使用吸氣和呼氣電腦斷層影像進行「慢性阻塞性肺病」之異質性和嚴重性的定量、分類和預測

Abstract

慢性阻塞性肺疾病（COPD）是一種進行性肺部疾病，據世界衛生組織（WHO）統計顯示，慢性阻塞性肺病（COPD）將於2030年成為全球第三大死亡原因。COPD會引起全身性炎症和合併症，包括缺血性心髒病，肺動脈高壓，骨質疏鬆症，貧血，憂鬱和惡病質。慢性咳嗽和暴露於空氣汙染危險因子（例如吸煙）可用於評估COPD，COPD的診斷基於肺功能檢查;對氣流受限的測量。然而COPD更像是一種綜合徵而不是一種疾病，肺功能檢查只能用於功能性的評估而不能用於探討疾病的異質性。造成氣道阻塞的主要原因是肺氣腫，慢性支氣管炎和慢性哮喘性支氣管炎。電腦斷層掃描（CT）成為表徵和量化COPD惡化的標準工具。儘管低衰減體積百分比（LAV％）代表CT上肺氣腫體積與整個肺體積的比例;雖然該指數與COPD患者的肺功能相關，但由於肺氣腫異質性，LAV％測量值相似的患者肺功能可能會有較大差異。除了透過測量吸氣和呼氣CT的低衰減區域對肺氣腫和功能性細支氣管疾病進行量化和分類外，橫膈膜的檢測還提供了有關細小支氣管功能障礙程度的有用資訊。例如橫膈肌偏移量的減少間接顯示出呼吸衰竭的現象。儘管進行了診斷和治療，但COPD的特徵是持續性氣流受阻且不可完全逆轉。目前有效的改善的治療方針是使用支氣管鏡植入氣管瓣膜（EBV）進行支氣管鏡肺減容治療（BLVR）。此手術將成為治療嚴重肺氣腫患者的常規方法。儘管治療的長期療效取決於瓣膜位置的準確定位，但目前的術前計畫仍未能有效定位出可有效治療的肺葉。為了精準治療，術前需要透徹分析區域性肺氣腫的破壞程度以及該區域對肺功能的影響程度。本文根據肺氣腫密度（ED）的大小變化及其空間分佈，為三維肺氣腫大皰建立了肺氣腫異質性描述子。第二個目的是基於肺氣腫和功能性細支氣管疾病（fSAD）的區域異質性，得出氣流受限的預測模型。進而推導出利用球狀結構的特徵因子描述位於上，下肺葉的多個尺度的低衰減簇（LAC），使用線性回歸模型計算出預測模型，透過影像對位的吸氣和呼氣CT估計肺功能的嚴重性。

Chronic obstructive pulmonary disease (COPD) is a progressive lung disease and the estimates show that COPD becomes in 2030 the third leading cause of death worldwide according to WHO statistics. Body of evidence demonstrates that COPD causes systemic inflammation and co-morbidities, including ischemic heart disease, pulmonary hypertension, osteoporosis, anemia, depression, and cachexia. The symptoms of chronic cough and exposure to the risk factors, such as cigarette smoking can be used to assess COPD, the diagnosis of COPD is based on the airflow limitation measurement by spirometry. However, COPD is more like a syndrome than a disease that the global measurement of spirometry cannot be used to assess the heterogeneous disorder. The main causes of this airway obstruction are emphysema, chronic bronchitis, and chronic asthmatic bronchitis. Computed tomography (CT) is becoming a standard tool for characterizing and quantifying COPD exacerbation. Although low attenuation volume percentage (LAV%) which represents the proportion of emphysema volume to whole lung volume on CT, correlates with lung function in patients with COPD, patients with similar measurements of LAV% may have large variations in lung function due to emphysema heterogeneity. Other than quantification and classification of emphysema and functional small airway disease by measuring the low attenuation regions on inspiratory and expiratory CT, the detection of diaphragm excursion also provides useful information on the extent of small airway dysfunction. Decreased diaphragmatic excursion prolongs expiration in patients with respiratory failure due to COPD. Despite the diagnosis and treatment, COPD is characterized by progressive airflow limitation and not fully reversible. Advanced bronchoscopy lung volume reduction treatment (BLVR) using endobronchial valves (EBV) is now a routine care option for treating patients with severe emphysema. While the long-term efficacy of the treatment relies on the accurate positioning of the valve placement, the method for target lobe selection remains unmet medical need. To target the treatment candidate, the functional information for regional emphysema destruction is required for supporting the choice of optimal target. In this thesis, we develop an emphysema heterogeneity descriptor for the three-dimensional emphysematous bullae according to the size variations of emphysematous density (ED) and their spatial distribution. The second purpose is to derive a predictive model of airflow limitation based on the regional heterogeneity of emphysema and functional small airway disease (fSAD). Deriving the bullous representation and grouping them into multiple scales of low attenuation clusters (LACs) in the upper and lower lobes, a predictive model is computed using a linear model fitting to estimate the severity of lung function using co-registered inspiratory and expiratory CT.