利用X光影像特徵值評估髖關節置換術後緊密性之研究

The study of evaluated features for X-ray image after total hip replacement

林幸元(Shing-Yuan Lin)

指導教授 ： 黃詠暉

共同指導教授 ： 陳泰賓(Tai-Been Chen)

義守大學/電機資訊學院/資訊工程學系/碩士(2017年)

摘要

人工髖關節置換術(Total Hip Replacement, THR)主要作為髖關節關節炎、類風濕性關節炎、股骨頭缺血性壞死、股骨頸骨折、等退化性與外傷性髖關節病變之治療。髖關節置換金屬植入人體，由於緊密性不足，使得髖臼與腿骨元件以及骨水泥膠與組織產生變形、磨損、鬆脫、感染及骨溶解，造成患者髖部疼痛、步伐不穩、髖關節支撐力不足、以及其他不適情形。
本研究採用回顧性配對實驗設計。取樣時間為2004年至2016年，樣本數為共38例，其中男性21例、女性17例。第一次THR術後追蹤之平均年齡為52.5±13.2歲；第二次THR術前追蹤之平均年齡55.3±13.2歲，收集人工髖關節置換的患者為主。第一次THR術後的影像為對照組；第二次THR術前影像為實驗組。利用選定5條影像剖面線形各萃取4種特徵比值，包括剖面線形圖(Line Profile)影像強度最大與最小比值(Max_Min)、剖面線形圖影像強度中位數與最小比值(Med_Min)、剖面線形圖影像強度平均數與最小比值(Mean_Min)、剖面線形圖影像強度平均數與標準差比值(Mean_SD)。經由統計分析檢定控制組與實驗組特徵值差異性與再置換天數關連性；統計分析包括描述性統計、無母數魏克森檢定、皮爾森相關係數以及迴歸分析。。
本研究對照組至實驗組間隔平均天數為男性870.4±647.9天、女性為1318.2±979.7天。藉由魏克森無母數檢定各區位4種特徵比值差異性，發現4條剖面線形圖區位各有1個顯著影像特徵比值，包括區位一(L1)為影像強度最大與最小比值(Max_Min)、區位二(L2)為影像強度中位數與最小比值(Med_Min)、區位三(L3)為影像強度平均數與最小比值(Mean_Min)、區位四(L4)為影像強度平均數與標準差比值(Mean_SD) (P<0.05)。間隔天數與對照組影像特徵值在L1與L2有低度正相關(P<0.05)；L3與 L4區位對照組影像特徵值與間隔天數無明顯相關。迴歸分析結果發現年齡與對照組影像特徵值具有顯著性，可用於預估間隔天數。
本次研究初步了解各區位顯著性特徵比值；可藉由對照組影像特徵值及年齡預估髖關節鬆脫天數。唯本次研究樣本稍嫌不足，建議增加樣本收案數，以提高分析可靠性與準確性。

關鍵字

髖關節置換術 ； 剖面線形圖 ； 回顧性配對實驗設計 ； 影像特徵比值

並列摘要

Total hip replacement has been the primary choice of treatment for patients with degenerative or traumatic conditions such as advanced osteoarthritis, rheumatoid arthritis of the hip joints, avascular necrosis and fracture of the femur. The metallic implantation from hip replacement due to insufficient tightness could result in tissue deformation, wear, loosening, infection and bone dissolution between the acetabulum and the bone cement glue that cause pain, unstable gait, lack of joint support and other discomfort to the patients.
This is a retrospective study with matching experiments. Total of 38 cases of patient with artificial hip arthroplasty were enrolled from 2004 to 2016, including 21 males and 17 females. The mean age of the first THR postoperative follow-up was 52.5 ± 13.2 years; the mean age of the second THR was 55.3 ± 13.2 years. The first THR post-operative images were designated as the control group. The second THR preoperative image was selected as the experimental group. Four characteristic ratios were extracted by utilizing 5 imaging profiles, including maximum to minimum ratios (Max_Min), the median to minimum ratio (Med_Min), mean to minimum ratio (Mean_Min), mean to standard deviation ratio (Mean_SD)of the image intensities of the line profile graphs. The statistical analysis included the descriptive statistics, the nonparametric Wilcoxon rank test, the Pearson correlation coefficient and the regression analysis.
The average days between the images taken for control group and the experimental group was 870.4 ± 647.9 days for men and 1318.2 ± 979.7 days for women. By means of nonparametric Wilcoxon rank test, the difference of the four characteristic radios of each location is found. It is found that there is one significant image feature ratio in the four line. Including (L1) is the ratio of the maximum to minimum (Max_Min), (L2) is the ratio of the median and to minimum (Med_Min), (L3) the ratio of mean to minimum (Mean_Min), and the (L4) is the ratio of average to the standard deviation (Mean_SD) of the image intensity(P <0.05). There was a low positive correlation between the number of days and the control group (P <0.05). There was no significant correlation between the characteristic ratios of L3 and L4 location in the control group. The results of regression analysis showed that the age and control group had significant difference and could be used to estimate the number of days.
This study was to understand the significance ratio of each location. The number of days of hip joint after became loose was estimated by the characteristic ratios of images and age from the control group. The challenge of this study is the inadequate numbers of enrolled cases. It is recommended to increase the number of samples to improve the reliability and accuracy in the future.

並列關鍵字

Total Hip Replacement ； Line Profile ； Retrospective Study with Matching Experiments ； Image Feature Ratio

參考文獻

[1] Kawaji H, Uematsu T, Hoshikawa N, Watanabe H, Takai S. Clinical Results of VerSys Hip System (Zimmer) Uncemented Total Hip Replacement Arthroplasty. J Nippon Med Sch.2016;83(5):184-187.

[2] Röder C, Eggli S, Münger P, Melloh M, Busato A. Patient characteristics differently affect early cup and stem loosening in THA: acase-control study on 7,535 patients. Int Orthop.2008;2(1):33-8.

[3] Al-Azzani WA. Adverse reaction to metal bearing leading to femoral stem fractures: a literaturereview and report of two cases. J Surg Case Rep.2016 1-3;2016(2).

[4] Sasagawa T, Nakamura T. Associated Factors for Lumbar Degenerative Spondylolisthesis in Japanese Patients with Osteoarthritis of the Hip: A Radiographic Study. Asian Spine J.2016;10(5):935-939.

[5] Chammout G, Muren O, Bodén H, Salemyr M, Sköldenberg O. Cemented compared to uncemented femoral stems in total hip replacement for displaced femoral neck fractures in the elderly: study protocol for a single-blinded, randomized controlled trial (CHANCE-trial).BMC Musculoskelet Disord.2016 20;17(1):398-402.