No abstract

Justification Automatic brain tumor classification by MRS has been under development for more than a decade. Nonetheless, to our knowledge, there are no published evaluations of predictive models with unseen cases that are subsequently acquired in different centers. The multicenter eTUMOUR project (2004)(2005)(2006)(2007)(2008)(2009), which builds upon previous expertise from the INTERPRET project (2000INTERPRET project ( -2002 has allowed such an evaluation to take place. Materials and Methods A total of 253 pairwise classifiers for glioblastoma, meningioma, metastasis, and low-grade glial diagnosis were inferred based on 211 SV short TE INTERPRET MR spectra obtained at 1.5 T (PRESS or STEAM, 20-32 ms) and automatically pre-processed. Afterwards, the classifiers were tested with 97 spectra, which were subsequently compiled during eTUMOUR. ResultsIn our results based on subsequently acquired spectra, accuracies of around 90% were achieved for most of the pairwise discrimination problems. The exception was for the glioblastoma versus metastasis discrimination, which was below 78%. A more clear definition of metastases may be obtained by other approaches, such as MRSI + MRI. Conclusions The prediction of the tumor type of in-vivo MRS is possible using classifiers developed from previously acquired data, in different hospitals with different instrumentation under the same acquisition protocols. This methodology may find application for assisting in the diagnosis of new brain tumor cases and for the quality control of multicenter MRS databases.

No abstract

This paper proposes the use of least-squares support vector machines (LS-SVMs) as a relatively new nonlinear multivariate calibration method, capable of dealing with ill-posed problems. LS-SVMs are an extension of "traditional" SVMs that have been introduced recently in the field of chemistry and chemometrics. The advantages of SVM-based methods over many other methods are that these lead to global models that are often unique, and nonlinear regression can be performed easily as an extension to linear regression. An additional advantage of LS-SVM (compared to SVM) is that model calculation and optimization can be performed relatively fast. As a test case to study the use of LS-SVM, the well-known and important chemical problem is considered in which spectra are affected by nonlinear interferences. As one specific example, a commonly used case is studied in which near-infrared spectra are affected by temperatureinduced spectral variation. Using this test case, model optimization, pruning, and model interpretation of the LS-SVM have been demonstrated. Furthermore, excellent performance of the LS-SVM, compared to other approaches, has been presented on the specific example. Therefore, it can be concluded that LS-SVMs can be seen as very promising techniques to solve ill-posed problems. Furthermore, these have been shown to lead to robust models in cases of spectral variations due to nonlinear interferences.