Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques

Luke Macyszyn, Hamed Akbari, Jared M. Pisapia, Xiao Da, Mark Attiah, Vadim Pigrish, Yingtao Bi, Sharmistha Pal, Ramana V. Davuluri, Laura Roccograndi, Nadia Dahmane, Maria Martinez-Lage, George Biros, Ronald L. Wolf, Michel Bilello, Donald M. O'Rourke, Christos Davatzikos

Research output: Contribution to journalArticle

Abstract

Background MRI characteristics of brain gliomas have been used to predict clinical outcome and molecular tumor characteristics. However, previously reported imaging biomarkers have not been sufficiently accurate or reproducible to enter routine clinical practice and often rely on relatively simple MRI measures. The current study leverages advanced image analysis and machine learning algorithms to identify complex and reproducible imaging patterns predictive of overall survival and molecular subtype in glioblastoma (GB). Methods One hundred five patients with GB were first used to extract approximately 60 diverse features from preoperative multiparametric MRIs. These imaging features were used by a machine learning algorithm to derive imaging predictors of patient survival and molecular subtype. Cross-validation ensured generalizability of these predictors to new patients. Subsequently, the predictors were evaluated in a prospective cohort of 29 new patients. Results Survival curves yielded a hazard ratio of 10.64 for predicted long versus short survivors. The overall, 3-way (long/medium/short survival) accuracy in the prospective cohort approached 80%. Classification of patients into the 4 molecular subtypes of GB achieved 76% accuracy. Conclusions By employing machine learning techniques, we were able to demonstrate that imaging patterns are highly predictive of patient survival. Additionally, we found that GB subtypes have distinctive imaging phenotypes. These results reveal that when imaging markers related to infiltration, cell density, microvascularity, and blood-brain barrier compromise are integrated via advanced pattern analysis methods, they form very accurate predictive biomarkers. These predictive markers used solely preoperative images, hence they can significantly augment diagnosis and treatment of GB patients.

Original languageEnglish (US)
Pages (from-to)417-425
Number of pages9
JournalNeuro-Oncology
Volume18
Issue number3
DOIs
StatePublished - Mar 1 2016
Externally publishedYes

Fingerprint

Glioblastoma
Survival
Biomarkers
Machine Learning
Blood-Brain Barrier
Glioma
Survivors
Cell Count
Phenotype
Brain
Neoplasms

Keywords

  • glioblastoma
  • imaging
  • machine learning
  • predict
  • survival

ASJC Scopus subject areas

  • Cancer Research
  • Oncology
  • Clinical Neurology

Cite this

Macyszyn, L., Akbari, H., Pisapia, J. M., Da, X., Attiah, M., Pigrish, V., ... Davatzikos, C. (2016). Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques. Neuro-Oncology, 18(3), 417-425. https://doi.org/10.1093/neuonc/nov127

Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques. / Macyszyn, Luke; Akbari, Hamed; Pisapia, Jared M.; Da, Xiao; Attiah, Mark; Pigrish, Vadim; Bi, Yingtao; Pal, Sharmistha; Davuluri, Ramana V.; Roccograndi, Laura; Dahmane, Nadia; Martinez-Lage, Maria; Biros, George; Wolf, Ronald L.; Bilello, Michel; O'Rourke, Donald M.; Davatzikos, Christos.

In: Neuro-Oncology, Vol. 18, No. 3, 01.03.2016, p. 417-425.

Research output: Contribution to journalArticle

Macyszyn, L, Akbari, H, Pisapia, JM, Da, X, Attiah, M, Pigrish, V, Bi, Y, Pal, S, Davuluri, RV, Roccograndi, L, Dahmane, N, Martinez-Lage, M, Biros, G, Wolf, RL, Bilello, M, O'Rourke, DM & Davatzikos, C 2016, 'Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques', Neuro-Oncology, vol. 18, no. 3, pp. 417-425. https://doi.org/10.1093/neuonc/nov127
Macyszyn, Luke ; Akbari, Hamed ; Pisapia, Jared M. ; Da, Xiao ; Attiah, Mark ; Pigrish, Vadim ; Bi, Yingtao ; Pal, Sharmistha ; Davuluri, Ramana V. ; Roccograndi, Laura ; Dahmane, Nadia ; Martinez-Lage, Maria ; Biros, George ; Wolf, Ronald L. ; Bilello, Michel ; O'Rourke, Donald M. ; Davatzikos, Christos. / Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques. In: Neuro-Oncology. 2016 ; Vol. 18, No. 3. pp. 417-425.
@article{448c7f553b0f401ea029adbce4ab0d97,
title = "Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques",
abstract = "Background MRI characteristics of brain gliomas have been used to predict clinical outcome and molecular tumor characteristics. However, previously reported imaging biomarkers have not been sufficiently accurate or reproducible to enter routine clinical practice and often rely on relatively simple MRI measures. The current study leverages advanced image analysis and machine learning algorithms to identify complex and reproducible imaging patterns predictive of overall survival and molecular subtype in glioblastoma (GB). Methods One hundred five patients with GB were first used to extract approximately 60 diverse features from preoperative multiparametric MRIs. These imaging features were used by a machine learning algorithm to derive imaging predictors of patient survival and molecular subtype. Cross-validation ensured generalizability of these predictors to new patients. Subsequently, the predictors were evaluated in a prospective cohort of 29 new patients. Results Survival curves yielded a hazard ratio of 10.64 for predicted long versus short survivors. The overall, 3-way (long/medium/short survival) accuracy in the prospective cohort approached 80{\%}. Classification of patients into the 4 molecular subtypes of GB achieved 76{\%} accuracy. Conclusions By employing machine learning techniques, we were able to demonstrate that imaging patterns are highly predictive of patient survival. Additionally, we found that GB subtypes have distinctive imaging phenotypes. These results reveal that when imaging markers related to infiltration, cell density, microvascularity, and blood-brain barrier compromise are integrated via advanced pattern analysis methods, they form very accurate predictive biomarkers. These predictive markers used solely preoperative images, hence they can significantly augment diagnosis and treatment of GB patients.",
keywords = "glioblastoma, imaging, machine learning, predict, survival",
author = "Luke Macyszyn and Hamed Akbari and Pisapia, {Jared M.} and Xiao Da and Mark Attiah and Vadim Pigrish and Yingtao Bi and Sharmistha Pal and Davuluri, {Ramana V.} and Laura Roccograndi and Nadia Dahmane and Maria Martinez-Lage and George Biros and Wolf, {Ronald L.} and Michel Bilello and O'Rourke, {Donald M.} and Christos Davatzikos",
year = "2016",
month = "3",
day = "1",
doi = "10.1093/neuonc/nov127",
language = "English (US)",
volume = "18",
pages = "417--425",
journal = "Neuro-Oncology",
issn = "1522-8517",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques

AU - Macyszyn, Luke

AU - Akbari, Hamed

AU - Pisapia, Jared M.

AU - Da, Xiao

AU - Attiah, Mark

AU - Pigrish, Vadim

AU - Bi, Yingtao

AU - Pal, Sharmistha

AU - Davuluri, Ramana V.

AU - Roccograndi, Laura

AU - Dahmane, Nadia

AU - Martinez-Lage, Maria

AU - Biros, George

AU - Wolf, Ronald L.

AU - Bilello, Michel

AU - O'Rourke, Donald M.

AU - Davatzikos, Christos

PY - 2016/3/1

Y1 - 2016/3/1

N2 - Background MRI characteristics of brain gliomas have been used to predict clinical outcome and molecular tumor characteristics. However, previously reported imaging biomarkers have not been sufficiently accurate or reproducible to enter routine clinical practice and often rely on relatively simple MRI measures. The current study leverages advanced image analysis and machine learning algorithms to identify complex and reproducible imaging patterns predictive of overall survival and molecular subtype in glioblastoma (GB). Methods One hundred five patients with GB were first used to extract approximately 60 diverse features from preoperative multiparametric MRIs. These imaging features were used by a machine learning algorithm to derive imaging predictors of patient survival and molecular subtype. Cross-validation ensured generalizability of these predictors to new patients. Subsequently, the predictors were evaluated in a prospective cohort of 29 new patients. Results Survival curves yielded a hazard ratio of 10.64 for predicted long versus short survivors. The overall, 3-way (long/medium/short survival) accuracy in the prospective cohort approached 80%. Classification of patients into the 4 molecular subtypes of GB achieved 76% accuracy. Conclusions By employing machine learning techniques, we were able to demonstrate that imaging patterns are highly predictive of patient survival. Additionally, we found that GB subtypes have distinctive imaging phenotypes. These results reveal that when imaging markers related to infiltration, cell density, microvascularity, and blood-brain barrier compromise are integrated via advanced pattern analysis methods, they form very accurate predictive biomarkers. These predictive markers used solely preoperative images, hence they can significantly augment diagnosis and treatment of GB patients.

AB - Background MRI characteristics of brain gliomas have been used to predict clinical outcome and molecular tumor characteristics. However, previously reported imaging biomarkers have not been sufficiently accurate or reproducible to enter routine clinical practice and often rely on relatively simple MRI measures. The current study leverages advanced image analysis and machine learning algorithms to identify complex and reproducible imaging patterns predictive of overall survival and molecular subtype in glioblastoma (GB). Methods One hundred five patients with GB were first used to extract approximately 60 diverse features from preoperative multiparametric MRIs. These imaging features were used by a machine learning algorithm to derive imaging predictors of patient survival and molecular subtype. Cross-validation ensured generalizability of these predictors to new patients. Subsequently, the predictors were evaluated in a prospective cohort of 29 new patients. Results Survival curves yielded a hazard ratio of 10.64 for predicted long versus short survivors. The overall, 3-way (long/medium/short survival) accuracy in the prospective cohort approached 80%. Classification of patients into the 4 molecular subtypes of GB achieved 76% accuracy. Conclusions By employing machine learning techniques, we were able to demonstrate that imaging patterns are highly predictive of patient survival. Additionally, we found that GB subtypes have distinctive imaging phenotypes. These results reveal that when imaging markers related to infiltration, cell density, microvascularity, and blood-brain barrier compromise are integrated via advanced pattern analysis methods, they form very accurate predictive biomarkers. These predictive markers used solely preoperative images, hence they can significantly augment diagnosis and treatment of GB patients.

KW - glioblastoma

KW - imaging

KW - machine learning

KW - predict

KW - survival

UR - http://www.scopus.com/inward/record.url?scp=84960112384&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84960112384&partnerID=8YFLogxK

U2 - 10.1093/neuonc/nov127

DO - 10.1093/neuonc/nov127

M3 - Article

VL - 18

SP - 417

EP - 425

JO - Neuro-Oncology

JF - Neuro-Oncology

SN - 1522-8517

IS - 3

ER -