Personalized and adaptive neural networks for pain detection from multi-modal physiological features

Mingzhe Jiang, Riitta Rosio, Sanna Salanterä, Amir M. Rahmani, Pasi Liljeberg, Daniel Santos da Silva, Victor Hugo C. de Albuquerque, Wanqing Wu

Abstract: Highlights•We proposed fusing personal pain sensitivity for personalized pain recognition.•We proposed using feature attention dynamic in time to enhance model adaptivity.•Within- and cross-datasets validations show better adaptivity to new pain events.•Cross-study comparisons indicate outperforming SOTA in recognizing 4-second pain.