Premature infants require specialized neonatal care, often through incubators that maintain ideal environmental conditions. However, conventional hospital incubators are costly and not readily available in remote regions. This project presents a low-cost, IoT-based portable incubator for premature infants using edge computing with the Raspberry Pi Zero 2W. The system monitors physiological parameters such as body temperature, blood oxygen, and heart rate, along with cabin temperature and humidity. Actuators like a heater and fan are controlled to maintain optimal conditions. A multi-output neural network trained with sensor data enables prediction of medical needs and environmental adjustments. The system supports local decision-making, real-time control, and remote access via a Telegram chatbot, enhancing neonatal care in low-resource settings.