A Lightweight Transformer for Pain Recognition from Brain Activity
It addresses the need for reliable automated pain assessment in clinical settings by offering a compact model that can run on both GPU and CPU.
The paper introduces a lightweight transformer for pain recognition from fNIRS brain activity that fuses multiple representations via a unified tokenization mechanism, achieving competitive performance with computational efficiency suitable for real-time inference.
Pain is a multifaceted and widespread phenomenon with substantial clinical and societal burden, making reliable automated assessment a critical objective. This paper presents a lightweight transformer architecture that fuses multiple fNIRS representations through a unified tokenization mechanism, enabling joint modeling of complementary signal views without requiring modality-specific adaptations or increasing architectural complexity. The proposed token-mixing strategy preserves spatial, temporal, and time-frequency characteristics by projecting heterogeneous inputs onto a shared latent representation, using a structured segmentation scheme to control the granularity of local aggregation and global interaction. The model is evaluated on the AI4Pain dataset using stacked raw waveform and power spectral density representations of fNIRS inputs. Experimental results demonstrate competitive pain recognition performance while remaining computationally compact, making the approach suitable for real-time inference on both GPU and CPU hardware.