Search

This study aims to improve the accuracy of detecting acute tonic cold pain (CP) perception from microwave transcranial transmission (MTT) signals. Two different types of CP and no-pain (NP) MTT signals are obtained from 15 subjects. Four features, namely, power spectral exponential entropy, improved multiscale permutation entropy, refined composite multiscale dispersion entropy, and refined composite multiscale fuzzy entropy, are extracted in the variational modal decomposition domain. The feature datasets are divided into training datasets and test datasets in a 3:1 ratio. Random forest (RF) and support vector machine (SVM) are selected as classifiers. The training datasets are imported into the classifier, and the optimal training dataset is obtained with a 10-fold cross validation strategy. The feature dimension reduction algorithm of the principal component analysis is used to reduce the complexity of the feature datasets and select the most recognizable features. The classification performance of the test datasets is evaluated by the optimal classifiers. Results showed that the RF classifier performs better than the SVM classifier. The RF classifier provides high values of specificity (91.67%), sensitivity (95.83%), positive predictive value (92.00%), accuracy (93.75%), and area under curve (0.867). The combination of the microwave detection approach and machine learning algorithm can effectively detect brain activity induced by nociceptive stimulation. This approach is important in improving the accuracy of pain detection.

To investigate the feasibility of detecting acute tonic cold pain (CP) perception from recordable microwave transcranial transmission (MTT) signals by using machine learning techniques. CP and no-pain (NP) MTT signals collected from 15 young subjects are analyzed in the wavelet packet transformation (WPT) and variational mode decomposition (VMD) domains. In addition, features such as relative energy change, refined composite multiscale dispersion entropy, refined composite multiscale fuzzy entropy, and autoregressive model coefficients are extracted in the WPD, VMD, VMD-WPD, and WPD-VMD domains. Simultaneously, support vector machine (SVM) is selected as the classifier, and feature indexes are input into the classifier by using the 10-fold cross validation method to obtain the best training and test datasets. Principal component analysis is used to reduce the feature dimensions of the training and test datasets and to improve classification accuracy. Then, the test dataset is imported into the trained classifier for the calculation and evaluation of the model's classification performance. In the validation of the SVM classifier, feature extraction in the WPD-VMD domain is the best pain detection algorithm. It provides high values of sensitivity (91.30%), specificity (90.47%), positive predictive value (91.30%), accuracy (90.90%), and area under curve (0.806). The microwave scattering technique can be used as a direct, objective, and experimentally stable method to detect acute CP perception, this approach has a high application prospect for clinical real-time diagnosis.