Background： Conventional weaning assessments are widely used but often affected by factors such as poor patient cooperation, neuromuscular issues, or unstable consciousness, leading to uncertainty in extubation timing. Studies have shown that HRV differs significantly between successful and failed extubation cases. However, most research features small sample sizes, focuses on single HRV indicators, and lacks integration with traditional assessments, limiting HRV's clinical application.

Effect： The study aims to develop a predictive model combining HRV and traditional weaning indicators to provide a more objective and sensitive physiological reference for ventilator weaning decisions in critical care settings.

Inclusion Criteria:

• Age between 18 and 85 years.
• Received mechanical ventilation for more than 24 hours and clinically assessed as ready for a Spontaneous Breathing Trial (SBT).
• Stable ECG signal quality before extubation, meeting the basic requirements for HRV analysis.
• Obtained informed consent from the patient or their legal representative, and agreed to participate in this study and cooperate with the 72-hour follow-up after extubation.

Exclusion Criteria:

• Individuals with permanent pacemakers, defibrillators, or other devices that affect monitoring data.
• Individuals on continuous deep sedation (RAS ≤ -4) and unable to regain consciousness in the short term.
• Individuals with severe arrhythmias or unstable cardiovascular conditions that prevent reliable HRV analysis.
• Individuals with cognitive impairment or mental illness that prevents them from cooperating with the research process, or individuals deemed unsuitable for participation in this study by their clinicians.

This study will recruit participants from the Intensive Care Unit (ICU) of Fu Jen Catholic University Hospital, including the Medical Intensive Care Unit, Surgical Intensive Care Unit, and Respiratory Care Center.