This chapter explores the application of machine learning (ML) to infectious disease surveillance during emergencies, such as pandemics, natural disasters, and armed conflicts. It examines the challenges and limitations of data collection in these settings, including data quality, completeness, and integration into existing healthcare infrastructures. The discussion highlights how ML models, utilizing advanced techniques such as data augmentation, feature engineering, and anomaly detection, can significantly improve disease prediction and management accuracy and timeliness. Key points include the potential of ML to revolutionize public health responses through real-time data processing and the integration of diverse data sources. The chapter also addresses the future landscape of ML in public health crises, emphasizing the need for enhanced data collection methods, real-time analytics, and the development of more interpretable and secure ML models. Ultimately, this chapter underscores the transformative potential of ML technologies in improving public health surveillance and response during emergencies while acknowledging the critical need to overcome current challenges to fully realize these benefits.