Prospects for theranostics in neurosurgical imaging

Empowering confocal laser endomicroscopy diagnostics via deep learning

Mohammadhassan Izadyyazdanabadi, Evgenii Belykh, Michael A. Mooney, Jennifer M. Eschbacher, Peter Nakaji, Yezhou Yang, Mark C. Preul

Research output: Contribution to journalReview article

5 Citations (Scopus)

Abstract

Confocal laser endomicroscopy (CLE) is an advanced optical fluorescence imaging technology that has potential to increase intraoperative precision, extend resection, and tailor surgery for malignant invasive brain tumors because of its subcellular dimension resolution. Despite its promising diagnostic potential, interpreting the gray tone fluorescence images can be difficult for untrained users. CLE images can be distorted by motion artifacts, fluorescence signals out of detector dynamic range, or may be obscured by red blood cells, and thus interpreted as nondiagnostic (ND). However, just a single CLE image with a detectable pathognomonic histological tissue signature can suffice for intraoperative diagnosis. Dealing with the abundance of images from CLE is not unlike sifting through a myriad of genes, proteins, or other structural or metabolic markers to find something of commonality or uniqueness in cancer that might indicate a potential treatment scheme or target. In this review, we provide a detailed description of bioinformatical analysis methodology of CLE images that begins to assist the neurosurgeon and pathologist to rapidly connect on-the-fly intraoperative imaging, pathology, and surgical observation into a conclusionary system within the concept of theranostics. We present an overview and discuss deep learning models for automatic detection of the diagnostic CLE images and discuss various training regimes and ensemble modeling effect on power of deep learning predictive models. Two major approaches reviewed in this paper include the models that can automatically classify CLE images into diagnostic/ND, glioma/nonglioma, tumor/injury/normal categories, and models that can localize histological features on the CLE images using weakly supervised methods. We also briefly review advances in the deep learning approaches used for CLE image analysis in other organs. Significant advances in speed and precision of automated diagnostic frame selection would augment the diagnostic potential of CLE, improve operative workflow, and integration into brain tumor surgery. Such technology and bioinformatics analytics lend themselves to improved precision, personalization, and theranostics in brain tumor treatment.

Original languageEnglish (US)
Article number240
JournalFrontiers in Oncology
Volume8
Issue numberJUL
DOIs
StatePublished - Jul 3 2018

Fingerprint

Lasers
Learning
Brain Neoplasms
Optical Imaging
Fluorescence
Theranostic Nanomedicine
Technology
Surgical Pathology
Workflow
Computational Biology
Glioma
Artifacts
Neoplasms
Erythrocytes
Observation
Wounds and Injuries
Therapeutics
Proteins

Keywords

  • Brain neoplasm
  • Brain tumor imaging
  • Cancer
  • Confocal laser endomicroscopy
  • Convolutional neural networks
  • Deep learning
  • Fluorescence
  • Theranostics

ASJC Scopus subject areas

  • Oncology
  • Cancer Research

Cite this

Prospects for theranostics in neurosurgical imaging : Empowering confocal laser endomicroscopy diagnostics via deep learning. / Izadyyazdanabadi, Mohammadhassan; Belykh, Evgenii; Mooney, Michael A.; Eschbacher, Jennifer M.; Nakaji, Peter; Yang, Yezhou; Preul, Mark C.

In: Frontiers in Oncology, Vol. 8, No. JUL, 240, 03.07.2018.

Research output: Contribution to journalReview article

Izadyyazdanabadi, Mohammadhassan ; Belykh, Evgenii ; Mooney, Michael A. ; Eschbacher, Jennifer M. ; Nakaji, Peter ; Yang, Yezhou ; Preul, Mark C. / Prospects for theranostics in neurosurgical imaging : Empowering confocal laser endomicroscopy diagnostics via deep learning. In: Frontiers in Oncology. 2018 ; Vol. 8, No. JUL.
@article{0a29215ea6714065ba27ba38164d3a93,
title = "Prospects for theranostics in neurosurgical imaging: Empowering confocal laser endomicroscopy diagnostics via deep learning",
abstract = "Confocal laser endomicroscopy (CLE) is an advanced optical fluorescence imaging technology that has potential to increase intraoperative precision, extend resection, and tailor surgery for malignant invasive brain tumors because of its subcellular dimension resolution. Despite its promising diagnostic potential, interpreting the gray tone fluorescence images can be difficult for untrained users. CLE images can be distorted by motion artifacts, fluorescence signals out of detector dynamic range, or may be obscured by red blood cells, and thus interpreted as nondiagnostic (ND). However, just a single CLE image with a detectable pathognomonic histological tissue signature can suffice for intraoperative diagnosis. Dealing with the abundance of images from CLE is not unlike sifting through a myriad of genes, proteins, or other structural or metabolic markers to find something of commonality or uniqueness in cancer that might indicate a potential treatment scheme or target. In this review, we provide a detailed description of bioinformatical analysis methodology of CLE images that begins to assist the neurosurgeon and pathologist to rapidly connect on-the-fly intraoperative imaging, pathology, and surgical observation into a conclusionary system within the concept of theranostics. We present an overview and discuss deep learning models for automatic detection of the diagnostic CLE images and discuss various training regimes and ensemble modeling effect on power of deep learning predictive models. Two major approaches reviewed in this paper include the models that can automatically classify CLE images into diagnostic/ND, glioma/nonglioma, tumor/injury/normal categories, and models that can localize histological features on the CLE images using weakly supervised methods. We also briefly review advances in the deep learning approaches used for CLE image analysis in other organs. Significant advances in speed and precision of automated diagnostic frame selection would augment the diagnostic potential of CLE, improve operative workflow, and integration into brain tumor surgery. Such technology and bioinformatics analytics lend themselves to improved precision, personalization, and theranostics in brain tumor treatment.",
keywords = "Brain neoplasm, Brain tumor imaging, Cancer, Confocal laser endomicroscopy, Convolutional neural networks, Deep learning, Fluorescence, Theranostics",
author = "Mohammadhassan Izadyyazdanabadi and Evgenii Belykh and Mooney, {Michael A.} and Eschbacher, {Jennifer M.} and Peter Nakaji and Yezhou Yang and Preul, {Mark C.}",
year = "2018",
month = "7",
day = "3",
doi = "10.3389/fonc.2018.00240",
language = "English (US)",
volume = "8",
journal = "Frontiers in Oncology",
issn = "2234-943X",
publisher = "Frontiers Media S. A.",
number = "JUL",

}

TY - JOUR

T1 - Prospects for theranostics in neurosurgical imaging

T2 - Empowering confocal laser endomicroscopy diagnostics via deep learning

AU - Izadyyazdanabadi, Mohammadhassan

AU - Belykh, Evgenii

AU - Mooney, Michael A.

AU - Eschbacher, Jennifer M.

AU - Nakaji, Peter

AU - Yang, Yezhou

AU - Preul, Mark C.

PY - 2018/7/3

Y1 - 2018/7/3

N2 - Confocal laser endomicroscopy (CLE) is an advanced optical fluorescence imaging technology that has potential to increase intraoperative precision, extend resection, and tailor surgery for malignant invasive brain tumors because of its subcellular dimension resolution. Despite its promising diagnostic potential, interpreting the gray tone fluorescence images can be difficult for untrained users. CLE images can be distorted by motion artifacts, fluorescence signals out of detector dynamic range, or may be obscured by red blood cells, and thus interpreted as nondiagnostic (ND). However, just a single CLE image with a detectable pathognomonic histological tissue signature can suffice for intraoperative diagnosis. Dealing with the abundance of images from CLE is not unlike sifting through a myriad of genes, proteins, or other structural or metabolic markers to find something of commonality or uniqueness in cancer that might indicate a potential treatment scheme or target. In this review, we provide a detailed description of bioinformatical analysis methodology of CLE images that begins to assist the neurosurgeon and pathologist to rapidly connect on-the-fly intraoperative imaging, pathology, and surgical observation into a conclusionary system within the concept of theranostics. We present an overview and discuss deep learning models for automatic detection of the diagnostic CLE images and discuss various training regimes and ensemble modeling effect on power of deep learning predictive models. Two major approaches reviewed in this paper include the models that can automatically classify CLE images into diagnostic/ND, glioma/nonglioma, tumor/injury/normal categories, and models that can localize histological features on the CLE images using weakly supervised methods. We also briefly review advances in the deep learning approaches used for CLE image analysis in other organs. Significant advances in speed and precision of automated diagnostic frame selection would augment the diagnostic potential of CLE, improve operative workflow, and integration into brain tumor surgery. Such technology and bioinformatics analytics lend themselves to improved precision, personalization, and theranostics in brain tumor treatment.

AB - Confocal laser endomicroscopy (CLE) is an advanced optical fluorescence imaging technology that has potential to increase intraoperative precision, extend resection, and tailor surgery for malignant invasive brain tumors because of its subcellular dimension resolution. Despite its promising diagnostic potential, interpreting the gray tone fluorescence images can be difficult for untrained users. CLE images can be distorted by motion artifacts, fluorescence signals out of detector dynamic range, or may be obscured by red blood cells, and thus interpreted as nondiagnostic (ND). However, just a single CLE image with a detectable pathognomonic histological tissue signature can suffice for intraoperative diagnosis. Dealing with the abundance of images from CLE is not unlike sifting through a myriad of genes, proteins, or other structural or metabolic markers to find something of commonality or uniqueness in cancer that might indicate a potential treatment scheme or target. In this review, we provide a detailed description of bioinformatical analysis methodology of CLE images that begins to assist the neurosurgeon and pathologist to rapidly connect on-the-fly intraoperative imaging, pathology, and surgical observation into a conclusionary system within the concept of theranostics. We present an overview and discuss deep learning models for automatic detection of the diagnostic CLE images and discuss various training regimes and ensemble modeling effect on power of deep learning predictive models. Two major approaches reviewed in this paper include the models that can automatically classify CLE images into diagnostic/ND, glioma/nonglioma, tumor/injury/normal categories, and models that can localize histological features on the CLE images using weakly supervised methods. We also briefly review advances in the deep learning approaches used for CLE image analysis in other organs. Significant advances in speed and precision of automated diagnostic frame selection would augment the diagnostic potential of CLE, improve operative workflow, and integration into brain tumor surgery. Such technology and bioinformatics analytics lend themselves to improved precision, personalization, and theranostics in brain tumor treatment.

KW - Brain neoplasm

KW - Brain tumor imaging

KW - Cancer

KW - Confocal laser endomicroscopy

KW - Convolutional neural networks

KW - Deep learning

KW - Fluorescence

KW - Theranostics

UR - http://www.scopus.com/inward/record.url?scp=85049612589&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049612589&partnerID=8YFLogxK

U2 - 10.3389/fonc.2018.00240

DO - 10.3389/fonc.2018.00240

M3 - Review article

VL - 8

JO - Frontiers in Oncology

JF - Frontiers in Oncology

SN - 2234-943X

IS - JUL

M1 - 240

ER -