Premium
Automated classification of short echo time in in vivo 1 H brain tumor spectra: A multicenter study
Author(s) -
Tate A. Rosemary,
Majós Carles,
Moreno Angel,
Howe Franklyn A.,
Griffiths John R.,
Arús Carles
Publication year - 2003
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.10315
Subject(s) - echo time , echo (communications protocol) , spectral line , pattern recognition (psychology) , linear discriminant analysis , artificial intelligence , astrocytoma , computer science , categorization , set (abstract data type) , training set , nuclear medicine , medicine , radiology , magnetic resonance imaging , glioma , physics , computer network , cancer research , astronomy , programming language
Automated pattern recognition techniques are needed to help radiologists categorize MRS data of brain tumors according to histological type and grade. A major question is whether a computer program “trained” on spectra from one hospital will be able to classify those from another, particularly if the acquisition protocol is different. A subset of 144 histopathologically validated brain tumor spectra in the INTERPRET database, obtained from three of the collaborating centers, was grouped into meningiomas, low‐grade astrocytomas, and “aggressive tumors” (glioblastomas and metastases). Spectra from two centers formed the training set (94 spectra) while the third acted as the test set (50 spectra). Linear discriminant analysis successfully classified 48/50 in the test set; the remaining two were atypical cases. When the training and test sets were combined, 133 of the 144 spectra were correctly classified using the leave‐one‐out procedure. These spectra had been obtained using different sequences (STEAM and PRESS), different echo times (20, 30, 31, and 32 ms), different repetition times (1600 and 2000 ms), and different manufacturers' instruments (GE and Philips). Pattern recognition algorithms are less sensitive to acquisition parameters than had been expected. Magn Reson Med 49:29–36, 2003. © 2003 Wiley‐Liss, Inc.