本研究計畫的主要目的在研發電腦輔助偵測及影像分析系統來協助放射科醫師診斷三度立體人體掃描資料。這十年來因為三度立體掃描技術像是電腦斷層掃描(CT)和核磁共振掃描(MRI)的快速進步，放射科醫師有極詳細的三度立體圖形以利病症的早期診斷。雖然這些高解析度的三度立體圖形資料大幅度的提升了診斷正確性，但是缺乏有效影像處理軟體來過濾，偵測，並加強處理病徵嫌疑區域，放射科醫師被迫要在倍增的圖形資料裡去尋找細微病徵。因此大幅增加放射科醫師的讀圖時間，也減低醫院的看診效率。我們主張利用電腦影像分析系統來預先過濾分析原始資料並以最佳的視覺化方式以協助放射科醫師讀圖診斷。我們將以電腦斷層血管造影(CTA)為研究重點，主要目標在研發腦血管病變的電腦輔助分析系統。本計畫全程二年六個月，包含三個步驟及目標：(1)CTA成像的最佳化(2)血管影像強化及分割處理(3)血管形狀分析方法開發。我們的主要目的在自行研發電腦輔助醫學影像分析的重要核心技術以提升國人自行開發醫學應用軟體的能力。最終希望能發展為一套完整的電腦輔助偵測及分析系統以提升放射科醫師看診正確性及效率。 ; The main goal of this research project is to develop computer-aided detection (CAD) and image analysis systems to assist radiologists in reading three-dimensional (3D) scanning data. Since 3D scanning technologies such as computed tomography (CT) and magnetic resonance imaging (MRI) have improved dramatically for the past 10 years, radiologists are able to detect smaller lesions much earlier with unprecedentedly high-resolution 3D images. However, this improvement does come with a price. Without effective software to localize suspected regions first, radiologists are forced to detect small lesions from much increased raw data volumes. This significantly lengthens radiologists’ reading time and consequently lowers hospital throughput. In order to improve radiologists’ performance, we propose using CAD to pre-screen raw data and applying modern visualization technologies to enhance suspected regions. This 2-year-and-6-month pilot project will focus on processing cerebral CT angiography (CTA) data. It consists of three components: 1) CTA scanning protocol optimization; 2) blood vessel enhancement and segmentation; 3) vessel morphology analysis. Our main goal is to build a solid foundation which we can rely on to develop a fully-fledged, workable CAD system. Ultimately, we intend to build functional and commercially available software systems to improve radiologists’ diagnostic accuracy and efficiency. ; 研究期間 9806 ~ 9907
