Discriminative spectral pattern analysis for positive margin detection of prostate cancer specimens using light reflectance spectroscopy

Rahilsadat Hosseini, Henry Chan, Payal Kapur, Jeffrey A Cadeddu, Hani Liu, Shouyi Wang

Research output: Contribution to journalArticle

Abstract

For localized prostate cancer, one treatment is prostatectomy which surgically removes the prostate gland. However, some undetectable cancer cells may be left as positive surgical margins, leading to a high risk of cancer recurrence. It is highly desirable to develop a portable and accurate classification methodology that detects positive margins on human prostate specimens immediately after their removal during surgery. This study applied data mining techniques on the light reflectance spectroscopy (LRS) data taken from ex-vivo human specimens and developed a novel classification algorithm that could enable real-time, positive-margin identification during surgery. Specifically, the LRS measurements taken from human prostate specimens ex vivo were classified to normal or cancerous tissue with support vector machines and were also classified to normal, cancerous and transition-to-cancer class with an ensemble of trees. The data in this study were highly overlapped and imbalanced among classes. We solved the overlapping issue by defining a middle class (transition-to-cancer), and by optimizing a moving spectral window through the range of LRS. To solve the imbalanced problem, we removed irregular tissue measurements, followed by application of random undersampling from the majority class. We achieved sensitivity and specificity of 100% and 82% for binary classification.

Original languageEnglish (US)
Pages (from-to)156-166
Number of pages11
JournalIISE Transactions on Healthcare Systems Engineering
DOIs
StateAccepted/In press - Mar 30 2018

Fingerprint

Spectrum Analysis
Prostatic Neoplasms
cancer
Spectroscopy
Light
Prostate
Surgery
Tissue
Neoplasms
surgery
Data Mining
Support vector machines
Data mining
Prostatectomy
Cells
middle class
Recurrence
Sensitivity and Specificity
methodology

Keywords

  • data mining
  • ensemble
  • Light Reflectance Spectroscopy (LRS)
  • positive surgical margin
  • prostate cancer
  • random undersampling (RUS)
  • support vector machines (SVM)

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Safety Research
  • Public Health, Environmental and Occupational Health

Cite this

@article{bd811cc8873945f8a504f373ff5cb838,
title = "Discriminative spectral pattern analysis for positive margin detection of prostate cancer specimens using light reflectance spectroscopy",
abstract = "For localized prostate cancer, one treatment is prostatectomy which surgically removes the prostate gland. However, some undetectable cancer cells may be left as positive surgical margins, leading to a high risk of cancer recurrence. It is highly desirable to develop a portable and accurate classification methodology that detects positive margins on human prostate specimens immediately after their removal during surgery. This study applied data mining techniques on the light reflectance spectroscopy (LRS) data taken from ex-vivo human specimens and developed a novel classification algorithm that could enable real-time, positive-margin identification during surgery. Specifically, the LRS measurements taken from human prostate specimens ex vivo were classified to normal or cancerous tissue with support vector machines and were also classified to normal, cancerous and transition-to-cancer class with an ensemble of trees. The data in this study were highly overlapped and imbalanced among classes. We solved the overlapping issue by defining a middle class (transition-to-cancer), and by optimizing a moving spectral window through the range of LRS. To solve the imbalanced problem, we removed irregular tissue measurements, followed by application of random undersampling from the majority class. We achieved sensitivity and specificity of 100{\%} and 82{\%} for binary classification.",
keywords = "data mining, ensemble, Light Reflectance Spectroscopy (LRS), positive surgical margin, prostate cancer, random undersampling (RUS), support vector machines (SVM)",
author = "Rahilsadat Hosseini and Henry Chan and Payal Kapur and Cadeddu, {Jeffrey A} and Hani Liu and Shouyi Wang",
year = "2018",
month = "3",
day = "30",
doi = "10.1080/24725579.2018.1442378",
language = "English (US)",
pages = "156--166",
journal = "IISE Transactions on Healthcare Systems Engineering",
issn = "2472-5579",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - Discriminative spectral pattern analysis for positive margin detection of prostate cancer specimens using light reflectance spectroscopy

AU - Hosseini, Rahilsadat

AU - Chan, Henry

AU - Kapur, Payal

AU - Cadeddu, Jeffrey A

AU - Liu, Hani

AU - Wang, Shouyi

PY - 2018/3/30

Y1 - 2018/3/30

N2 - For localized prostate cancer, one treatment is prostatectomy which surgically removes the prostate gland. However, some undetectable cancer cells may be left as positive surgical margins, leading to a high risk of cancer recurrence. It is highly desirable to develop a portable and accurate classification methodology that detects positive margins on human prostate specimens immediately after their removal during surgery. This study applied data mining techniques on the light reflectance spectroscopy (LRS) data taken from ex-vivo human specimens and developed a novel classification algorithm that could enable real-time, positive-margin identification during surgery. Specifically, the LRS measurements taken from human prostate specimens ex vivo were classified to normal or cancerous tissue with support vector machines and were also classified to normal, cancerous and transition-to-cancer class with an ensemble of trees. The data in this study were highly overlapped and imbalanced among classes. We solved the overlapping issue by defining a middle class (transition-to-cancer), and by optimizing a moving spectral window through the range of LRS. To solve the imbalanced problem, we removed irregular tissue measurements, followed by application of random undersampling from the majority class. We achieved sensitivity and specificity of 100% and 82% for binary classification.

AB - For localized prostate cancer, one treatment is prostatectomy which surgically removes the prostate gland. However, some undetectable cancer cells may be left as positive surgical margins, leading to a high risk of cancer recurrence. It is highly desirable to develop a portable and accurate classification methodology that detects positive margins on human prostate specimens immediately after their removal during surgery. This study applied data mining techniques on the light reflectance spectroscopy (LRS) data taken from ex-vivo human specimens and developed a novel classification algorithm that could enable real-time, positive-margin identification during surgery. Specifically, the LRS measurements taken from human prostate specimens ex vivo were classified to normal or cancerous tissue with support vector machines and were also classified to normal, cancerous and transition-to-cancer class with an ensemble of trees. The data in this study were highly overlapped and imbalanced among classes. We solved the overlapping issue by defining a middle class (transition-to-cancer), and by optimizing a moving spectral window through the range of LRS. To solve the imbalanced problem, we removed irregular tissue measurements, followed by application of random undersampling from the majority class. We achieved sensitivity and specificity of 100% and 82% for binary classification.

KW - data mining

KW - ensemble

KW - Light Reflectance Spectroscopy (LRS)

KW - positive surgical margin

KW - prostate cancer

KW - random undersampling (RUS)

KW - support vector machines (SVM)

UR - http://www.scopus.com/inward/record.url?scp=85044787256&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85044787256&partnerID=8YFLogxK

U2 - 10.1080/24725579.2018.1442378

DO - 10.1080/24725579.2018.1442378

M3 - Article

AN - SCOPUS:85044787256

SP - 156

EP - 166

JO - IISE Transactions on Healthcare Systems Engineering

JF - IISE Transactions on Healthcare Systems Engineering

SN - 2472-5579

ER -