Elderly individuals with dementia experience significant cognitive and emotional impairments, motivating research into technology-driven therapies to improve their quality of life. However, the effectiveness of such systems is often limited by poor-quality electroencephalographic (EEG) data, which can be distorted or incomplete due to artefacts. This study introduces a novel brain-computer interface (BCI)-based rehabilitation framework that combines neural network-assisted EEG data restoration with personalised therapy modules. The proposed method employs a multilayer perception (MLP) enhanced with a custom activation function to reconstruct missing EEG values by modelling spatial and temporal dependencies among adjacent electrodes. Experimental evaluation on benchmark EEG datasets shows that the proposed approach reduces Mean Absolute Error (MAE) by 15% and increases the Correlation Coefficient (CC) by 10% compared to traditional imputation techniques such as mean substitution and k-nearest neighbours (KNN). The restored EEG data are further integrated into a generative AI-powered rehabilitation system that delivers adaptive treatments through virtual reality (VR) environments and social interaction activities. By incorporating patient-specific affective profiles and preferences, the system dynamically personalises interventions such as cognitive games, reminiscence sessions, and immersive simulations. Overall, this framework bridges computational neuroscience and patient-centred healthcare, highlighting EEG imputation as a core technology for next-generation intelligent dementia care solutions, particularly in rural and resource-limited settings.

1.         Arpaia P, Coyle D, Donnarumma F, et al. Visual and haptic feedback in detecting motor imagery within a wearable brain–computer interface. Measurement. 2023; 206: 112304. doi: 10.1016/j.measurement.2022.112304

2.         Velasco-Álvarez F, Fernández-Rodríguez Á, Ron-Angevin R. Brain-computer interface (BCI)-generated speech to control domotic devices. Neurocomputing. 2022; 509: 121–136. doi: 10.1016/j.neucom.2022.08.068

3.         Orban M, Elsamanty M, Guo K, et al. A review of brain activity and EEG-based brain–computer interfaces for rehabilitation application. Bioengineering. 2022; 9(12): 768. doi: 10.3390/bioengineering9120768

4.         Kanemura A, Cheng Y, Kaneko T, et al. Imputing missing values in EEG with multivariate autoregressive models. 2018 40th annual international conference of the IEEE engineering in medicine and biology society (EMBC). 2018: 2639–2642. doi: 10.1109/embc.2018.8512790

5.         Mack C, Su Z, Weistreich D. Managing missing data in patient registries. Agency for Healthcare Research and Quality (AHRQ); 2018. doi: 10.23970/ahrqregistriesmissingdata

6.         SKhairuddin IMohd, Abdullah MA, Ab. Nasir AF, et al. Enabling industry 4.0 through advances in mechatronics. Springer Nature Singapore; 2022. doi: 10.1007/978-981-19-2095-0

7.         Bansal D, Mahajan R. EEG-Based brain-computer interfacing (BCI). EEG-based brain-computer interfaces. 2019: 2171. doi: 10.1016/b978-0-12-814687-3.00002-8

8.         Glas CAW. International encyclopedia of education, 3rd ed. Elsevier; 2010. pp. 283–288.

9.         Singh SP, Pritamdas K, Devi KJ, et al. Custom convolutional neural network for detection and classification of rice plant diseases. Procedia Computer Science. 2023; 218: 2026–2040. doi: 10.1016/j.procs.2023.01.179

10.      Barbera T, Burger J, D’Amelio A, et al. On using AI for EEG-based BCI applications: Problems, current challenges and future trends. International Journal of Human–Computer Interaction. 2025: 1–20. doi: 10.1080/10447318.2025.2561185

11.      Wahul RM, Ambadekar S, Dhanvijay DM, et al. Multimodal approaches and AI-driven innovations in dementia diagnosis: a systematic review. Discover artificial intelligence. 2025; 5(1). doi: 10.1007/s44163-025-00358-x

12.      Jiao D. AI-Enhanced digital therapeutics for cognitive impairment: Integrating mobile applications, virtual reality, and wearable devices. discover artificial intelligence. 2025; 5(1). doi: 10.1007/s44163-025-00325-6

13.      Noor NM, Al Bakri Abdullah MM, Yahaya AS, et al. Comparison of linear interpolation method and mean method to replace the missing values in environmental data set. Materials Science Forum. 2014; 803: 278–281. doi: 10.4028/www.scientific.net/msf.803.278

14.      Hippert-Ferrer A, El Korso MN, Breloy A, et al. Robust low-rank covariance matrix estimation with a general pattern of missing values. Signal Processing. 2022; 195: 108460. doi: 10.1016/j.sigpro.2022.108460

15.      Lin J, Li N, Alam MA, et al. Data-driven missing data imputation in cluster monitoring system based on deep neural network. Applied Intelligence. 2019; 50(3): 860–877. doi: 10.1007/s10489-019-01560-y

16.      Cheng CY, Tseng WL, Chang CF, et al. A deep learning approach for missing data imputation of rating scales assessing attention-deficit hyperactivity disorder. Frontiers in Psychiatry. 2020; 11. doi: 10.3389/fpsyt.2020.00673

17.      Wang H, Tang J, Wu M, et al. Application of machine learning missing data imputation techniques in clinical decision making: taking the discharge assessment of patients with spontaneous supratentorial intracerebral hemorrhage as an example. BMC Medical Informatics and Decision Making. 2022; 22(1). doi: 10.1186/s12911-022-01752-6

18.      Emmanuel T, Maupong T, Mpoeleng D, et al. A survey on missing data in machine learning. Journal of Big Data. 2021; 8(1). doi: 10.1186/s40537-021-00516-9

19.      Hosseini S, Guo X. Deep convolutional neural network for automated detection of mind wandering using eeg signals. Proceedings of the 10th acm international conference on bioinformatics, computational biology and health informatics. 2019: 314–319. doi: 10.1145/3307339.3342176

20.      Gardner MW, Dorling SR. Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmospheric Environment. 1998; 32(14–15): 2627-2636. doi: 10.1016/S1352-2310(97)00447-0

21.      Delorme A. Publicly available EEG/ERP data. Available online: http://sccn.ucsd.edu/~arno/fam2data/publicly_available _EEG_data.html (accessed on 3 August 2025).

22.      García S, Fernández A, Luengo J, et al. A study of statistical techniques and performance measures for genetics-based machine learning: accuracy and interpretability. Soft Computing. 2008; 13(10): 959–977. doi: 10.1007/s00500-008-0392-y

23.      Popescu MC, Balas VE, Perescu-Popescu L, et al. Multilayer Perceptron and Neural Networks. WSEAS Transactions on Circuits and Systems. 2009; 8(7): 579–588. doi: 10.5555/1639537.1639542

24.      Rojas R. Neural Networks. Springer berlin heidelberg; 1996. doi: 10.1007/978-3-642-61068-4

25.      Huisman M. Imputation of missing item responses: some simple techniques. Quality & Quantity. 2000; 34: 331–351. doi: 10.1023/A:1004782230065