AI-Assisted Simple Scoring Algorithm Was Helpful in the Risk Assessment of Cardiac Involvement in Patients with...

Area of Science:

• Cardiology
• Pulmonology
• Medical Imaging

Background:

• Cardiac sarcoidosis (CS) is a severe complication of sarcoidosis, often presenting with subtle symptoms.
• Early and accurate diagnosis of CS is crucial for timely intervention and improved patient outcomes.

Purpose of the Study:

• To evaluate clinical predictors for the development of cardiac sarcoidosis in patients with pulmonary sarcoidosis.
• To develop and validate an AI-assisted scoring system for predicting cardiac involvement.

Main Methods:

• Retrospective analysis of 393 pulmonary sarcoidosis patients undergoing cardiac magnetic resonance (CMR).
• Application of original Lake Louise criteria for active myocarditis identification.
• Development of an AI scoring system using logistic regression, incorporating ECG, Holter, liver/spleen involvement, gender, and disease stage.

Main Results:

• Cardiac sarcoidosis confirmed in 52% (48/92) of patients who underwent CMR.
• Holter ECG abnormalities and liver/spleen sarcoidosis were significantly associated with CS.
• The AI scoring system achieved 76% sensitivity and 74% specificity for CS prediction.

Conclusions:

• An AI-assisted scoring algorithm effectively predicts cardiac involvement in pulmonary sarcoidosis patients.
• Holter ECG abnormalities and liver/spleen involvement are significant predictors of CS.
• Prospective validation is required to confirm the clinical utility of the AI scoring system.