Technical Name 人工智慧輔助胰臟癌偵測工具-PANCREASaver
Project Operator National Taiwan University
Project Host 王偉仲
Summary PANCREASaver contains a “PC automatic segmentation model” (image segmentation)a “PC analysis AI model” (image classification) that can read the DICOM format of postcontrast CT images directly for the automatic analysis process. After conducting prep-processing with image processing algorithms, C2FNAS is employed to illustrate the tumor position prior to the diagnosis conducted by CNN. The results can be provided to the physician for diagnostic reference so as to reduce early omissionsincrease the detection rate of pancreatic cancer.
Technical Film
Scientific Breakthrough The overall five-year survival rate of pancreatic cancer is less than 10, which turn it the most lethal cancer. However, if the tumor is detectedtreated sooner when it is smaller than 2 cm, the five-year survival rate can be increased to 80. PANCREASaver is the world’s first computer-assisted detection system that uses CNN models to detect pancreatic cancer on CT imagesassists physicians when interpreting. It increases the tumor detection rate of pancreatic cancer on CT imagesfulfills the urgent clinical needs of diagnosing pancreatic cancer that is often ignored on CT images.
Industrial Applicability PANCREASaver is the world's first AI pancreatic cancer (PC) detection model that can detect 92.1 of tumors 2 cm on CTsassists physicians in interpretation to improve patient survival. The usage of abdominal CT scans is up to 920,000 per year in Taiwan, thus, a combination of PANCREASaverCT scan is a possible new diagnostic tool that can create innovative medical servicesbusiness opportunities for medical institutionsphysical examination centers. Besides, health insurance-related units can also benefit from saving the high medical expenses of advanced PC patients.
Matching Needs 天使投資人、策略合作夥伴
Keyword PANCREASaver Pancreatic cancer diagnosis Computed tomography Image interpretation Computer-aided detection system Convolutional neural network (CNN) Coarse-to-fine neural architecture search (C2FNAS) Deep learning Tumor detection Image segmentation and classification
  • Contact
  • 潘美慈
Other people also saw
本網站使用您的Cookie於優化網站。繼續瀏覽網站即表示您同意本公司隱私權政策,您可至隱私權政策了解詳細資訊。