基於電腦斷層成像分析腦組織成分的系統及其運作方法

專利類型

發明

專利國別 (專利申請國家)

中華民國

專利申請案號

109104583

專利證號

I 767188

專利獲證名稱

基於電腦斷層成像分析腦組織成分的系統及其運作方法

專利所屬機關 (申請機關)

臺北醫學大學

獲證日期

2022/06/11

技術說明

近年來國人腦中風發生率與死亡率居高不下，一直佔國人十大死因前三位。臨床上發現腦 部血管梗塞與中風息息相關且普遍存在於急性梗塞性腦中風病患。目前醫院使用軟體輔助 之半自動分割腦梗塞區域，處理與分析耗時過久且較容易產生不同量測者之偏差。大部分 的軟體也都是運用在拍攝時間較久且成本較高MRI 影像上，本專利開發急性缺血性腦中風 患者之電腦斷層影像上梗塞區域的偵測，藉助神經影像影像後處理與統計分析發展新穎的 腦梗塞區域偵測的方法，此方法能在臨床上更有效率的輔助醫師在電腦斷層影像上的診 斷。 Cerebral infarction (CI) refers to necrosis of brain tissue due to a blockage of cerebral blood vessels. This study aimed at the development of a method for detecting the presence and location of acute CI on computed tomographic (CT) images. Currently, available software algorithms mostly detect CI on magnetic resonance images (MRI), the acquisition of which is time-consuming and costly. On the other hand, the published CI detection algorithms on CT images are limited to whole-brain classification function without being able to pinpoint the location of CIs. In this study, the new algorithm is developed with the application of the deep learning method. CT images from acute ischemic stroke patients and healthy subjects were collected and preprocessed and then used to train the artificial neural network for the developed algorithm. This algorithm provides a facile way for physicians to diagnose CI from low-cost CT images. It could enhance clinical efficiency in the diagnosis and treatment of cerebrovascular diseases.

備註

連絡單位 (專責單位/部門名稱)

技術移轉中心

連絡電話

02-6638-2736-2006

網址