To address the lack of objective and quantitative criteria for the “acupoint stagnation” state in cervical spine palpation in traditional Chinese medicine (TCM), this study proposed a quantitative method for assessing acupoint tissue stiffness based on a flexible sensing array, and evaluated its effectiveness in the TCM pathological classification of cervical spondylosis. A wearable, vertically distributed flexible sensing array made of gallium-indium alloy and polyvinylidene fluoride (GaIn-PVDF) was employed and attached to the cervical acupoint region in a micro-contact and low physiological burden manner. By leveraging the high-sensitivity response of the GaIn strain unit to skin deformation under force and incorporating Hertz contact theory, a quantitative model was established to convert sensor resistance changes into the equivalent stiffness of tissue at the acupoint. The tissue stiffness output by the model showed a strong correlation with clinical Shore hardness values (Pearson correlation coefficient = 0.96), indicating that the proposed model can effectively reflect the true stiffness of acupoint tissue. On this basis, acupoint electrophysiological parameters and pressure sensitivity thresholds were integrated, and a multi-parameter TCM pathological classification model based on principal component analysis (PCA) and K-nearest neighbor (KNN) was constructed, achieving an overall multi-class prediction accuracy of 85.7%. Furthermore, the proposed sensing array was applied to monitor cervical range of motion, and a comprehensive evaluation framework integrating TCM pathological classification with Western medicine cervical range of motion was established. In addition, the piezoelectric effect of the PVDF unit was utilized to map acupoint pressing signals into the actions of a virtual game character, realizing human-computer interaction and enhancing patient engagement in rehabilitation training. This work provides a new technical pathway for the objectification and digitalization of TCM acupoint diagnosis and demonstrated good potential for use in home health management and smart rehabilitation scenarios.