Room: Exhibit Hall | Forum 8
Purpose: T2-weighted MRI provides differential diagnostic information. The Double Echo Steady State (DESS) acquisition provides T2 contrast with 3D acquisitions, higher resolution and shorter scan times than fast Spin Echo (CUBE) T2 methods, which is used as current T2 clinical protocol. In this work, we investigate the internal tumor heterogeneity in DESS T2 versus CUBE T2 acquisitions to explore the clinical value of DESS acquisition.
Methods: DESS, CUBE, and Dynamic Contrast Enhanced (DCE) acquisitions were performed in 7 patients, with a total of 11 biopsy-proven tumors as part of a research protocol following IRB policies. Scans were acquired on a GE Discovery MR750 3T scanner with a 16-channel Sentinelle breast coil. (CUBE: 320 x 320, 36 cm FOV, 3-mm sl thick, TE 80 ms, DESS: 256 x 256, 36 cm FOV, 3-mm sl thick, TE 15 ms, DCE-MRI: 512 x 320, 27 cm FOV, 1-mm sl thick). A single central tumor slice was registered between the DCE and DESS and CUBE and DESS. Tumors were first segmented from DCE subtracted images via Fuzzy-C-Mean algorithm and then aligned to registered DESS and CUBE images for heterogeneity measurements. Registration was validated with the Dice overlapping ratios on the breast skin-tissue interface. Tumor heterogeneity was measured with entropy and uniformity.
Results: The dice ratio calculated for DESS versus CUBE and DESS versus DCE for all seven patients are 96.5±3.5% and 98.5±2.1%. The entropy and uniformity were highly correlated between DESS and CUBE images for all 11 tumors with Pearson correlation coefficient of 0.99(p <0.01) and 0.99(p <0.01).
Conclusion: The high correlation in tumor heterogeneity between DESS and CUBE indicates the T2 contrast is not greatly affected by the echo time difference between the two methods. The results of this work indicate that DESS may be an alternative for T2-weighted acquisitions for differential diagnosis.
Not Applicable / None Entered.
Not Applicable / None Entered.