透過您的圖書館登入
IP:3.149.229.253
  • 學位論文

醫療影像分析辨識與應用

The Analysis of Medical Images and Its Application

指導教授 : 游子宜

摘要


近年來台灣女性乳癌發生率逐年攀升,乳癌在台灣女性十大癌症死亡率中一直都名列前茅。乳癌最常使用的篩檢方式之一為乳房X光攝影,根據我國衛福部資料顯示乳癌在第零期與第一期的治癒率大概為95%,如何在女性乳癌罹患的第零期與第一期就偵測到乳癌,一直是醫學界和學術界很努力的話題。因此本研究開發一套醫療影像處理系統,專注於乳房X光攝影的影像處理和鈣化點偵測,並快速且大量處理醫療影像。研究方法使用影像處理的技術,對醫療影像進行處理,並透過本研究中的演算法計算出最佳閥值的方法(thresholding),再以分群分類的演算法,篩選可能是乳癌鈣化點症兆部位。透過本研究的演算法方法篩選出可能異常鈣化點區域,協助醫師對於乳癌可疑病兆判別,藉此減少醫療人員的負擔,降低對於疾病的誤判,提高治療乳癌的機會,達到提早發現提早治療。

並列摘要


The disease of the breast cancer in Taiwan has increased in recent year. This disease is among the top ten causes of death among women. One of the most commonly used detecting methods for breast cancer is Mammography. The statistics show that the successful treatment rate for the zero and first stage of breast cancer is nearby 95%. How to detect breast cancer of zero or the first stage has been a difficult research. This research has utilized imaging processing methods and applied to Mammography images, and aimed to detect the early calcification area in those mammography images. This reach can process many Mammography images quickly at the same time. We can calculate the optimal thresholding method through the algorithm, and then utilize the classification algorithm to identify the possible breast calcifications area. This study is aimed to find the area of possible abnormal breast calcifications and to assist physicians in identifying suspicious breast cancer. This study can help to reduce the burden for medical personnel by avoid miscalculation, and improving the chance of treatment of breast cancer by early detection.

參考文獻


[1] 衛生福利部統計處,十大癌症發生率,取自https://dep.mohw.gov.tw/DOS/cp-1720-7336-113.html。
[2] 行政院性別平等會,重要性別統計資料庫,女性乳癌死亡率,取自https://www.gender.ey.gov.tw/gecdb/Stat_Statistics_Category.aspx?fs=fTQP3HmkUvd1PbnmtSP3rw%3D%3D&cs1=2iYDMFM0KRukRMtXpZ73IQ%3D%3D。
[3] 衛生福利部國民健康署,衛生福利部公布癌症發生資料,取自https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=4141&pid=12682
[4] World Health Organization,Latest global cancer data: Cancer burden rises to 19.3 million new cases and 10.0 million cancer deaths in 2020,取自https://www.iarc.who.int/news-events/latest-global-cancer-data-cancer-burden-rises-to-19-3-million-new-cases-and-10-0-million-cancer-deaths-in-2020
[5] 國家生技醫療產業策進會,世衛:乳癌擠下肺癌成全球最常見癌症,取自https://ibmi.taiwan-healthcare.org/news_detail.php?REFDOCTYPID=0o4dd9ctwhtyumw0&REFDOCID=0qnxhczsj12w5h65&PageNO=1

延伸閱讀