Computer-aided Diagnosis System for Lung Nodule Detection

[[abstract]]本文在發展一演算法用於胸腔放射影像之肺部腫瘤，偵測可疑腫瘤的可能存在部位，提供醫師做為參考。方法上以差值影像為基礎，以似圓性來圈選可疑腫瘤區域，然後再利用生理資訊（灰階度判斷是否鈣化，位置判斷是否為縱膈…等）及利用監督式的倒傳遞類神經網路對圈選出來的可疑區域進行處理而減少FP 數目。此外，也藉由臨床醫師之協助比較系統與傳統人工判讀之區別。系統除了提供面積之計算外，在假體測試中均成功的測得目標物。在真實影像之測量中，在閥值為30％條件下，對一般腫瘤較少的病人其敏感度均能達到1。對腫瘤較多之影像，在閥值為32％條件下，其敏感度可以達到0.96667。與其他系統比較中，本系統的FP 數較少，且準確性較高。 In this paper, algorithms were developed to detect lung nodules on chest radiological image, and mark possible region of suspicious nodules for doctor's diagnosis. The detecting method was based on difference image and marked all suspicious nodules regions by circularity at difference image. Then, two algorithms were used on the suspicious regions for reducing the number of false positive(FP). A biological information which is determined whether calcification by gray scale and verify whether mediastinum by position was used to reduce the number of FP. If it didn't show great effect, artificial neural network (ANN) was applied to reduce FP number. Besides, by the aids of clinical doctors, we found that the difference of performance between our system and traditional method could be distinguished. Our system provided the calculation of suspicious area and successfully detected the nodules in phantom. In real image, the sensitivity approaches to 100% for less nodules patients when applied the gray level of 30% in cumulative histogram as a threshold. For more nodules image, the sensitivity approached to 0.96667 when applied 32% gray level as a threshold. Comparing with other systems, this system obtained less FP and higher accuracy