Quantitative assessment of auditory cortex responses induced by imager acoustic noise.

A clustered volume acquisition functional MRI pulse sequence was modified to assess the response to the acoustic noise of echo-planar imaging in the auditory cortex and to determine whether it is possible to obtain data which is relatively free of acoustic contamination. The spatial location and strength (percent signal change) of cortical responses to the imager noise were examined by introducing extra gradient readouts, without slice excitation, to provide acoustic stimulation immediately prior to acquisition of a cerebral volume. The duration of acoustic stimulation was controlled by varying the number of extra gradient readouts. Slice acquisitions were clustered at the end of the repetition time (TR) period to prevent a response from being induced by the volume acquisition itself ("Intra-Acquisition Response"). The cerebral volumes were acquired using a long TR in order to limit the integration of the cortical response across volume acquisitions ("Inter-Acquisition Response"). Cortical responses were observed to be largest and most significant on the medial two-thirds of Heschl's gyrus, the location of primary auditory cortex. Mean signal changes induced by the imager noise were observed to be as high as 0.95%. A 2 sec delay prior to onset of the BOLD response was empirically determined. These results demonstrate that clustered volume acquisitions may be utilized for up to 2 sec of volume acquisition without inducing an appreciable Intra-Acquisition Response and can be used, with a sufficiently long TR, to provide data which are similarly free of any Inter-Acquisition Response.

Improved auditory cortex imaging using clustered volume acquisitions.

The effects of the noise of echo-planar functional magnetic resonance imaging on auditory cortex responses were compared for two methods of acquiring functional MR data. Responses observed with a distributed volume acquisition sequence were compared to those obtained with a clustered volume acquisition sequence. In the former case, slices from the volume were acquired at equal intervals within the repetition time, whereas the latter acquired all slices in rapid succession at the end of the imaging period. The clustered volume acquisition provides a period of quiet during which a stimulus may be presented uninterrupted and uncontaminated by the noise of echo-planar imaging. Both sequences were implemented on a General Electric Signa imager retrofitted for echo-planar imaging by Advanced NMR Systems, Inc. The sequences were used to acquire 60 images per slice of a fixed volume of cerebral cortex while subjects were presented an instrumental music stimulus in an On vs. Off paradigm. Data were acquired for both sequences using TR values of 2, 3, 4, 6 and 8 sec. The clustered volume acquisition sequence was found to yield greater measures of dynamic range (percent signal change, mean statistical power per unit imaging time) across the tested range of TR values. Observations of more consistent spatial extent of responses, greater mean signal changes, and higher and more consistent values of mean t-statistic per unit imaging time demonstrate the efficacy of using a clustered volume acquisition for fMRI of auditory cortex.

Dynamic echo-planar imaging of the breast: experience in diagnosing breast carcinoma and correlation with tumor angiogenesis.

PURPOSE: To correlate quantitative echo-planar magnetic resonance (MR) imaging measures of gadopentetate dimeglumine tumor uptake with histologic diagnoses and microvessel density (MVD) and to compare dynamic echo-planar imaging of breast lesions with conventional dynamic MR imaging techniques. MATERIALS AND METHODS: The study group comprised 63 patients (aged 13-70 years) with 71 breast lesions who underwent conventional and echo-planar MR imaging. The T1 values, change in gadopentetate dimeglumine concentration, and extraction-flow products were calculated with the echo-planar imaging data and were correlated with histologic findings and MVD estimates. Extraction-flow product data normalized to pectoral muscle gadopentetate dimeglumine concentration in invasive cancers was also correlated with MVD. RESULTS: On average, cancer T1 values were shorter than benign values, but there was substantial overlap between the two groups. Cancers had higher extraction-flow products than benign lesions (P < .001). Sensitivity, specificity, positive predictive value, and negative predictive value were 83%, 79%, 67%, and 90%, respectively. Receiver operating characteristic analysis showed improved performance with extraction-flow products than with percentages of signal intensity change. Among the invasive cancers, there was no significant correlation between extraction-flow product and MVD. CONCLUSION: The T1 value remains important in more precise quantitative estimation of gadopentetate dimeglumine uptake in breast tumors, which helps improve the specificity of dynamic imaging. Tumor MVD affects the contrast medium enhancement of breast lesions, but other factors contribute.

[Brain structures of reading Japanese words with functional MRI].

This review examines recent progress in understanding mechanisms involved in language related brain functions using functional magnetic resonance imaging (fMRI). The main focus is to detect differences in reading processes between the ideogram (Japanese kanji) and the phonogram (Indo-European language or Japanese kana). Inferior temporal (IT) areas on the language dominant side are involved in reading both characters. A plasticity model is introduced to explain the different localizations that have been ascribed to kana or kanji, respectively. FMRI will be an important clinical application for the preoperative evaluation of patients with lesions adjacent to these areas in the near future. Activation studies involving visual imagery and other relevant clinical issues are also briefly reviewed.

Benign and malignant breast lesions: differentiation with echo-planar MR imaging.

PURPOSE: To quantify dynamic enhancement of breast lesions with echo-planar and conventional magnetic resonance (MR) imaging, to correlate these data with histologic findings and vessel density, and to evaluate MATERIALS AND METHODS: Twenty female patients with 22 breast lesions underwent conventional and MR echo-planar imaging T1 values, change in gadopentetate dimeglumine concentration, and extraction-flow products were calculated with echo-planar imaging data and were correlated with histologic findings and microvessel density. RESULTS: T1 values of cancers were not statistically significantly shorter. Cancers had more rapid uptake and higher extraction-flow products (P < .02). Sensitivity was 86% and specificity was 93% for diagnosis of malignancy. Microvessel density was higher for malignant lesions (P < .02) with an overall positive (not statistically significant) correlation between extraction-flow product and microvessel density. CONCLUSION: Echo-planar imaging appears promising for quantification of breast lesion enhancement. Microvessel data indicate that tumor angiogenesis affects enhancement.