Yingbo Ma develops machine learning for multimodal data in healthcare and education, combining text, speech, vision, and EHRs. Current work spans federated clinical risk prediction, transparent AI interfaces, and signal processing for active noise control.