Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections

Garrett Winkelmaier, Mina Khoshdeli, Qingsu Cheng, Alexander D Borowsky, Bahram Parvin

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

Abstract

Quantum Cascade Laser InfRared (QCL-IR) microscopy has the potential to emerge as a unique modality for the diagnostics of histology sections. A pilot experiment is designed to evaluate whether benign and malignant breast histology sections can be differentiated using chemical profiling and without the use of spatial information. The experiment consists of 22 independent samples that were randomly selected from paraffin-embedded blocks with an equal number of benign and malignant labels. Spectral data are normalized and then used for both visualization and classification. Visualization is based on consensus clustering of the spectral signature within each group of the benign or malignant phenotype. Classification has been evaluated with three methods of tree-based, convolutional neural networks, and an encoder module for compression followed by softmax classification. The latter has the best performance with using only 20% of the data for training to arrive at the classification accuracy of 100%. Direct analysis of the spectral signatures indicates that both malignant and benign samples express the similar spectral peaks; however, peaks corresponding to several nucleic acid sequences and protein groups are overexpressed in malignant tissues.

Original languageEnglish (US)
Title of host publication2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
PublisherIEEE Computer Society
Pages831-834
Number of pages4
Volume2018-April
ISBN (Electronic)9781538636367
DOIs
StatePublished - May 23 2018
Event15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, United States
Duration: Apr 4 2018Apr 7 2018

Other

Other15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
CountryUnited States
CityWashington
Period4/4/184/7/18

Fingerprint

Semiconductor Lasers
Quantum cascade lasers
Histology
Confocal Microscopy
Microscopic examination
Breast
Infrared radiation
Phenotype
Visualization
Nucleic acid sequences
Paraffin
Paraffins
Nucleic Acids
Cluster Analysis
Labels
Experiments
Tissue
Neural networks
Proteins

Keywords

  • Breast cancer
  • Machine learning
  • QCL-IR
  • Spectral pathology

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Winkelmaier, G., Khoshdeli, M., Cheng, Q., Borowsky, A. D., & Parvin, B. (2018). Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections. In 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018 (Vol. 2018-April, pp. 831-834). IEEE Computer Society. https://doi.org/10.1109/ISBI.2018.8363700

Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections. / Winkelmaier, Garrett; Khoshdeli, Mina; Cheng, Qingsu; Borowsky, Alexander D; Parvin, Bahram.

2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Vol. 2018-April IEEE Computer Society, 2018. p. 831-834.

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

Winkelmaier, G, Khoshdeli, M, Cheng, Q, Borowsky, AD & Parvin, B 2018, Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections. in 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. vol. 2018-April, IEEE Computer Society, pp. 831-834, 15th IEEE International Symposium on Biomedical Imaging, ISBI 2018, Washington, United States, 4/4/18. https://doi.org/10.1109/ISBI.2018.8363700
Winkelmaier G, Khoshdeli M, Cheng Q, Borowsky AD, Parvin B. Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections. In 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Vol. 2018-April. IEEE Computer Society. 2018. p. 831-834 https://doi.org/10.1109/ISBI.2018.8363700
Winkelmaier, Garrett ; Khoshdeli, Mina ; Cheng, Qingsu ; Borowsky, Alexander D ; Parvin, Bahram. / Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections. 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018. Vol. 2018-April IEEE Computer Society, 2018. pp. 831-834
@inproceedings{26d6bc817db24736b4ad865079aa121e,
title = "Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections",
abstract = "Quantum Cascade Laser InfRared (QCL-IR) microscopy has the potential to emerge as a unique modality for the diagnostics of histology sections. A pilot experiment is designed to evaluate whether benign and malignant breast histology sections can be differentiated using chemical profiling and without the use of spatial information. The experiment consists of 22 independent samples that were randomly selected from paraffin-embedded blocks with an equal number of benign and malignant labels. Spectral data are normalized and then used for both visualization and classification. Visualization is based on consensus clustering of the spectral signature within each group of the benign or malignant phenotype. Classification has been evaluated with three methods of tree-based, convolutional neural networks, and an encoder module for compression followed by softmax classification. The latter has the best performance with using only 20{\%} of the data for training to arrive at the classification accuracy of 100{\%}. Direct analysis of the spectral signatures indicates that both malignant and benign samples express the similar spectral peaks; however, peaks corresponding to several nucleic acid sequences and protein groups are overexpressed in malignant tissues.",
keywords = "Breast cancer, Machine learning, QCL-IR, Spectral pathology",
author = "Garrett Winkelmaier and Mina Khoshdeli and Qingsu Cheng and Borowsky, {Alexander D} and Bahram Parvin",
year = "2018",
month = "5",
day = "23",
doi = "10.1109/ISBI.2018.8363700",
language = "English (US)",
volume = "2018-April",
pages = "831--834",
booktitle = "2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Quantum cascade laser infrared microscopy differentiates malignant phenotypes in breast histology sections

AU - Winkelmaier, Garrett

AU - Khoshdeli, Mina

AU - Cheng, Qingsu

AU - Borowsky, Alexander D

AU - Parvin, Bahram

PY - 2018/5/23

Y1 - 2018/5/23

N2 - Quantum Cascade Laser InfRared (QCL-IR) microscopy has the potential to emerge as a unique modality for the diagnostics of histology sections. A pilot experiment is designed to evaluate whether benign and malignant breast histology sections can be differentiated using chemical profiling and without the use of spatial information. The experiment consists of 22 independent samples that were randomly selected from paraffin-embedded blocks with an equal number of benign and malignant labels. Spectral data are normalized and then used for both visualization and classification. Visualization is based on consensus clustering of the spectral signature within each group of the benign or malignant phenotype. Classification has been evaluated with three methods of tree-based, convolutional neural networks, and an encoder module for compression followed by softmax classification. The latter has the best performance with using only 20% of the data for training to arrive at the classification accuracy of 100%. Direct analysis of the spectral signatures indicates that both malignant and benign samples express the similar spectral peaks; however, peaks corresponding to several nucleic acid sequences and protein groups are overexpressed in malignant tissues.

AB - Quantum Cascade Laser InfRared (QCL-IR) microscopy has the potential to emerge as a unique modality for the diagnostics of histology sections. A pilot experiment is designed to evaluate whether benign and malignant breast histology sections can be differentiated using chemical profiling and without the use of spatial information. The experiment consists of 22 independent samples that were randomly selected from paraffin-embedded blocks with an equal number of benign and malignant labels. Spectral data are normalized and then used for both visualization and classification. Visualization is based on consensus clustering of the spectral signature within each group of the benign or malignant phenotype. Classification has been evaluated with three methods of tree-based, convolutional neural networks, and an encoder module for compression followed by softmax classification. The latter has the best performance with using only 20% of the data for training to arrive at the classification accuracy of 100%. Direct analysis of the spectral signatures indicates that both malignant and benign samples express the similar spectral peaks; however, peaks corresponding to several nucleic acid sequences and protein groups are overexpressed in malignant tissues.

KW - Breast cancer

KW - Machine learning

KW - QCL-IR

KW - Spectral pathology

UR - http://www.scopus.com/inward/record.url?scp=85048101441&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048101441&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2018.8363700

DO - 10.1109/ISBI.2018.8363700

M3 - Conference contribution

VL - 2018-April

SP - 831

EP - 834

BT - 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018

PB - IEEE Computer Society

ER -