Lithium-ion batteries are essential in numerous applications, with their performance strongly influenced by the instantaneous value of internal resistance (R0). Over time, increases in internal resistance reduce efficiency and cause voltage fluctuations, directly affecting the battery’s State of Health (SOH). This study introduces an AI-driven SOH prediction system that uses a Long Short-Term Memory (LSTM) neural network to forecast SOH by analyzing instantaneous internal resistance along with real-time voltage, current, and temperature data. A Processor-in-Loop (PIL) simulation framework is employed to validate the predictive model under realistic operating conditions, enhancing the accuracy and reliability of SOH predictions. Comprehensive simulations and experimental results demonstrate the system’s effectiveness in monitoring battery health, showing that accurate SOH prediction through real-time internal resistance analysis can significantly improve battery performance assessments.