AI-based Frailty Assessment Tool for Patients Undergoing Cardiac Surgery

Frailty Assessment Tool for Patients Undergoing Cardiac Surgery Based on Artificial Intelligence Multimodal Non-contact Monitoring.

The main purpose of this study is to establish a frailty automatic evaluation index "AI frailty index" based on artificial intelligence multi-modal non-contact monitoring information analysis. At the same time, the study will explore the correlation between ' AI weakness index ' and perioperative and long-term prognosis and quality of life.

This is a prospective cohort study. Patients who undergoes elective cardiac surgery will be enrolled. We collect frailty scales, clinical information and multi-modal, non-contact monitoring information during hospitalization. One-year follow-up will be done. The non-contact monitoring information are used to predict the frailty scales to establish and validate an AI-based frailty assessment model "AI frailty index". The AI frailty index will be further used to predict perioperative and long-term outcomes.

Inclusion Criteria:

• Elective cardiac surgery

Exclusion Criteria:

• Emergency surgery
• Age ≤ 18 years
• The clinical situation is unstable, such as frequent occurrence of coronary ischemia events, unstable heart failure, or acute events leading to severe discomfort symptoms or changes in vital signs
• Unable to cooperate in completing research data collection due to severe neurological and psychiatric abnormalities or other reasons
• Duration of postoperative ventilation > 48 hours or inability to obtain scale data on the first day after surgery