AI Based Approach to Predict Ovarian Cancer Type
Principal Investigator
Name
Kezban Alpan
Degrees
Ph.D.
Institution
Middlesex University
Position Title
PhD Student
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCOI-1202
Initial CDAS Request Approval
Apr 19, 2023
Title
AI Based Approach to Predict Ovarian Cancer Type
Summary
Ovarian cancer is one of the most dangerous women's cancers. This disease, which is called "Silent Killer" due to its insidious progress, has an average of 14,000 women to die annually. There is an average of 4,200 ovarian cancer deaths in the UK every year. In advanced stages, metastasis has already occurred but it doesn’t mean that early-diagnosed patients are safe as ovarian cancer cells can get drug-resistant and the disease can progress even during chemotherapy. In this case, it is very important to know the existence of the danger in order to take precautions. In recent years, the Image Enhancement and AI based approaches draws attention with its successful prediction rate. The method is capable to combine different techniques. In this research, AI-integrated systems will be created by combining the output of image enhancement and U-Net. Numerical data that includes meta-data for image dataset was previously sent by PLCO at our request. The current goal is to deepen the research and bring it to the level of real artificial intelligence. For this purpose; numerical and image data will be processed with image enhancement and AI techniques.
Aims
• Making images more visible and clear with image enhancement method.
• Classifying ovarian cancer type with machine learning techniques using image data.
• Creating an image segmentation for prediction success.
• To predict the risk of possible metastasis in the patient with artificial intelligence by combining the results obtained with the multimodal method.
• To support the decision-making process of oncologists with the AI-powered system to be created.
• Prolonging the patient's life or providing full recovery by taking precautions for possible metastases.
Collaborators
Dr. Britta Stordal - Middlesex University