透過您的圖書館登入
IP:3.19.234.150
  • 期刊

應用碎形維度檢測乳房放大影像與磁振造影之病灶

Lesion Detection in Magnification Mammography and Magnetic Resource Imaging Using Fractal Dimension

摘要


乳腺癌是目前女性中最常見的癌症類型之一,藉由乳房攝影診斷提高早期篩檢率,當影像發現乳腺局部增生時,利用放大局部壓迫(spot compression magnification view, SCM)乳房評估邊緣結構為現今標準輔助方式,本研究目的為使用碎形維度分析放大乳房攝影與乳房磁振造影的病灶區域以評估惡性與良性,並與乳房磁振造影與臨床切片結果進行比較。結果顯示放大局部壓迫乳房與乳房磁振造影的碎形維度為1.46±0.12與1.40±0.13,與臨床切片結果相比,惡性與良性的準確度為80%與83%,敏感度皆超過90%。本研究建立SCM影像病灶與碎形維度之關聯,進一步強化影像特徵與病灶關聯,以協助臨床醫師診斷惡性腫瘤。

並列摘要


Breast cancer is the most common types of cancer among women. The early screening rate is improved by mammography diagnosis. When the image shows local hyperplasia of the breast, the evaluation of the edge structure by using the spot compression magnification view (spot compression magnification view, SCM) is the current standard procedure. This study aimed to use fractal dimensions to analyze the SCM and breast MRI (bMRI) to assess malignancy and benign and compare with the results of breast MRI and clinical slices. The results show that the fractal dimensions for SCM and bMRI are 1.46±0.12 and 1.40±0.13. Compared with the biopsy, malignant and benign accuracy is 80% and 83% for SCM and bMRI, respectively. The sensitivity is up to 90%. This study establishes the association between SCM and fractal dimensions. Further, it strengthens the association between imaging features and lesions to assist breast radiologist in diagnosing malignant tumors.

延伸閱讀