MRS Feature Extraction: Time-Frequency and Wavelet Analysis

Magnetic Resonance Spectroscopy (MRS) signals are being used for diagnosis of various brain diseases. Feature extraction of the MRS data is the most important step in analyzing the data. In this study a fully automated system is developed to analyze the MRS signals. The wavelet analysis is utilized in extracting signal features and the time-frequency representations. Consulting specialists in the field, the sensitivity of 87.11% plusmn 18 and the positive predictivity of 88.97% plusmn 15 in extracting features have been achieved.