由核磁共振影像做基於人體解剖資料之自動化腹腔及腹腔內臟器脂肪測量

Automatic measurement of adipose tissue inside visceral cavity and organs using MRI and visible human data

簡欣永(SHIN-YUNG JIAN)

指導教授 ： 林康平

中原大學/工學院/電機工程研究所/碩士(2010年)

https://doi.org/10.6840/cycu201000845

摘要

本論文希望能提出一個便捷的方法計算出核磁共振影像的腹腔內脂肪面積、肝臟脂肪面積與脊椎的脂肪面積，讓醫生可以更方便做診斷，並能節省醫療人力資源，不要讓醫療人力浪費在繪製內臟脂肪的工作上。本論文的核磁共振影像從橫隔膜以下到骨盆底。其中包含三種不同的PROTOCOL，照影重點分別為肝臟、腹腔與骨盆腔。

本論文共探討七筆資料，每ㄧ筆資料約含70張影像。本論文利用形態學、直方圖均化與POWELL演算法等方法，輔以圖譜資訊，計算出每張影像的腹腔內脂肪面積、肝臟脂肪面積與脊椎內脂肪面積。並將結果和誤差利用統計中的正確率、準確率與敏感度三種參數呈現出來。

本論文使用直方圖均化與型態學等方法輔以圖譜，成功發展出一套便捷可以計算出人體腹腔磁振影像內的脂肪組織，平均誤差在24.4%，在未來或許可以輔助醫生診斷有關腹腔內脂肪上的疾病或是幫助到心血管方面的研究。

關鍵字

直方圖均化 ； 腹腔內脂肪 ； 肝臟脂肪 ； 核磁造影

並列摘要

This thesis hopes to present a convenient way to calculate the magnetic resonance imaging of intra-abdominal fat area and liver fat area and fat area of the spine, so that doctors can more easily make the diagnosis, and can save medical human resources, do not let doctors waste time in visceral fat rendering work. Magnetic resonance imaging of this paper is from the diaphragm to the pelvic floor. Which contains three different PROTOCOL, liver, abdomen and pelvis.

The thesis of seven patients about each data contains 70 images. In this thesis, morphology, histogram equalization and the POWELL algorithm and other methods, supplemented by visible human data for each image to calculate an area of intra-abdominal fat, liver fat, and spine fat. The results and the error rate using the correct statistics, the accuracy and sensitivity shown by three parameters.

This paper uses histogram equalization and morphological patterns and other methods supported by the successful development of a convenient can be calculated within the human abdominal adipose tissue magnetic resonance imaging, with an average error of 24.4%, in the future may be able to assist doctors to diagnose the intra-abdominal fat on cardiovascular disease or help to the study.

並列關鍵字

intra-abdominal fat ； liver fat ； magnetic resonance imaging ； histogram equalization

參考文獻

subcutaneous abdominal adipose tissue by magnetic resonance imaging. International Journal of Obesity 30, 844–852.

[3] NEMA,(2003) Digital Imaging and Communications in Medicine

(DICOM): National Electrical Manufacturers Association.

[4] “The Visible Human Project”,

DeFronzo RA. Abdominal fat distribution and peripheral and hepatic insulin resistance in type 2 diabetes mellitus. Am J Physiol Endocrinol Metab 2002; 283: E1135–E1143.