本研究採用相位對比磁振造影為基礎之非侵入性量測顱內壓檢測技術。將量測腦脊髓液區域及六條血管，並遵循Alperin的研究方法去計算顱內順應性指數，且分成兩實驗組進行，一組將直接量測顱內順應性指數；另一組受測者將重複兩次實驗，一次在不干涉受測者狀態下進行掃描，另一次則在每次掃描前確定受測者保持清醒狀態。
本研究使用自行開發的全自動化血管分割系統，藉由全自動化定位內頸動脈、內頸靜脈及椎動脈位置，並對於不同血管採取對應的分割技術，達到精確的自動化分割結果以減少操作時間及降低人為誤差。
實驗結果中，第一組實驗我們測量出的原始數據結果，腦脊髓液的峰值間壓力梯度變化(Peak-to-peak pressure gradient, PGp-p)平均值為0.0624 (mmHg/cm)；峰值間顱內體積變化(Peak-to-peak intracranial volume change, ICVCp-p)平均值為0.422 (ml)；顱內順應性指數(Intracranial compliance index, ICCI)平均值為7.12 (ml/(mmHg/cm))，在第二組實驗當中，從結果的變異係數(Coefficient of variation, CV)可知在顱內順應性指數中，清醒組(CV=10.0%)的重複性較未干擾組(CV=25.7%)高，根據實驗，總結在量測ICCI時須確保受試者保持清醒狀態。睡眠狀態是影響顱內順應性指數之因素之一，進而導致重複性下降。本系統可做為長期非侵入性監測自發性顱內低壓患者使用。
In this study we use phase contrast (PC) MRI to measure the blood flow in ICA (internal carotid artery), VA (vertebral artery) and IJV (internal jugular vein). Also we measure the cerebral spinal fluid (CSF) flow. Combined these measurement we are able to calculate the intracranial compliance index (ICCI), which is defined by previous study.
In order to measure blood flow, we use different vessel segmentation techniques. The 3D dynamic programming is applied to segment ICA, the active contour model is used to segment IJV and the Hough circle transform is used to segment VA. With a proposed pre-process algorithm the positions of ICA, VA, and IJV are fully automated detected and segmented.
In the first experiment, the average value of CSF peak-to-peak pressure gradient (PGp-p) is 0.0624 (mmHg/cm), the average value of the peak-to-peak intracranial volume change (ICVCp-p) is 0.422 (ml), and the average ICCI is 7.12 (ml/(mmHg/cm)). In the second experiment, the reproducibility of the ICCI in awake group (CV: coefficient of variation = 10.0%) is better than that of the non-interfering group (CV = 25.7%). Based on the experiment, we conclude that the ICCI measurement should be done under the state of consciousness during the measurement. The conscious state is a factor which will affect the intracranial compliance index and reduce the reproducibility. The system can be used as a long-term non-invasive monitoring of patients suffering spontaneous intracranial hypertension.
