The objective of this research project is to develop an automated muscle mass measurement system based on computed tomography (CT) images using deep learning and artificial intelligence, and to construct a prognostic prediction model for patients who have experienced cardiac arrest and undergone resuscitation. The model will be validated using multicenter registry data and medical imaging. Furthermore, a clinical trial involving core muscle rehabilitation for post-resuscitation patients will be conducted to evaluate both short-term and long-term outcomes, aiming to comprehensively understand the impact of muscle mass on the prognosis of patients after cardiac arrest.

Inclusion Criteria:

• Non-traumatic adult patients who experienced sudden cardiac arrest
• Achieved sustained return of spontaneous circulation (ROSC) for more than five minutes following successful resuscitation
• Admitted to the ICU
• Survived beyond 72 hours without successful weaning from mechanical ventilation

Exclusion Criteria:

• Age under 18 years
• Pregnancy
• Expected to be weaned from mechanical ventilation within 48 hours
• Terminal-stage cancer or unwillingness to receive aggressive treatment
• Unavailability of ICU care due to overcrowding in the emergency intensive care unit
• Refusal to participate in the study by the patient or family members
• Presence of an implanted pacemaker or defibrillator
• Open wounds, severe edema, or deep vein thrombosis in the lower limbs
• Pre-arrest status limited to wheelchair use or bedridden, unable to ambulate