Attention Deficit Hyperactivity Disorder (ADHD) is a common neuropsychiatric disorder in childhood, characterized by inattention, impulsiveness, and hyperactivity, which adversely affects the ability to control emotions. To overcome the limitations proposed, this study investigates the use of advanced technologies such as Convolutional Neural Networks (CNN) and embedded systems. A real-time emotion detection system based on emotive expression analysis was constructed through a webcam, improving existing tools by integrating color theory to assist lighting parameters in the emotion detection process. Based on the emotional state of the ADHD individuals using the system, it uses Digital Ceiling technology with PoE to change the lighting environment. An adaptive system previously piloted with patients with ADHD is able to facilitate a non-disruptive environment while jointly improving emotional and attentional dynamics. Drawbacks include the reliance on good quality video footage and the possibility of varying accuracy due to individual differences in facial expressions. However, despite these limitations, the results highlight the promise of combining neural networks with lighting technology to create therapeutic environments. This work adds to the emerging intersection of neuroscience, embedded systems, and color theory, providing new insights into the role of environmental modulation in treating ADHD symptoms.