Multi-task multiple kernel machines for personalized pain recognition\n from functional near-infrared spectroscopy brain signals

2018/08/21 by Daniel Lopez-Martinez, Lopez-Martinez, Daniel, Ke Peng +9
Medicine · #FOS: Computer and information sciences #Heart Rate Variability and Autonomic Control #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Optical Imaging and Spectroscopy Techniques #Pain Mechanisms and Treatments

paper · pdf · doi:10.48550/arxiv.1808.06774

openalex publication_date 2018/08/21 · openalex created_date 2022/08/04 · openalex updated_date 2026/07/28

Abstract

Currently there is no validated objective measure of pain. Recent\nneuroimaging studies have explored the feasibility of using functional\nnear-infrared spectroscopy (fNIRS) to measure alterations in brain function in\nevoked and ongoing pain. In this study, we applied multi-task machine learning\nmethods to derive a practical algorithm for pain detection derived from fNIRS\nsignals in healthy volunteers exposed to a painful stimulus. Especially, we\nemployed multi-task multiple kernel learning to account for the inter-subject\nvariability in pain response. Our results support the use of fNIRS and machine\nlearning techniques in developing objective pain detection, and also highlight\nthe importance of adopting personalized analysis in the process.\n

Related

• Pain Detection with fNIRS-Measured Brain Signals: A Personalized Machine Learning Approach Using the Wavelet Transform and Bayesian Hierarchical Modeling with Dirichlet Process Priors
• Pain Assessment based on fNIRS using Bidirectional LSTMs
• Multi-task Neural Networks for Personalized Pain Recognition from Physiological Signals
• Detecting Pain During Cognitive Tasks via Visibility Graphs of Cortical Signals
• PainDECOG: Machine Learning-Based Identification of Pain Biomarkers from sEEG Signals