Technical Name AI Liver Tumor Detector
Project Operator National Cheng Kung University
Project Host 詹寶珠
Our technique is a liver tumor detection method using machine learning which is the most popular AI technique in these years. the high-magnification image provides more details, but the view is smaller in contrast, a low-magnification image has the larger view but less details. Thus, our method combines the two magnificationsyields more accurate tumor detection results.
Scientific Breakthrough
In tradition, the input to a CNN is often a single-scale image. A high-scale image have more details but lack surrounding features a low-scale image have more surrounding features but lack details. Thus, combine the advantages of both. First use CNNs to extract feature maps from highlow scale images respectively,then combine the two feature maps for analysis. The result will be better.
Industrial Applicability
In medical images, the input of multi-magnification images can be applied to the detection of tumor, portal areanecrosis, etc. Besides, combined with computer-aided calculations, some objective medical assessments can be obtained, which be assistants for doctors to assess the patient's situation. In addition to economic benefits, bring healthy well-being for all humans.
Keyword multi-magnification images convolutional neural network deep learning feature maps activation function Atrous spatial pyramid pooling fully connected layer field of view computer-aided diagnosis electric equipments
  • Contact