A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging

Shachi Mittal, Tomasz P. Wrobel, L. S. Leslie, Andre Kadjacsy-Balla, Rohit Bhargava

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

Abstract

High-definition (HD) Fourier transform infrared (FT-IR) spectroscopic imaging is an emerging technique that not only enables chemistry-based visualization of tissue constituents, and label free extraction of biochemical information but its higher spatial detail makes it a potentially useful platform to conduct digital pathology. This methodology, along with fast and efficient data analysis, can enable both quantitative and automated pathology. Here we demonstrate a combination of HD FT-IR spectroscopic imaging of breast tissue microarrays (TMAs) with data analysis algorithms to perform histologic analysis. The samples comprise four tissue states, namely hyperplasia, dysplasia, cancerous and normal. We identify various cell types which would act as biomarkers for breast cancer detection and differentiate between them using statistical pattern recognition tools i.e. Random Forest (RF) and Bayesian algorithms. Feature optimization is integrally carried out for the RF algorithm, reducing computation time as well as redundant spectral features. We achieved an order of magnitude reduction in the number of features with comparable prediction accuracy to that of the original feature set. Together, the demonstration of histology and selection of features paves the way for future applications in more complex models and rapid data acquisition.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2016
Subtitle of host publicationDigital Pathology
EditorsAnant Madabhushi, Metin N. Gurcan
PublisherSPIE
ISBN (Electronic)9781510600263
DOIs
StatePublished - Jan 1 2016
Event4th Medical Imaging 2016: Digital Pathology - San Diego, United States
Duration: Mar 2 2016Mar 3 2016

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume9791
ISSN (Print)1605-7422

Other

Other4th Medical Imaging 2016: Digital Pathology
CountryUnited States
CitySan Diego
Period3/2/163/3/16

Fingerprint

Fourier Analysis
breast
Fourier transforms
Breast
pathology
Pathology
Tissue
Infrared radiation
Imaging techniques
Histology
histology
biomarkers
Information Storage and Retrieval
Biomarkers
Microarrays
pattern recognition
Pattern recognition
data acquisition
Hyperplasia
Labels

Keywords

  • FTIR
  • High-Definition
  • Random Forest
  • automated pathology
  • breast cancer
  • four class classifier

ASJC Scopus subject areas

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

Cite this

Mittal, S., Wrobel, T. P., Leslie, L. S., Kadjacsy-Balla, A., & Bhargava, R. (2016). A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging. In A. Madabhushi, & M. N. Gurcan (Eds.), Medical Imaging 2016: Digital Pathology [979118] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 9791). SPIE. https://doi.org/10.1117/12.2217358

A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging. / Mittal, Shachi; Wrobel, Tomasz P.; Leslie, L. S.; Kadjacsy-Balla, Andre; Bhargava, Rohit.

Medical Imaging 2016: Digital Pathology. ed. / Anant Madabhushi; Metin N. Gurcan. SPIE, 2016. 979118 (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 9791).

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

Mittal, S, Wrobel, TP, Leslie, LS, Kadjacsy-Balla, A & Bhargava, R 2016, A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging. in A Madabhushi & MN Gurcan (eds), Medical Imaging 2016: Digital Pathology., 979118, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 9791, SPIE, 4th Medical Imaging 2016: Digital Pathology, San Diego, United States, 3/2/16. https://doi.org/10.1117/12.2217358
Mittal S, Wrobel TP, Leslie LS, Kadjacsy-Balla A, Bhargava R. A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging. In Madabhushi A, Gurcan MN, editors, Medical Imaging 2016: Digital Pathology. SPIE. 2016. 979118. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.2217358
Mittal, Shachi ; Wrobel, Tomasz P. ; Leslie, L. S. ; Kadjacsy-Balla, Andre ; Bhargava, Rohit. / A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging. Medical Imaging 2016: Digital Pathology. editor / Anant Madabhushi ; Metin N. Gurcan. SPIE, 2016. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{7de685782e3b4242897270311cab6141,
title = "A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging",
abstract = "High-definition (HD) Fourier transform infrared (FT-IR) spectroscopic imaging is an emerging technique that not only enables chemistry-based visualization of tissue constituents, and label free extraction of biochemical information but its higher spatial detail makes it a potentially useful platform to conduct digital pathology. This methodology, along with fast and efficient data analysis, can enable both quantitative and automated pathology. Here we demonstrate a combination of HD FT-IR spectroscopic imaging of breast tissue microarrays (TMAs) with data analysis algorithms to perform histologic analysis. The samples comprise four tissue states, namely hyperplasia, dysplasia, cancerous and normal. We identify various cell types which would act as biomarkers for breast cancer detection and differentiate between them using statistical pattern recognition tools i.e. Random Forest (RF) and Bayesian algorithms. Feature optimization is integrally carried out for the RF algorithm, reducing computation time as well as redundant spectral features. We achieved an order of magnitude reduction in the number of features with comparable prediction accuracy to that of the original feature set. Together, the demonstration of histology and selection of features paves the way for future applications in more complex models and rapid data acquisition.",
keywords = "FTIR, High-Definition, Random Forest, automated pathology, breast cancer, four class classifier",
author = "Shachi Mittal and Wrobel, {Tomasz P.} and Leslie, {L. S.} and Andre Kadjacsy-Balla and Rohit Bhargava",
year = "2016",
month = "1",
day = "1",
doi = "10.1117/12.2217358",
language = "English (US)",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",
editor = "Anant Madabhushi and Gurcan, {Metin N.}",
booktitle = "Medical Imaging 2016",

}

TY - GEN

T1 - A four class model for digital breast histopathology using high-definition Fourier transform infrared (FT-IR) spectroscopic imaging

AU - Mittal, Shachi

AU - Wrobel, Tomasz P.

AU - Leslie, L. S.

AU - Kadjacsy-Balla, Andre

AU - Bhargava, Rohit

PY - 2016/1/1

Y1 - 2016/1/1

N2 - High-definition (HD) Fourier transform infrared (FT-IR) spectroscopic imaging is an emerging technique that not only enables chemistry-based visualization of tissue constituents, and label free extraction of biochemical information but its higher spatial detail makes it a potentially useful platform to conduct digital pathology. This methodology, along with fast and efficient data analysis, can enable both quantitative and automated pathology. Here we demonstrate a combination of HD FT-IR spectroscopic imaging of breast tissue microarrays (TMAs) with data analysis algorithms to perform histologic analysis. The samples comprise four tissue states, namely hyperplasia, dysplasia, cancerous and normal. We identify various cell types which would act as biomarkers for breast cancer detection and differentiate between them using statistical pattern recognition tools i.e. Random Forest (RF) and Bayesian algorithms. Feature optimization is integrally carried out for the RF algorithm, reducing computation time as well as redundant spectral features. We achieved an order of magnitude reduction in the number of features with comparable prediction accuracy to that of the original feature set. Together, the demonstration of histology and selection of features paves the way for future applications in more complex models and rapid data acquisition.

AB - High-definition (HD) Fourier transform infrared (FT-IR) spectroscopic imaging is an emerging technique that not only enables chemistry-based visualization of tissue constituents, and label free extraction of biochemical information but its higher spatial detail makes it a potentially useful platform to conduct digital pathology. This methodology, along with fast and efficient data analysis, can enable both quantitative and automated pathology. Here we demonstrate a combination of HD FT-IR spectroscopic imaging of breast tissue microarrays (TMAs) with data analysis algorithms to perform histologic analysis. The samples comprise four tissue states, namely hyperplasia, dysplasia, cancerous and normal. We identify various cell types which would act as biomarkers for breast cancer detection and differentiate between them using statistical pattern recognition tools i.e. Random Forest (RF) and Bayesian algorithms. Feature optimization is integrally carried out for the RF algorithm, reducing computation time as well as redundant spectral features. We achieved an order of magnitude reduction in the number of features with comparable prediction accuracy to that of the original feature set. Together, the demonstration of histology and selection of features paves the way for future applications in more complex models and rapid data acquisition.

KW - FTIR

KW - High-Definition

KW - Random Forest

KW - automated pathology

KW - breast cancer

KW - four class classifier

UR - http://www.scopus.com/inward/record.url?scp=84989954093&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84989954093&partnerID=8YFLogxK

U2 - 10.1117/12.2217358

DO - 10.1117/12.2217358

M3 - Conference contribution

AN - SCOPUS:84989954093

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2016

A2 - Madabhushi, Anant

A2 - Gurcan, Metin N.

PB - SPIE

ER -