scispace - formally typeset
Search or ask a question
Book Chapter•DOI•

A Case-Based Reasoning Approach to GBM Evolution

Ana Mendonça1, Joana Pereira1, Rita Reis1, Victor Alves1, António Abelha1, Filipa Ferraz1, João Neves, Jorge Ribeiro2, Henrique Vicente1, José Neves1 •
University of Minho1, Instituto Politécnico Nacional2
05 Sep 2018-pp 489-498
TL;DR: This study proposes a new Case Based Reasoning approach to problem solving that attempts to predict a patient’s GBM volume after five months of treatment based on features extracted from MR images and patient attributes such as age, gender, and type of treatment.
Abstract: GlioBastoma Multiforme (GBM) is an aggressive primary brain tumor characterized by a heterogeneous cell population that is genetically unstable and resistant to chemotherapy. Indeed, despite advances in medicine, patients diagnosed with GBM have a median survival of just one year. Magnetic Resonance Imaging (MRI) is the most widely used imaging technique for determining the location and size of brain tumors. Indisputably, this technique plays a major role in the diagnosis, treatment planning, and prognosis of GBM. Therefore, this study proposes a new Case Based Reasoning approach to problem solving that attempts to predict a patient’s GBM volume after five months of treatment based on features extracted from MR images and patient attributes such as age, gender, and type of treatment.

Summary (1 min read)

Jump to:  and [Summary]

Summary

  • GlioBastoma Multiforme (GBM) is an aggressive primary brain tumor characterized by a heterogeneous cell population that is genetically unstable and resistant to chemotherapy.
  • Indeed, despite advances in medicine, patients diagnosed with GBM have a median survival of just one year.
  • Magnetic Resonance Imaging (MRI) is the most widely used imaging technique for determining the location and size of brain tumors.
  • Indisputably, this technique plays a major role in the diagnosis, treatment planning, and prognosis of GBM.
  • Therefore, this study proposes a new Case Based Reasoning approach to problem solving that attempts to predict a patient’s GBM volume after five months of treatment based on features extracted from MR images and patient attributes such as age, gender, and type of treatment.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

A Case-Based Reasoning Approach to GBM
Evolution
Ana Mendonça
1
, Joana Pereira
1
, Rita Reis
1
, Victor Alves
2
,
António Abelha
2
, Filipa Ferraz
1,2
,João Neves
3
,
Jorge Ribeiro
4
, Henrique Vicente
2,5
, and José Neves
2(&)
1
Departamento de Informática, Escola de Engenharia, Universidade do Minho,
Braga, Portugal
{a70606,a73302,a71983}@alunos.uminho.pt,
filipatferraz@gmail.com
2
Centro Algoritmi, Universidade do Minho, Braga, Portugal
{valves,jneves}@di.uminho.pt
3
Mediclinic Arabian Ranches, PO Box 282602, Dubai, United Arab Emirates
joaocpneves@gmail.com
4
Escola Superior de Tecnologia e Gestão,
Instituto Politécnico de Viana do Castelo, Viana do Castelo, Portugal
jribeiro@estg.ipvc.pt
5
Departamento de Química, Escola de Ciências e Tecnologia,
Centro de Química de Évora, Universidade de Évora, Évora, Portugal
hvicente@uevora.pt
Abstract. GlioBastoma Multiforme (GBM) is an aggressive primary brain
tumor characterized by a heterogeneous cell population that is genetically
unstable and resistant to chemotherapy. Indeed, despite advances in medicine,
patients diagnosed with GBM have a median survival of just one year. Magnetic
Resonance Imaging (MRI) is the most widely used imaging technique for
determining the location and size of brain tumors. Indisputably, this technique
plays a major role in the diagnosis, treatment planning, and prognosis of GBM.
Therefore, this study proposes a new Case Based Reasoning approach to
problem solving that attempts to predict a patient’s GBM volume after five
months of treatment based on features extracted from MR images and patient
attributes such as age, gender, and type of treatment.
Keywords: Artificial Intelligence
GlioBlastoma Multiforme
Logic Programming
Knowledge Representation and Reasoning
Case Based Reasoning
1 Introduction
Brain tumors are defined as abnormal cells that tend to proliferate without control,
being possible to group the different types of brain tumors into two different classes,
i.e., primary and secondary tumors. The former one develops in the brain itself, while
secondary brain tumors spread through metastasis from other locations to the brain [1].
GlioBlastoma Multiforme (GBM) is most common – about 30% of primary brain
© Springer Nature Switzerland AG 2018
N. T. Nguyen et al. (Eds.): ICCCI 2018, LNAI 11056, pp. 489–498, 2018.
https://doi.org/10.1007/978-3-319-98446-9_46
Citations
More filters
Dissertation•
Formalisation de la démarche de conception d'un système de production mobile : intégration des concepts de mobilité et de reconfigurabilité

[...]

Youssef Benama
12 Feb 2016
TL;DR: In this paper, BenAM et al. propose a demarche d'analyse and de conception of a systeme de production mobile, which is based on the concept of mobilite peut.
Abstract: Dans cette these nous analysons dans quelle mesure le concept de mobilite peut etre pris en compte dans la demarche d'analyse et de conception de systemes de production. Notre apport vise a formaliser la demarche d'analyse et de conception de ce dernier, explicitant les decisions a prendre, les informations necessaires et les criteres de decision a mettre en place. Dans cet objectif, deux niveaux d'analyse ont ete distingues : un niveau local concernant un site de production et un niveau global comprenant un ensemble de sites.Le premier niveau local considere un seul site de production. A ce niveau nous avons propose une approche prenant en compte les caracteristiques du site de production. Dans notre contexte, le choix de la localisation geographique de production est impose par le client. De ce fait, la conception du systeme de production doit s'adapter a cette contrainte. D'un point de vue conception, quatre questions sont abordees : (1) dans quelle mesure le concept de mobilite peut etre integre dans une demarche de conception de systeme de production mobile ? (2) quelles caracteristiques de l'environnement de production doivent etre prises en compte ? (3) comment determiner ce qu'il faut produire sur site ou ce qu'il serait opportun d'externaliser ? et (4) compte tenu des informations obtenues quelle est la meilleure configuration du SPM a envisager et selon quels criteres de choix ? La reponse a ces questions conduit a la proposition d'une configuration du SPM adaptee pour un seul site de production.Le deuxieme niveau global traite la problematique de mobilite successive multi sites. En effet, pour etre rentabilise le systeme de production doit etre mobilise sur plusieurs sites de production. A chaque changement de site de production, une reconfiguration du systeme de production s'impose en se basant sur la configuration existante (version i-1). LaThese de Youssef BENAMAreconfigurabilite concerne d'une part l'architecture interne du systeme (choix des machines, recrutement de nouvelles equipes locales, etc) et d'autre part l'organisation de la chaine d'approvisionnement du SPM (faire en interne ou externaliser, fournisseur local, etc.). A ce niveau global d'analyse, nous proposons deux modeles d'analyse : (1) un premier modele pour l'analyse de la reconfigurabilite interne. Ce modele d'analyse permet d'adapter le nombre de lignes de production et le nombre d'equipes en fonction d'un scenario de demande (localisations geographiques, capacite necessaire par site). L'originalite de notre proposition consiste d'une part en l'evaluation des couts de reconfiguration necessaires et d'autre part l'appreciation du niveau d'adequation de la configuration proposee avec le contexte du site de production via l'utilisation de l'indicateur de mobilite. (2) Le deuxieme modele d'analyse concerne la reconfigurabilite de la chaine d'approvisionnement amont du SPM. Il consiste en une adaptation du modele d'aide a la decision "faire ou faire faire" par l'integration d'un cote de l'importance du site de production et d'un autre cote des specificites de chaque site de production.La demarche d'analyse proposee est illustree sur le cas industriel concernant la conception d'une usine mobile pour la fabrication et l'installation sur site de composants de centrales solaires thermodynamiques.

8 citations

References
More filters
Journal Article•DOI•
Brain tumor mri image classification with feature selection and extraction using linear discriminant analysis

[...]

V. P. Gladis Pushpa Rathi, S. Palani
31 Jul 2012
TL;DR: The problem of classification MRI brain images is addressed by creating a robust and more accurate classifier which can act as an expert assistant to medical practitioners by presenting a novel method of feature selection and extraction.
Abstract: Feature extraction is a method of capturing visual content of an image. The feature extraction is the process to represent raw image in its reduced form to facilitate decision making such as pattern classification. We have tried to address the problem of classification MRI brain images by creating a robust and more accurate classifier which can act as an expert assistant to medical practitioners. The objective of this paper is to present a novel method of feature selection and extraction. This approach combines the Intensity, Texture, shape based features and classifies the tumor as white matter, Gray matter, CSF, abnormal and normal area. The experiment is performed on 140 tumor contained brain MR images from the Internet Brain Segmentation Repository. The proposed technique has been carried out over a larger database as compare to any previous work and is more robust and effective. PCA and Linear Discriminant Analysis (LDA) were applied on the training sets. The Support Vector Machine (SVM) classifier served as a comparison of nonlinear techniques Vs linear ones. PCA and LDA methods are used to reduce the number of features used. The feature selection using the proposed technique is more beneficial as it analyses the data according to grouping class variable and gives reduced feature set with high classification accuracy.

55 citations

Proceedings Article•DOI•
Artificial neural networks in diabetes control

[...]

Filipe Fernandes1, Henrique Vicente2, António Abelha1, José Machado1, Paulo Novais1, José Neves1 •
University of Minho1, University of Évora2
28 Jul 2015
TL;DR: This work will focus on the development of a diagnosis support system, in terms of its knowledge representation and reasoning procedures, under a formal framework based on Logic Programming, complemented with an approach to computing centred on Artificial Neural Networks to evaluate the Diabetes states and the Degree ofConfidence that one has on such a happening.
Abstract: Diabetes Mellitus is now a prevalent disease in both developed and underdeveloped countries, being a major cause of morbidity and mortality. Overweight/obesity and hypertension are potentially modifiable risk factors for diabetes mellitus, and persist during the course of the disease. Despite the evidence from large controlled trials establishing the benefit of intensive diabetes management in reducing microvasculars and macrovasculars complications, high proportions of patients remain poorly controlled. Poor and inadequate glycemic control among patients with Type 2 diabetes constitutes a major public health problem and a risk factor for the development of diabetes complications. In clinical practice, optimal glycemic control is difficult to obtain on a long-term basis, once the reasons for feebly glycemic control are complex. Therefore, this work will focus on the development of a diagnosis support system, in terms of its knowledge representation and reasoning procedures, under a formal framework based on Logic Programming, complemented with an approach to computing centred on Artificial Neural Networks, to evaluate the Diabetes states and the Degree-of-Confidence that one has on such a happening.

52 citations

Journal Article•DOI•
Extracted magnetic resonance texture features discriminate between phenotypes and are associated with overall survival in glioblastoma multiforme patients

[...]

Ahmad Chaddad1, Camel Tanougast1•
University of Lorraine1
10 Mar 2016-Medical & Biological Engineering & Computing
TL;DR: Among 22 features examined, three texture features have the ability to predict overall survival for GBM patients demonstrating the utility of GLCM analyses in both the diagnosis and prognosis of this patient population.
Abstract: GBM is a markedly heterogeneous brain tumor consisting of three main volumetric phenotypes identifiable on magnetic resonance imaging: necrosis (vN), active tumor (vAT), and edema/invasion (vE). The goal of this study is to identify the three glioblastoma multiforme (GBM) phenotypes using a texture-based gray-level co-occurrence matrix (GLCM) approach and determine whether the texture features of phenotypes are related to patient survival. MR imaging data in 40 GBM patients were analyzed. Phenotypes vN, vAT, and vE were segmented in a preprocessing step using 3D Slicer for rigid registration by T1-weighted imaging and corresponding fluid attenuation inversion recovery images. The GBM phenotypes were segmented using 3D Slicer tools. Texture features were extracted from GLCM of GBM phenotypes. Thereafter, Kruskal–Wallis test was employed to select the significant features. Robust predictive GBM features were identified and underwent numerous classifier analyses to distinguish phenotypes. Kaplan–Meier analysis was also performed to determine the relationship, if any, between phenotype texture features and survival rate. The simulation results showed that the 22 texture features were significant with p value <0.05. GBM phenotype discrimination based on texture features showed the best accuracy, sensitivity, and specificity of 79.31, 91.67, and 98.75 %, respectively. Three texture features derived from active tumor parts: difference entropy, information measure of correlation, and inverse difference were statistically significant in the prediction of survival, with log-rank p values of 0.001, 0.001, and 0.008, respectively. Among 22 features examined, three texture features have the ability to predict overall survival for GBM patients demonstrating the utility of GLCM analyses in both the diagnosis and prognosis of this patient population.

44 citations

Journal Article•DOI•
A framework for multimodal imaging-based prognostic model building: Preliminary study on multimodal MRI in Glioblastoma Multiforme

[...]

Taman Upadhaya1, Taman Upadhaya2, Yannick Morvan, Eric Stindel2, P.-J. Le Reste, Mathieu Hatt2 •
University of Western Brittany1, French Institute of Health and Medical Research2
01 Nov 2015-Irbm
TL;DR: The results suggest that several textural features in each MR sequence have prognostic value in GBM.
Abstract: In Glioblastoma Multiforme (GBM) image-derived features (“radiomics”) could help in individualizing patient management. Simple geometric features of tumors (necrosis, edema, active tumor) and first-order statistics in Magnetic Resonance Imaging (MRI) are used in clinical practice. However, these features provide limited characterization power because they do not incorporate spatial information and thus cannot differentiate patterns. The aim of this work is to develop and evaluate a methodological framework dedicated to building a prognostic model based on heterogeneity textural features of multimodal MRI sequences (T1, T1-contrast, T2 and FLAIR) in GBM. The proposed workflow consists in i) registering the available 3D multimodal MR images and segmenting the tumor volume, ii) extracting image features such as heterogeneity metrics and iii) building a prognostic model by selecting, ranking and combining optimal features through machine learning (Support Vector Machine). This framework was applied to 40 histologically proven GBM patients with the endpoint being overall survival (OS) classified as above or below the median survival (15 months). The models combining features from a maximum of two modalities were evaluated using leave-one-out cross-validation (LOOCV). A classification accuracy of 90% (sensitivity 85%, specificity 95%) was obtained by combining features from T1 pre-contrast and T1 post-contrast sequences. Our results suggest that several textural features in each MR sequence have prognostic value in GBM.

22 citations

Journal Article•DOI•
Classifying Glioblastoma Multiforme Follow-Up Progressive vs. Responsive Forms Using Multi-Parametric MRI Features.

[...]

Adrian Ion-Margineanu1, Sofie Van Cauter1, Diana M. Sima1, Frederik Maes1, Stefan Sunaert1, Uwe Himmelreich1, Sabine Van Huffel1 •
Katholieke Universiteit Leuven1
11 Jan 2017-Frontiers in Neuroscience
TL;DR: The findings show that AdaBoost, SVM-rbf, and RUSBoost trained on T1pc and CBV features can differentiate progressive from responsive GBM patients with very high accuracy.
Abstract: The purpose of this paper is discriminating between tumour progression and response to treatment based on follow-up multi-parametric magnetic resonance imaging (MRI) data retrieved from glioblastoma multiforme (GBM) patients. Multi-parametric MRI data consisting of conventional MRI (cMRI) and advanced MRI (i.e.\@ perfusion weighted MRI (PWI) and diffusion kurtosis MRI (DKI)) were acquired from 29 GBM patients treated with adjuvant therapy after surgery. We propose an automatic pipeline for processing advanced MRI data and extracting intensity-based histogram features and 3-D texture features using manually and semi-manually delineated regions of interest (ROIs). Classifiers are trained using a leave-one-patient-out cross validation scheme on complete MRI data. Balanced accuracy rate (BAR) values are computed and compared between different ROIs, MR modalities, and classifiers, using non-parametric multiple comparison tests. Maximum BAR values using manual delineations are 0.956, 0.85, 0.879, and 0.932, for cMRI, PWI, DKI and all three MRI modalities combined, respectively. Maximum BAR values using semi-manual delineations are 0.932, 0.894, 0.885, and 0.947, for cMRI, PWI, DKI and all three MR modalities combined, respectively. After statistical testing using Kruskal-Wallis and post-hoc Dunn-{\v{S}}id{\'a}k analysis we conclude that training a RUSBoost classifier on features extracted using semi-manual delineations on cMRI or on all MRI modalities combined performs best. We present two main conclusions: (1) using T1 post-contrast (T1pc) features extracted from manual total delineations, AdaBoost achieves the highest BAR value, 0.956; (2) using T1pc-average, T1pc-90$^{th}$ percentile and Cerebral Blood Volume (CBV) 90$^{th}$ percentile extracted from semi-manually delineated contrast enhancing ROIs, SVM-rbf and RUSBoost achieve BAR values of 0.947 and 0.932, respectively. Our findings show that AdaBoost, SVM-rbf, and RUSBoost trained on T1pc and CBV features can differentiate progressive from responsive GBM patients with very high accuracy.

19 citations

Related Papers (5)
Improve Glioblastoma Multiforme Prognosis Prediction by Using Feature Selection and Multiple Kernel Learning

[...]

01 Sep 2016-IEEE/ACM Transactions on Computational Biology and Bioinformatics
Ya Zhang, Ao Li, Chen Peng, Minghui Wang 
Deep Learning of Imaging Phenotype and Genotype for Predicting Overall Survival Time of Glioblastoma Patients

[...]

06 Jan 2020-IEEE Transactions on Medical Imaging
Zhenyu Tang, Yuyun Xu, Lei Jin, Abudumijiti Aibaidula, Junfeng Lu, Zhicheng Jiao, Jinsong Wu, Han Zhang, Dinggang Shen 
A Novel MKL Method for GBM Prognosis Prediction by Integrating Histopathological Image and Multi-Omics Data

[...]

01 Jan 2020-IEEE Journal of Biomedical and Health Informatics
Ya Zhang, Ao Li, Jie He, Minghui Wang 
Predicting Clinical Outcomes in Glioblastoma: An Application of Topological and Functional Data Analysis

[...]

21 Nov 2016-arXiv: Applications
Lorin Crawford, Anthea Monod, Andrew X. Chen, Sayan Mukherjee, Raul Rabadan 
Constraint based temporal event sequence mining for Glioblastoma survival prediction

[...]

01 Jun 2016-Journal of Biomedical Informatics
Kunal Malhotra, Shamkant B. Navathe, Duen Horng Chau, Costas G. Hadjipanayis, Jimeng Sun