Stress is the body's response to stressful situations, and it can harm a person's health, especially chronic disease patients. Self-assessment questionnaires are the gold standard for stress assessment, but they're time-consuming and limiting. Facial analysis, voice recognition, thermography, and electrocardiography are used in several stress estimate models, although patients may not accept them. In this study, a multichannel detection system with a distributed computing platform is proposed to extract multiple physiological parameters using photoplethysmography signal processing algorithms. Body temperature and galvanic skin response acquisition was also added to provide the basis for the proposed model for stress levels estimation based on Fuzzy Logic. Thus, the stress is classified into five levels. To validate the proposed work, a two-phase experimental protocol was established, with the first phase consisting of evaluating the wearable devices' viability through a series of tasks. The second phase assessed the thermal stress induction model for stress classification. At the end of the exposure period, each participant was asked to selfevaluate their stress level on a scale from 1 to 5, providing reference values to compare with the model's estimates, resulting in good accuracy indicators. Participants could track their stress and wellness via a mobile application also proposed