1. Build an efficient and accurate deep learning model: By analyzing a large amount of ovarian cancer pathological image data, build a deep learning model with efficient and accurate prediction capability, which can automatically extract features from pathological images and predict patients' prognosis.
2. Improve the accuracy of ovarian cancer prognosis evaluation: Use deep learning models to predict the prognosis of ovarian cancer patients based on pathological images, improve the accuracy of prognosis evaluation, and provide more reliable evidence for clinical doctors to guide treatment plans and personalized medical decisions.
3. Explore new features related to prognosis: Through the analysis of features extracted by deep learning models from ovarian cancer pathological images, explore new features related to prognosis, and further provide a more comprehensive reference for ovarian cancer prognosis evaluation.
4. Establish an open-access prognosis prediction model: Based on the research results of this project, establish an open-access prognosis prediction model, so that other researchers and clinical doctors can benefit from it, and promote the development of ovarian cancer prognosis evaluation research and clinical applications.

Chen Ziqiang, School of Basic Medical Sciences, Fudan University.