Yasuyuki Yamashita

Impact of FDG-PET/CT Fused Imaging on Tumor Volume Assessment of Head-and-Neck Squamous Cell Carcinoma: Intermethod and Interobserver Variations

Background: Although gross tumor volume (GTV) at the primary site can predict local control of head-and-neck squamous cell carcinoma (SCC) in patients who are treated with organ-preservation therapy, GTV assessment does not eliminate substantial interobserver variation. Purpose: To evaluate whether F-18-fluorodeoxyglucose positron emission tomography (FDG-PET)/computed tomography (CT) fused imaging provides additional information for GTV assessment. Material and Methods: We obtained FDG-PET/CT fused images on 20 patients with head-and-neck SCC. All had undergone preoperative conventional workup, including contrast-enhanced CT and magnetic resonance imaging (MRI). The GTV of the primary tumors was designed by two independent observers who used routine clinical data. Observer A was a radiologist and observer B a radiation oncologist. GTV1 and GTV2 were designed without and with FDG-PET/CT, respectively. For geometric interobserver comparison, we calculated the concordance rate as the ratio of the intersection (A⋂B) of the GTVs to their union (A⋃B). Intermethod (GTV1 vs. GTV2) and interobserver (A vs. B) differences in the GTVs were assessed by Bland-Altman analysis and the Spearman rank-correlation test. The interobserver concordance rates for GTV1 and GTV2 were compared using a two-tailed paired-samples t test. Results: On FDG-PET/CT, all primary tumors were visualized. There was no systemic trend for a volume difference between GTV1 and GTV2. Although the 95% limits of agreement were wider for interobserver than intermethod differences, the 95% limits of interobserver agreement were narrower for GTV2 than GTV1. The mean interobserver concordance rate for GTV2 was higher than for GTV1 (54.5% vs. 39.1%, P=0.0002). Conclusion: FDG-PET/CT is a useful modality for consistent GTV assessment, which should not be used as a single modality but rather to obtain supplemental information in patients with head-and-neck SCC.

Dose profiles for lung and breast regions at prospective and retrospective CT coronary angiography using optically stimulated luminescence dosimeters on a 64-detector CT scanner

PURPOSEnThe purpose of our study was to acquire dose profiles at critical organs of lung and breast regions using optically stimulated luminescence (OSL) dosimeters; assess the actual radiation dose delivered at retrospective and prospective computed tomography coronary angiography (CTCA).nnnMATERIALS AND METHODSnUsing a chest CT phantom we applied a prospectively-gated step-and-shoot- and a retrospectively-gated helical mode on a 64-detector row CT scanner. Retrospective scan mode was used with and without electrocardiogram (ECG) based tube current modulation. OSL dosimeters were used to measure dose profiles. In the both scan modes we acquired dose profiles and determined the mean and maximum dose in left lung and in left breast regions.nnnRESULTSnIn prospective mode, the mean dose was 21.53 mGy in left lung- and 23.59 mGy in left breast region. With respect to the retrospective mode, the mean dose with tube current modulation was 38.63 mGy for left lung- and 46.02 mGy for left breast region, i.e. 0.56 and 0.55 times lower than the mean dose without modulation.nnnCONCLUSIONnThe OSL dosimeter is useful for measurement of the actual radiation dose along z-axis at lung and breast regions in the prospective and the retrospective CTCA.

3D TOF MRA of Intracranial Aneurysms at 1.5 T and 3 T. Influence of Matrix, Parallel Imaging, and Acquisition Time on Image Quality-A Vascular Phantom Study

RATIONALE AND OBJECTIVESnA 3-T magnetic resonance imaging system provides a better signal-to-noise ratio and inflow effect than 1.5 T in three-dimensional time-of-flight (3D TOF) magnetic resonance angiography (MRA). The purpose of this study is to analyze the influence of matrix, parallel imaging, and acquisition time on image quality of 3D TOF MRA at 1.5 and 3 T, and to illustrate whether the combination of larger matrixes with parallel imaging technique is feasible, by evaluating the visualization of simulated intracranial aneurysms and aneurysmal blebs using a vascular phantom with pulsatile flow.nnnMATERIALS AND METHODSnAn anthropomorphic vascular phantom was designed to simulate the various intracranial aneurysms with aneurysmal bleb. The vascular phantom was connected to an electromagnetic flow pump with pulsatile flow, and we obtained 1.5- and 3-T MRAs altering the parameters of 3D TOF sequences, including acquisition time. Two radiologists evaluated the depiction of simulated aneurysms and aneurysmal blebs.nnnRESULTSnThe aneurysmal blebs were not sufficiently visualized on the high-spatial resolution 1.5-T MRA (matrix size of 384 x 256 or 512 x 256), even with longer acquisition time (9 or 18 min). At 3 T with acquisition time of 4.5 min using parallel imaging technique, however, the depiction of aneurysmal blebs was significantly better for the high-spatial resolution sequence than for the standard resolution sequence. For the high-spatial resolution sequence, the longer acquisition times did not improve the depiction of aneurysmal blebs in comparison with 4.5 min at 3 T.nnnCONCLUSIONSnFor 3D TOF MRA, the combination of the large matrix with parallel imaging technique is feasible at 3 T, but not at 1.5 T.