Metabolite Quantification in MRS and Pattern Recognition

Poullet, Jean-Baptiste; Sima, Diana; Huffel, Sabine Van

doi:10.1002/9780470034590.emrstm1046

Cited by 2 publications

(2 citation statements)

References 95 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MRSI data processing was performed using SPID (Poullet, 2008; Poullet et al, 2008) in MatLab 2015a (MathWorks, Natick, MA, USA). AQSES-MRSI (Poullet et al, 2007; Sava et al, 2011) was used to quantify N -acetyl-aspartate, Choline (Cho), and Creatine (Cre), using a synthetic basis set.…”

Section: Methodsmentioning

confidence: 99%

Machine Learning Approach for Classifying Multiple Sclerosis Courses by Combining Clinical Data with Lesion Loads and Magnetic Resonance Metabolic Features

et al. 2017

Self Cite

View full text Add to dashboard Cite

Purpose: The purpose of this study is classifying multiple sclerosis (MS) patients in the four clinical forms as defined by the McDonald criteria using machine learning algorithms trained on clinical data combined with lesion loads and magnetic resonance metabolic features.Materials and Methods: Eighty-seven MS patients [12 Clinically Isolated Syndrome (CIS), 30 Relapse Remitting (RR), 17 Primary Progressive (PP), and 28 Secondary Progressive (SP)] and 18 healthy controls were included in this study. Longitudinal data available for each MS patient included clinical (e.g., age, disease duration, Expanded Disability Status Scale), conventional magnetic resonance imaging and spectroscopic imaging. We extract N-acetyl-aspartate (NAA), Choline (Cho), and Creatine (Cre) concentrations, and we compute three features for each spectroscopic grid by averaging metabolite ratios (NAA/Cho, NAA/Cre, Cho/Cre) over good quality voxels. We built linear mixed-effects models to test for statistically significant differences between MS forms. We test nine binary classification tasks on clinical data, lesion loads, and metabolic features, using a leave-one-patient-out cross-validation method based on 100 random patient-based bootstrap selections. We compute F1-scores and BAR values after tuning Linear Discriminant Analysis (LDA), Support Vector Machines with gaussian kernel (SVM-rbf), and Random Forests.Results: Statistically significant differences were found between the disease starting points of each MS form using four different response variables: Lesion Load, NAA/Cre, NAA/Cho, and Cho/Cre ratios. Training SVM-rbf on clinical and lesion loads yields F1-scores of 71–72% for CIS vs. RR and CIS vs. RR+SP, respectively. For RR vs. PP we obtained good classification results (maximum F1-score of 85%) after training LDA on clinical and metabolic features, while for RR vs. SP we obtained slightly higher classification results (maximum F1-score of 87%) after training LDA and SVM-rbf on clinical, lesion loads and metabolic features.Conclusions: Our results suggest that metabolic features are better at differentiating between relapsing-remitting and primary progressive forms, while lesion loads are better at differentiating between relapsing-remitting and secondary progressive forms. Therefore, combining clinical data with magnetic resonance lesion loads and metabolic features can improve the discrimination between relapsing-remitting and progressive forms.

show abstract

Section: Methodsmentioning

confidence: 99%

Machine Learning Approach for Classifying Multiple Sclerosis Courses by Combining Clinical Data with Lesion Loads and Magnetic Resonance Metabolic Features

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…The results demonstrate the biological relevance of MRS biomarkers in understanding the biology of the tumor, with potential in improving noninvasive diagnosis and in aiding clinicians in improving diagnosis and treatment and ongoing evaluation of the patients. Even if the diagnosis assignment is progressively improving by implementations of a variety of pattern recognition methodologies (35), still difficulties remain in the grading of glial tumor. This may be due to the fact that the methodologies proposed are not sensitive to the microheterogeneity of the tissue.…”

Section: Discussionmentioning

confidence: 99%

Ex vivo high resolution magic angle spinning metabolic profiles describe intratumoral histopathological tissue properties in adult human gliomas

Sava

Martínez-Bisbal

Huffel

et al. 2010

Magnetic Resonance in Med

View full text Add to dashboard Cite

In gliomas one can observe distinct histopathological tissue properties, such as viable tumor cells, necrotic tissue or regions where the tumor infiltrates normal brain. A first screening between the different intratumoral histopathological tissue properties would greatly assist in correctly diagnosing and prognosing gliomas. The potential of ex vivo high resolution magic angle spinning spectroscopy in characterizing these properties is analyzed and the biochemical differences between necrosis, high cellularity and border tumor regions in adult human gliomas are investigated. Statistical studies applied on sets of metabolite concentrations and metabolite ratios extracted from 52 high resolution magic angle spinning recordings coming from patients with different grades of glial tumors show a strong correlation between the histopathological tissue properties and the considered metabolic profiles, regardless of the malignancy grade. The results are in agreement with the pathology obtained by the histopathological examination that succeeded the high resolution magic angle spinning measurements. The metabolite concentration set can better differentiate between the considered histopathological tissue properties compared to the ratios. Representative reference tissue models describing the metabolic behavior are extracted for characterizing the intratumoral tissue properties. The proposed metabolic profiles reflect that the metabolites behavior is interconnected, and typical biochemical patterns emerge for each histopathological tissue property. Magn Reson Med 65:320-328,

show abstract

Metabolite Quantification in MRS and Pattern Recognition

Cited by 2 publications

References 95 publications

Machine Learning Approach for Classifying Multiple Sclerosis Courses by Combining Clinical Data with Lesion Loads and Magnetic Resonance Metabolic Features

Machine Learning Approach for Classifying Multiple Sclerosis Courses by Combining Clinical Data with Lesion Loads and Magnetic Resonance Metabolic Features

Ex vivo high resolution magic angle spinning metabolic profiles describe intratumoral histopathological tissue properties in adult human gliomas

Contact Info

Product

Resources

About