Unsupervised Cross-User Adaptation in Taste Sensation Recognition Based on Surface Electromyography

Human taste sensation can be qualitatively described with surface electromyography (sEMG). However, the pattern recognition models trained on one subject (the source domain) do not generalize well on other subjects (the target domain). To improve the generalizability and transferability of taste sensation models developed with sEMG data, two methods were innovatively applied in this study: domain regularized component analysis (DRCA) and conformal prediction with shrunken centroids (CPSC). The effectiveness of these two methods was investigated independently in an unlabeled data augmentation process with the unlabeled data from the target domain, and the same cross-user adaptation pipeline was conducted on six subjects. The results show that DRCA improved the classification accuracy on six subjects ( $p < 0.05$ ), compared with the baseline models trained only with the source domain data, while CPSC did not guarantee the accuracy improvement. Furthermore, the combination of DRCA and CPSC presented statistically significant improvement ( $p < 0.05$ ) in classification accuracy on six subjects. The proposed strategy of combining DRCA and CPSC showed its effectiveness in addressing the cross-user data distribution drift in sEMG-based taste sensation recognition application. It also shows the potential for more cross-user adaptation applications.

I. Introduction

Taste sensation is one key component of human feedback, which is not only closely related to the perception and evaluation of food quality and safety [1] but also related to an individual’s health status. For example, cancer patients, diabetes patients, and COVID-19 patients might have disordered taste sensations as one of the complication symptoms. [2]–[4]. To objectively and quantitatively describe the human taste sensation, recently, researchers have developed systems based on surface electromyography (sEMG) and machine learning models to recognize human taste sensations [5], [6]. This approach enables the researchers to recognize the taste sensation without verbal expression from the subjects, which can be very helpful in objectively detecting and then improving the feeding experience of those who lost their verbal abilities due to accidents, heredity, and diseases, such as those suffering traumatic brain injury [7] in the ICU who cannot verbally express their taste feelings. For example, our previous research have reached a fivefold cross-validation accuracy of 74.46% on classifying six types of human basic taste sensations [5].