Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients

Himar Fabelo, Martin Halicek, Samuel Ortega, Adam Szolna, Jesus Morera, Roberto Sarmiento, Gustavo M. Callico, Baowei Fei

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

1 Citation (Scopus)

Abstract

Brain cancer surgery has the goal of performing an accurate resection of the tumor and preserving as much as possible the quality of life of the patient. There is a clinical need to develop non-invasive techniques that can provide reliable assistance for tumor resection in real-time during surgical procedures. Hyperspectral imaging (HSI) arises as a new, noninvasive and non-ionizing technique that can assist neurosurgeons during this difficult task. In this paper, we explore the use of deep learning (DL) techniques for processing hyperspectral (HS) images of in-vivo human brain tissue. We developed a surgical aid visualization system capable of offering guidance to the operating surgeon to achieve a successful and accurate tumor resection. The employed HS database is composed of 26 in-vivo hypercubes from 16 different human patients, among which 258,810 labelled pixels were used for evaluation. The proposed DL methods achieve an overall accuracy of 95% and 85% for binary and multiclass classifications, respectively. The proposed visualization system is able to generate a classification map that is formed by the combination of the DL map and an unsupervised clustering via a majority voting algorithm. This map can be adjusted by the operating surgeon to find the suitable configuration for the current situation during the surgical procedure.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2019
Subtitle of host publicationImage-Guided Procedures, Robotic Interventions, and Modeling
EditorsBaowei Fei, Cristian A. Linte
PublisherSPIE
ISBN (Electronic)9781510625495
DOIs
StatePublished - Jan 1 2019
EventMedical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling - San Diego, United States
Duration: Feb 17 2019Feb 19 2019

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10951
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling
CountryUnited States
CitySan Diego
Period2/17/192/19/19

Fingerprint

Glioblastoma
learning
Tumors
surgeons
tumors
Visualization
Learning
brain
Brain
voting
Neoplasms
Politics
surgery
Brain Neoplasms
Surgery
preserving
Cluster Analysis
image processing
Pixels
cancer

Keywords

  • Brain tumor
  • Cancer surgery
  • Classifier
  • Convolutional neural network (CNN)
  • Deep learning
  • Hyperspectral imaging
  • Intraoperative imaging
  • Supervised classification

ASJC Scopus subject areas

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

Cite this

Fabelo, H., Halicek, M., Ortega, S., Szolna, A., Morera, J., Sarmiento, R., ... Fei, B. (2019). Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients. In B. Fei, & C. A. Linte (Eds.), Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling [1095110] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10951). SPIE. https://doi.org/10.1117/12.2512569

Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients. / Fabelo, Himar; Halicek, Martin; Ortega, Samuel; Szolna, Adam; Morera, Jesus; Sarmiento, Roberto; Callico, Gustavo M.; Fei, Baowei.

Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling. ed. / Baowei Fei; Cristian A. Linte. SPIE, 2019. 1095110 (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 10951).

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

Fabelo, H, Halicek, M, Ortega, S, Szolna, A, Morera, J, Sarmiento, R, Callico, GM & Fei, B 2019, Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients. in B Fei & CA Linte (eds), Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling., 1095110, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 10951, SPIE, Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling, San Diego, United States, 2/17/19. https://doi.org/10.1117/12.2512569
Fabelo H, Halicek M, Ortega S, Szolna A, Morera J, Sarmiento R et al. Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients. In Fei B, Linte CA, editors, Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling. SPIE. 2019. 1095110. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.2512569
Fabelo, Himar ; Halicek, Martin ; Ortega, Samuel ; Szolna, Adam ; Morera, Jesus ; Sarmiento, Roberto ; Callico, Gustavo M. ; Fei, Baowei. / Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients. Medical Imaging 2019: Image-Guided Procedures, Robotic Interventions, and Modeling. editor / Baowei Fei ; Cristian A. Linte. SPIE, 2019. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{2bd136c244e5449baca32066afcf2bfa,
title = "Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients",
abstract = "Brain cancer surgery has the goal of performing an accurate resection of the tumor and preserving as much as possible the quality of life of the patient. There is a clinical need to develop non-invasive techniques that can provide reliable assistance for tumor resection in real-time during surgical procedures. Hyperspectral imaging (HSI) arises as a new, noninvasive and non-ionizing technique that can assist neurosurgeons during this difficult task. In this paper, we explore the use of deep learning (DL) techniques for processing hyperspectral (HS) images of in-vivo human brain tissue. We developed a surgical aid visualization system capable of offering guidance to the operating surgeon to achieve a successful and accurate tumor resection. The employed HS database is composed of 26 in-vivo hypercubes from 16 different human patients, among which 258,810 labelled pixels were used for evaluation. The proposed DL methods achieve an overall accuracy of 95{\%} and 85{\%} for binary and multiclass classifications, respectively. The proposed visualization system is able to generate a classification map that is formed by the combination of the DL map and an unsupervised clustering via a majority voting algorithm. This map can be adjusted by the operating surgeon to find the suitable configuration for the current situation during the surgical procedure.",
keywords = "Brain tumor, Cancer surgery, Classifier, Convolutional neural network (CNN), Deep learning, Hyperspectral imaging, Intraoperative imaging, Supervised classification",
author = "Himar Fabelo and Martin Halicek and Samuel Ortega and Adam Szolna and Jesus Morera and Roberto Sarmiento and Callico, {Gustavo M.} and Baowei Fei",
year = "2019",
month = "1",
day = "1",
doi = "10.1117/12.2512569",
language = "English (US)",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",
editor = "Baowei Fei and Linte, {Cristian A.}",
booktitle = "Medical Imaging 2019",

}

TY - GEN

T1 - Surgical aid visualization system for glioblastoma tumor identification based on deep learning and in-vivo hyperspectral images of human patients

AU - Fabelo, Himar

AU - Halicek, Martin

AU - Ortega, Samuel

AU - Szolna, Adam

AU - Morera, Jesus

AU - Sarmiento, Roberto

AU - Callico, Gustavo M.

AU - Fei, Baowei

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Brain cancer surgery has the goal of performing an accurate resection of the tumor and preserving as much as possible the quality of life of the patient. There is a clinical need to develop non-invasive techniques that can provide reliable assistance for tumor resection in real-time during surgical procedures. Hyperspectral imaging (HSI) arises as a new, noninvasive and non-ionizing technique that can assist neurosurgeons during this difficult task. In this paper, we explore the use of deep learning (DL) techniques for processing hyperspectral (HS) images of in-vivo human brain tissue. We developed a surgical aid visualization system capable of offering guidance to the operating surgeon to achieve a successful and accurate tumor resection. The employed HS database is composed of 26 in-vivo hypercubes from 16 different human patients, among which 258,810 labelled pixels were used for evaluation. The proposed DL methods achieve an overall accuracy of 95% and 85% for binary and multiclass classifications, respectively. The proposed visualization system is able to generate a classification map that is formed by the combination of the DL map and an unsupervised clustering via a majority voting algorithm. This map can be adjusted by the operating surgeon to find the suitable configuration for the current situation during the surgical procedure.

AB - Brain cancer surgery has the goal of performing an accurate resection of the tumor and preserving as much as possible the quality of life of the patient. There is a clinical need to develop non-invasive techniques that can provide reliable assistance for tumor resection in real-time during surgical procedures. Hyperspectral imaging (HSI) arises as a new, noninvasive and non-ionizing technique that can assist neurosurgeons during this difficult task. In this paper, we explore the use of deep learning (DL) techniques for processing hyperspectral (HS) images of in-vivo human brain tissue. We developed a surgical aid visualization system capable of offering guidance to the operating surgeon to achieve a successful and accurate tumor resection. The employed HS database is composed of 26 in-vivo hypercubes from 16 different human patients, among which 258,810 labelled pixels were used for evaluation. The proposed DL methods achieve an overall accuracy of 95% and 85% for binary and multiclass classifications, respectively. The proposed visualization system is able to generate a classification map that is formed by the combination of the DL map and an unsupervised clustering via a majority voting algorithm. This map can be adjusted by the operating surgeon to find the suitable configuration for the current situation during the surgical procedure.

KW - Brain tumor

KW - Cancer surgery

KW - Classifier

KW - Convolutional neural network (CNN)

KW - Deep learning

KW - Hyperspectral imaging

KW - Intraoperative imaging

KW - Supervised classification

UR - http://www.scopus.com/inward/record.url?scp=85068937866&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068937866&partnerID=8YFLogxK

U2 - 10.1117/12.2512569

DO - 10.1117/12.2512569

M3 - Conference contribution

C2 - 31447494

AN - SCOPUS:85068937866

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

BT - Medical Imaging 2019

A2 - Fei, Baowei

A2 - Linte, Cristian A.

PB - SPIE

ER -