Development of made to order image diagnostic system for cervical metastaic lymph nodes in patients with oral cancer.

结果表明：1）MRI STIR对颈部淋巴结肿大的检出率与CT相当; 2）背景信号抑制弥散加权全身成像（DWIBS）对颈部淋巴结肿大的检出率上级优于CT。但DWI对转移灶的诊断价值不大。3)诊断为转移者的转移概率为70.2%，诊断为“可能不是转移”者的转移概率为4.1%。超声诊断为反应性肿大淋巴结的病例中，有4.1%的病例需要观察。4)MRI T2加权像（T2 WI）呈极高或高信号区，病理检查提示囊性变或角化。认为T_2WI对转移性淋巴结具有特异性。肿瘤细胞和纤维组织在MR图像上无特异性表现。认为肿瘤细胞与正常淋巴组织的鉴别是困难的。5)我们提出的动态MRI小感兴趣区法比常规感兴趣区法更有利于淋巴结内肿瘤细胞的检测。6)认为CT图像上低关注区提示脂肪组织、出血、充血或滤泡增生。7)提示淋巴结多普勒血流定量与牙颌炎症有关。8)我们开发了颈部淋巴结影像报告系统，通过医院信息系统局域网发送到200多台PC机上。

Results of this rematch is as fellows; 1) Detection ration of swelling cervical lymph nodes using STIR of MR imaging was equal to that of CT 2) Detection ration of swelling cervical lymph ncdes using Diffusion weighted whole body imaging with background signal suppression (DWIBS) was superior to that of CT. However, DWIBS was not useful for diagnosing metastasis. 3) Probability of metastasis was 70.2%in case of diagnosing as metastasis, 4.1% in case of diagnosing as “Perhaps, it was not metastasis. However, observation was necessary “and 4.1% in case of diagnosing as reactive swelling lymph node, using ultrasonography. 4) Very hyperintensity or hyperintensity area in T2 weighted image (T2WI) of MR imaging suggested cystic degeneration or keratinization in pathological fmdings. It was considered that these fmdings of T2WI were specific for matastatic lymph nodes. Tumor cells and fibrous tissue were not presented specific fmdings on MR images. It was considered that differentiation between tumor cell and normal lymphoid tissue was difficult. 5) Small ROI method of dynamic MRI, which was developed by us, was more useful for detecting tumor cells in lymph node than conventional ROI method. 6) It was considered that low attention area on CT images suggested fat tissue, hemorrhage, hyperemia or follicular hyperplasia. 7) I was suggested that vascular ration of lymph nodes on Doppler ultrasonograms depended on inflammation of teeth and jaw bones. 8) We developed imaging reporting system for cervical lymph node, which send to about200PCs through LAN of our hospital information system.