of EC in highly heterogeneous human tumors by a statistical classifier trained using examples selected by human experts performed poorly due to subjectivity and selection bias. We hypothesized that the analysis can be optimized by a more active process to aid experts in identifying informative training examples. To test this hypothesis, we incorporated a novel active learning (AL) algorithm into FARSIGHT image analysis software that aids the expert by seeking out informative examples for the operator to label. The resulting FARSIGHT-AL system identified EC with specificity and sensitivity consistently greater than 0.9 and outperformed traditional supervised classification algorithms. The system modeled individual operator preferences and generated reproducible results. Using the results of EC classification, we also quantified proliferation (Ki67) and activity in important signal transduction pathways (MAP kinase, STAT3) in immunostained human clear cell renal cell carcinoma and other tumors. FARSIGHT-AL enables characterization of EC in conventionally preserved human tumors in a more automated process suitable for testing and validating in clinical trials. The results of our study support a unique opportunity for quantifying angiogenesis in a manner that can now be tested for its ability to identify novel predictive and response biomarkers.

AB - Currently, no available pathological or molecular measures of tumor angiogenesis predict response to antiangiogenic therapies used in clinical practice. Recognizing that tumor endothelial cells (EC) and EC activation and survival signaling are the direct targets of these therapies, we sought to develop an automated platform for quantifying activity of critical signaling pathways and other biological events in EC of patient tumors by histopathology. Computer image analysis of EC in highly heterogeneous human tumors by a statistical classifier trained using examples selected by human experts performed poorly due to subjectivity and selection bias. We hypothesized that the analysis can be optimized by a more active process to aid experts in identifying informative training examples. To test this hypothesis, we incorporated a novel active learning (AL) algorithm into FARSIGHT image analysis software that aids the expert by seeking out informative examples for the operator to label. The resulting FARSIGHT-AL system identified EC with specificity and sensitivity consistently greater than 0.9 and outperformed traditional supervised classification algorithms. The system modeled individual operator preferences and generated reproducible results. Using the results of EC classification, we also quantified proliferation (Ki67) and activity in important signal transduction pathways (MAP kinase, STAT3) in immunostained human clear cell renal cell carcinoma and other tumors. FARSIGHT-AL enables characterization of EC in conventionally preserved human tumors in a more automated process suitable for testing and validating in clinical trials. The results of our study support a unique opportunity for quantifying angiogenesis in a manner that can now be tested for its ability to identify novel predictive and response biomarkers.

UR - http://www.scopus.com/inward/record.url?scp=84897412103&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897412103&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0090495

DO - 10.1371/journal.pone.0090495

M3 - Article

C2 - 24603893

AN - SCOPUS:84897412103

VL - 9

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 3

M1 - e90495

ER -