Facial Image Analysis for Parkinson’s Identification in Low-Data Contexts

• José Nazareno A. Rodrigues UFPI
• Caio B. A. A. Pereira UFPI
• Rodrigo de M. S. Veras UFPI
• Pedro H. X. R. Barros UFPI
• Kelson J. Almeida UFPI
• Kelson Rômulo T. Aires UFPI

DOI: https://doi.org/10.5753/sbcas.2025.7107

Abstract

This study investigates the use of 2D facial images to identify Parkinson’s Disease (PD) in data-limited scenarios. The exploration of facial expressions in 2D images remains scarce in the literature. To help bridge this gap, a complete pipeline was implemented. Predefined facial feature extraction approaches were compared with automatic extraction using pre-trained convolutional neural network models. Among the evaluated models, the combination of DenseNet-201 for feature extraction and SVM for classification achieved the best performance, reaching 94.75% accuracy. The results reinforce the potential of 2D facial image analysis for PD identification, despite challenges related to data scarcity.

References

Bey, R., Goussault, R., Grolleau, F., Benchoufi, M., and Porcher, R. (2020). Fold-stratified cross-validation for unbiased and privacy-preserving federated learning. Journal of the American Medical Informatics Association, 27(8):1244–1251.

Breiman, L. (2001). Random forests. Machine learning, 45(1):5–32. [Guan 2021] Guan, Y. (2021). Application of logistic regression algorithm in the diagnosis of expression disorder in parkinson’s disease. In 2021 IEEE 2nd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA), volume 2, pages 1117–1120. IEEE.

Hastie, T., Tibshirani, R., and Friedman, J. (2009). The elements of statistical learning: data mining, inference, and prediction. Springer Science & Business Media.

Jin, B., Qu, Y., Zhang, L., and Gao, Z. (2020). Diagnosing parkinson disease through facial expression recognition: video analysis. Journal of Medical Internet Research, 22(7):e18697.

Moustafa, K., Metawie, H., Saadoun, S., Sameh, N., Ibrahim, R., and Abdelsayed, A. (2023). Easy park: Mobile application for parkinson’s disease detection and severity level. In 2023 International Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC), pages 1–8. IEEE.

Pereira, C., Barros, P., Rodrigues, J., Araújo, P., Borges, R., Almeida, K., and Veras, R. (2024). Identificação de parkinson em imagens faciais usando modelos de deep learning pré-treinados. In Anais da XII Escola Regional de Computação do Ceará, Maranhão e Piauí, pages 169–178, Porto Alegre, RS, Brasil. SBC.

Rajnoha, M., Mekyska, J., Burget, R., Eliasova, I., Kostalova, M., and Rektorova, I. (2018). Towards identification of hypomimia in parkinson’s disease based on face recognition methods. In 2018 10th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), pages 1–4. IEEE.

Rodrigues, J. N. A. and Aires, K. R. T. (2021). Uma pesquisa exploratória sobre a utilização de soluções apoiadas em inteligência artificial e tecnologias móveis com portadores de transtorno do espectro autista. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 410–415. SBC.

Rodrigues, J. N. A., Veras, R. d. M. S., Neto, L. d. S. B., Barros, P. H. X. R., Moura, W. d. S., De Almeida, K. J. S., and Aires, K. R. T. (2025). Identification of parkinson’s disease through facial image classification: A systematic review. IEEE Access, pages 46720–46731.

Skaramagkas, V., Pentari, A., Kefalopoulou, Z., and Tsiknakis, M. (2023). Multi-modal deep learning diagnosis of parkinson’s disease-a systematic review. IEEE Transactions on Neural Systems and Rehabilitation Engineering.

Skibińska, J. and Burget, R. (2020). Parkinson’s disease detection based on changes of emotions during speech. In 2020 12th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), pages 124–130. IEEE.

Su, G., Lin, B., Yin, J., Luo, W., Xu, R., Xu, J., and Dong, K. (2021). Detection of hypomimia in patients with parkinson’s disease via smile videos. Annals of Translational Medicine, 9(16).

Warnakulasuriya, N., De Silva, S., Madushika, J., Gamage, S., Jayawardena, S., and Karunasena, A. (2023). Multimodal fusion for enhanced parkinson’s disease screening: Integrating brain mri, hand drawing, facial expressions, and voice analysis. In 2023 5th International Conference on Advancements in Computing (ICAC), pages 531–536. IEEE.