| Facial expression (FE) is the most natural and convincing source to communicate human emotions, providing valuable insides to the observer while assessing the emotional incongruities. In health care, the FE of the patient (specifically of neurological disorders (NDs) such as Parkinson's, Stroke, and Alzheimer's) can assist the medical doctor in evaluating the physical condition of a patient, such as fatigue, pain, and sadness. ND patients are usually going through proper observation and clinical tests, which are invasive, expensive and time-consuming. In this paper, an automatic lightweight deep learning (DL) based FEs recognition framework is developed that can classify the facial expression of ND patients with 93% accuracy. Initially, raw images of FEs are acquired from publicly available datasets according to the patient's most common expressions, such as normal, happy, sad, and anger. The framework cropped images through a face detector, extract high-level facial features through the convolutional layers and fed them to the dense layers for classification. The trained model is exported to an android based environment over a smart device and evaluated for real-time performance. The qualitative and quantitative results are evaluated on a standard dataset named Karolinska directed emotional faces (KDEF). Promising results are obtained of various NDs patients with Parkinson, Stroke, and Alzheimer that show the effectiveness of the proposed model. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.