利用擴散頻譜磁振造影神經纖維成像分析精神分裂症白質完整度：手動與自動化模板模式之比較

Diffusion spectrum imaging tractography analysis of white matter integrity in patients with schizophrenia: Comparison between manual approach and template-base automatic approach

何致韑(Jhih-Wei He)

指導教授 ： 林發暄 ； 曾文毅

國立臺灣大學/工學院/醫學工程學研究所/碩士(2012年)

https://doi.org/10.6342/NTU.2012.02774

摘要

精神分裂症是現代社會中影響身心靈、經濟很大的疾病，在台灣的盛行率約為0.3至1%。患者主要有幻覺、妄想、怪異行為等臨床表現。長期下來，也會有合併認知功能障礙以及造成情緒表達上的障礙。目前現有的研究讓我們了解到精神分裂症的病因之一可能是大腦各灰質區塊間聯繫出現阻礙，因此研究白質的神經完整性是當今的重要課題。再者，許多研究說明額葉眼眶面皮質與負向情緒以及焦慮的抑制有關，而這也是精神分裂症患者主要症狀。因此我們選定了下述三組連接至額葉眼眶面皮質的神經束來分析其完整性：兩側的前丘腦放射、兩側的鈎束以及胼胝體膝。此外，探討症狀輕重與神經完整性之關連亦為本研究之目標。
本研究共招募了二十位長期精神分裂症患者與二十位年齡性別相對應之對照組，每位受試者均接受擴散頻譜磁振造影以及T2權重影像二種磁振影像掃描，患者並於掃描前後門診接受活性與負性症狀量表之評估。首先使用手動模式神經纖維成像術來重建上述三組神經影像，並計算其總擴散不等向性。接著，比較病人組與對照組間之差異，並分析病人組中，總擴散不等向性與活性與負性症狀量表分數之相關。
組間結果顯示，相較於對照組，病人組除右側前丘腦放射之總擴散不等向性下降不夠顯著外，其餘神經束均有下降至顯著差異。總擴散不等向性與活性與負性症狀量表分數之相關結果顯示，胼胝體膝之總擴散不等向性與焦慮/憂鬱因子之分數有負相關。
此外，本研究另一目的為使用自動化模板模式來重現手動模式之結果。本模式中，病人組相較於對照組之總擴散不等向性均有下降但未達顯著。但由於擴散頻譜影像之標準化，我們可以得到等長的總擴散不等向性描繪圖，進而了解平均總擴散不等向性在解剖位置上高低值之分布並進行組間比較。結果顯示病人組的胼胝體膝與左側上縱束交錯位置之總擴散不等向性比對照組高。此外，前丘腦放射在丘腦位置之總擴散不等向性，對照組有左大於右的側化現象，而病人組沒有發現此現象。
在兩模式比較中，自動化模板模式所得到之總擴散不等向性與手動模式有高度正相關，且自動化模板模式所得到之總擴散不等向性描繪圖可與手動模式之結果互相呼應。
本研究顯現了擴散頻譜磁振造影與神經纖維成像術在精神分裂症中探討活體白質完整性的方便性，也讓我們對於精神分裂症的病理有更深一層的體認。另一方面，自動化模板模式中總擴散不等向性描繪圖的分析，也讓我們從整條神經束總擴散不等向性進而深入分析不同解剖位置。
未來手動模式與自動化模板模式將會相輔相成，融合個別特點以補足彼此之缺陷，以辟找到精神分裂症或其他疾病在神經影像診斷學上的生物標記。

關鍵字

精神分裂症 ； 額葉眼眶面皮質 ； 前丘腦放射鈎束 ； 胼胝體膝 ； 活性與負性症狀量表 ； 擴散頻譜磁振造影 ； 總擴散不等向性 ； 總擴散不等向性描繪圖

並列摘要

Schizophrenia is a high-impact mental disorder that affects approximately 0.3~1% of the population in Taiwan, with ravaging effects on both psychological and economical resources. This serious mental illness not only affects our cognition but also contributes to chronic emotional problems. Disturbed communication of white matter within the brain region may be the possible pathology of schizophrenia, which was mentioned previously. Therefore, it’s important to investigate the white matter’s integrity. Also, the orbitofontal cortex (OFC) was linked to negative emotions and anxiety, which were the symptoms of schzophrenia. There are three white matter tracts which connected to the OFC chosen to be studied in the investigation of the difference between patients with schizophrenia and normal control patients in our research: bilateral anterior thalamic radiation (ATR), bilateral uncinate fasciculus (UF) and genu.
We recruited 20 patients with chronic schizophrenia and age- and gender-matched healthy controls to acquire diffusion spectrum images and T2-weighted structure images. The patients’ PANSS scores were measured in outpatients one week before the MRI scan or one week after the scan. The first, tractogrphy of ATR, UF and genu was rebuilt in a manual approach, and the general fractional anisotropy (GFA) was calculated. Then, we compared the difference between two groups and analyzed the correlation between GFA values and PANSS scores.
In the results, the GFA values were significantly reduced in all tracts except the right ATR in schizophrenia. And there was a negative correlation between GFA values in the genu of a factor of 5 (anxiety/depression).
Moreover, we used an automatic template-based approach to compare results to the manual approach. In this approach, we found that there was a trend that all GFA values of tracts in schizophrenia were lower than normal controls’ even though it was not significant. But we could compare the profiles between groups because the lengths of the GFA profiles were the same after normalization. We found the mean GFA value in schizophrenia was higher in the genu in the intersection with left superior longitudinal fasciculus (SLF). Also, there was a left-right asymmetry in thalamus in normal controls.
There was a significant positive correlation of GFA values between these two approaches. These two approaches were auxiliary to each other because they possessed different traits and could be explained together.
Our findings showed the usefulness of applying DSI and tractography to investigate white matter fiber tracts in vivo in schizophrenia, and thus extended our understanding of the pathophysiology of this disease. Also, we could investigate the different parts of the GFA profile by the automatic template-based approach.
In the future, we will combine these two approaches in order to find the neuroimage biomarker of schizophrenia or other diseases.

並列關鍵字

schizophrenia ； orbitofrontal cortex ； anterior thalaic radiation ； uncinate fasciculus ； genu ； PANSS ； diffusion spectrum imaging ； tractography ； general fractional anisotropy ； GFA profile

參考文獻

Alexander, D. C., Pierpaoli, C., Basser, P. J., & Gee, J. C. (2001). Spatial transformations of diffusion tensor magnetic resonance images. IEEE Trans Med Imaging, 20(11), 1131-1139.

Ashburner, J. (2007). A fast diffeomorphic image registration algorithm. Neuroimage, 38(1), 95-113.

Barbas, H. (2000). Connections underlying the synthesis of cognition, memory, and emotion in primate prefrontal cortices. Brain Res Bull, 52(5), 319-330.

Basser, P. J., Mattiello, J., & LeBihan, D. (1994). MR diffusion tensor spectroscopy and imaging. Biophys J, 66(1), 259-267.

Bechara, A., Tranel, D., Damasio, H., & Damasio, A. R. (1996). Failure to respond autonomically to anticipated future outcomes following damage to prefrontal cortex. Cereb Cortex, 6(2), 215-225.