The T2-weighted half-Fourier acquired single-shot turbo-spin-echo technic compared to the conventional T2-weighted turbo-spin-echo technic for cerebral magnetic resonance tomography. A sequence comparison

To compare a T2-weighted half-fourier acquired single-shot turbo spin-echo (HF-TSE) sequence (HA-STE-sequence) for cerebral MRI with a standard T2-weighted fast spin-echo (TSE) sequence. Signal-to-noise (SNR) and contrast-to-noise ratios (CNR) for different cerebral structures, and intracerebral lesions as well as the detectability of intracerebral lesions depending on size and relaxation properties were evaluated on cranial MR examinations of 46 patients with both a TSE and a HF-TSE sequence. SNR and CNR were found to be significantly higher with the TSE sequence for all normal structures and lesions except CSF, and lesions with short relaxation time T2 (p < 0.001). The number of detected lesions larger than 10 mm was similar with both sequences. Thirty-six (TSE) and 34 (HF-TSE) hyperintense, and 7 (TSE) and 2 (HF-TSE) hypointense lesions of at least 5 mm but less than 10 mm in size were detected. Thirty-three (TSE) and 10 (HF-TSE) hyperintense, and 2 (TSE) and no (HF-TSE) hypointense lesions smaller than 5 mm were detected. Due to its short acquisition time, the HF-TSE sequence is an alternative for MR examinations of non-compliant or claustrophobic patients. The low SNR and CNR relative to the TSE-technique are limiting factors as to the detectability of small lesions or lesions with low contrast to surrounding structures, with the risk of an increasing number of false negative results in lesions with short T2 relaxation time smaller than 10 mm.