AUTOMATED CLASSIFICATION OF HUMAN BRAIN-TUMORS BY NEURAL-NETWORK ANALYSIS USING IN-VIVO H-1 MAGNETIC-RESONANCE SPECTROSCOPIC METABOLITE PHENOTYPES

WE present a novel method to integrate in vivo nuclear magnetic resona nce spectroscopy (MRS) information into the clinical diagnosis of brai n tumours. Water-suppressed H-1 MRS data were collected from 33 patien ts with brain tumours and 28 healthy controls in omo. The data were tr eated in the time domain for removal of residual water and a region fr om the frequency domain (from 3.4 to 0.3 p.p.m.) together with the uns uppressed water signal were used as inputs for artificial neural netwo rk (ANN) analysis. The ANN distinguished tumour and normal tissue in e ach case and was able to classify benign and malignant gliomas as well as other brain tumours to match histology in a clinically useful mann er with an accuracy of 82%. Thus the present data indicate existence o f tumour tissue-specific metabolite phenotypes that can be detected by in vivo H-1 MRS. We believe that a user-independent ANN analysis may provide an alternative method for tumour classification in clinical pr actice.