With the popularity of social media, cancer patients can share their experiences and emotions on social media platforms. These social media data contain rich emotional information, which is of great significance for understanding the psychological state and needs of cancer patients. This study aims to apply sentiment analysis algorithms to the social media of cancer patients and help medical institutions and caregivers better understand the emotional state of the patients and provide personalized care and support. We introduced sentiment analysis algorithms based on deep learning. Through the analysis of the social media data of cancer patients, the emotional status of the patients can be used for their treatment to improve their quality of life and social support. The results of this study provide a reference for medical institutions and nursing staff to improve the mental health of cancer patients, too.