Skip to main content
NCBI home page
Frontiers in Human Neuroscience logoLink to Frontiers in Human Neuroscience
. 2024 Jun 4;18:1426055. doi: 10.3389/fnhum.2024.1426055

ChatGPT and BCI-VR: a new integrated diagnostic and therapeutic perspective for the accurate diagnosis and personalized treatment of mild cognitive impairment

Yiduo Yao 1, W Z W Hasan 1,*, Wenlong Jiao 2, Xianling Dong 3, H R Ramli 1, N M H Norsahperi 1, Dong Wen 2,*
PMCID: PMC11183516  PMID: 38895167

1 Introduction

The large language model, ChatGPT (OpenAI, 2024), launched by OpenAI on November 30, 2022, rapidly propelled it into the world's fastest-growing consumer software applications, attracting over a hundred million users. This milestone not only marks a significant achievement in artificial intelligence (AI) but also has sparked widespread global interest in the potential applications of AI, particularly in the medical domain (Biswas, 2023; Wójcik et al., 2023). Especially in areas such as diagnostic assistance, treatment planning and recommendations, clinical practice and guidelines, clinical research and data analysis, telemedicine, and medical resource allocation (Ferdush et al., 2023), ChatGPT has shown immense potential.

The accurate diagnosis and effective treatment of Mild Cognitive Impairment (MCI) are fraught with numerous challenges, including the high costs and invasive nature of traditional diagnostic methods, side effects associated with treatment plans, and the limitations of resources and time for non-pharmacological interventions (Aniwattanapong, 2021; Reynolds et al., 2021). Given these challenges, improving the accuracy and efficiency of MCI diagnoses and optimizing patient treatment plans have emerged as crucial issues needing resolution in this field.

Simultaneously, the rapid advancement of Brain-Computer Interface (BCI) and Virtual Reality (VR) technologies offers new avenues for addressing these issues. These technologies not only make the diagnostic and treatment processes more convenient and intelligent but also enable BCI-VR-based diagnostic and therapeutic technologies to provide more rapid and accessible methods of diagnosis and assessment for patients with cognitive impairments (Lee et al., 2013).

This paper explores how BCI-VR technologies and ChatGPT can be integrated within medical applications, aiming to propose a comprehensive diagnostic and treatment framework specifically designed for patients with cognitive impairments. By integrating BCI, VR, and ChatGPT, this platform not only seeks to provide patients with new pathways for remote diagnostics, treatment, and follow-up visits but also aims to establish a scientific and systematic framework for comprehensively evaluating the benefits, potential risks, and challenges inherent in technological integration, paving the way for future advances in the diagnosis and treatment of cognitive impairments.

2 Basic overview of the diagnosis and therapeutic of MCI

MCI typically involves early-stage changes in cognition that precede more severe neurodegenerative disorders. These changes are often characterized by synaptic dysfunction and abnormal protein accumulations, such as amyloid-beta plaques and tau protein tangles, which affect memory and cognitive functions (Tremblay et al., 2007; Canuet et al., 2015; Gonzales et al., 2018; Yin et al., 2021; Contador et al., 2023; Carvalho et al., 2024).

MCI serves as a critical transitional stage between normal aging and dementia. Diagnosis involves clinical observation (Rodriguez-Santiago et al., 2024), neuroimaging (Chouliaras and O'Brien, 2023), biomarkers (An et al., 2024; Carpi et al., 2024; Kim S. et al., 2024; Kim S. K. et al., 2024; Perneczky et al., 2024; Plassman et al., 2024; Wu et al., 2024; Zuo et al., 2024), and electroencephalography (EEG) (Amato et al., 2024; Faghfouri et al., 2024; Khatun et al., 2024; Rutkowski et al., 2024; Sun et al., 2024), as well as neuropsychological assessments.

However, each diagnostic method has its limitations. Cognitive scales, while practical, are vulnerable to subjective inaccuracies (Chouliaras and O'Brien, 2023; Tsushima et al., 2024). Biomarkers, although precise, are invasive and expensive (An et al., 2024; Carpi et al., 2024). EEG can reveal subtle brain function changes but requires specialized expertise for analysis (Meghdadi et al., 2021). These challenges necessitate comprehensive data analysis, which is often laborious.

MCI treatment includes pharmacological and non-pharmacological interventions. Pharmacological methods can alleviate symptoms but are costly and carry adverse side effects (Kho et al., 2021; Nguyen et al., 2021; dos Santos Moreira et al., 2022; Zhang et al., 2022; Yu et al., 2023). Non-pharmacological treatments like cognitive training and lifestyle changes face resource and time limitations (Aniwattanapong, 2021; Reynolds et al., 2021).

Remote cognitive training and rehabilitation offer convenience but often lack sufficient rigor and structure (Niama Natta et al., 2015; Gately et al., 2019). As a result, improved diagnostic and treatment solutions that are both precise and individualized are urgently needed.

3 Current applications status of BCI-VR in MCI diagnosis and treatment

The combination of BCI and VR technologies has sparked a wave of technological innovation in the medical field, particularly demonstrating the immense value of their application in diagnosing and treating cognitive impairments. BCI technology interprets brain activity to generate control signals, while VR technology provides an immersive experience, showing practical potential in diagnosing and treating patients with MCI.

Research on diagnostic and therapeutic approaches for spatial orientation impairments has shown that targeted VR games integrated with BCI technology can assess and enhance patients' spatial cognition. For instance, simulations of supermarkets (Zygouris et al., 2017), hospital environments (Jiang and Li, 2007), and road navigation systems (Liu et al., 2022) provide real-time neural feedback, adapting interventions to individual needs.

In virtual life scenes and simulated driving tasks, BCI allows precise evaluation and targeted enhancement of episodic memory (Bellassen et al., 2012; Sauzéon et al., 2016). Similarly, with BCI data, VR simulations of daily environments like buses and kitchens facilitate personalized cognitive rehabilitation (Richard et al., 2010; Yamaguchi et al., 2012; Allain et al., 2014; Vallejo et al., 2017).

For motor and balance impairments, BCI-VR tasks, such as simulated cycling, provide neural monitoring for tailored training, revealing their value in improving executive functions and memory (Arciero et al., 2020).

Integrating BCI and VR technologies allows for a direct connection to the patient's nervous system and the creation of immersive therapeutic environments. However, the BCI-VR system cannot directly analyze the patient's condition nor adjust the treatment plan in real-time. By incorporating large language models into the diagnosis and treatment of MCI, it would be possible to enhance the accuracy of MCI diagnoses further and provide patients with more direct and personalized diagnostic and therapeutic methods.

4 Current application status of ChatGPT in MCI diagnosis and treatment

Integrating large-scale language models into BCI applications represents a significant advancement in neurotechnology. These models possess sophisticated computational abilities that can improve the interpretation and utilization of EEG signals. Accurate decoding and transformation of EEG data are crucial for effective brain-device communication, and large-scale language models, with their advanced pattern recognition and machine learning algorithms, are uniquely positioned to enhance this process. Learning from vast neural datasets enables the development of highly accurate and responsive BCI systems, which are essential for assistive technologies and advanced neurofeedback systems.

In the field of BCI applications, large-scale language models such as GPT (Cui et al., 2023; Kim J. W. et al., 2024), BERT (Devlin et al., 2018; Kostas et al., 2021), and Seq2Seq (Sutskever et al., 2024) have demonstrated their unique characteristics and applicability. The GPT series is renowned for generating coherent and contextually relevant text outputs, showcasing a deep understanding of complex contexts. However, its direct application in decoding electroencephalography signals is limited, requiring reliance on external conversion systems (Cui et al., 2023; Kim J. W. et al., 2024). In contrast, while the BERT series excels in text comprehension, it has comparatively weaker capabilities in generating text for specific BCI tasks, necessitating model fine-tuning for optimization (Ethayarajh, 2019; Rogers et al., 2021). Seq2Seq models can directly transform EEG signals into text sequences, supporting the realization of more natural language dialogue systems but face challenges in managing long-distance dependencies and high training complexity (Luong et al., 2015; Keneshloo et al., 2019; Shi et al., 2021; Wang et al., 2022).

Considering these models' performance, advantages, and disadvantages, this study has chosen to investigate the ChatGPT model further, finding it to exhibit significant potential in the medical field. The application of ChatGPT across various medical subdomains, including neurodegenerative diseases (Koga et al., 2023), neurosurgical procedures (Liu et al., 2023), psychiatric diagnoses, and oncology (Haemmerli et al., 2023; Li et al., 2023; Schulte, 2023; Gosak et al., 2024; Yang et al., 2024), has proven effective in assisting diagnoses and offering treatment suggestions (Au and Yang, 2023; Delsoz et al., 2023; Haemmerli et al., 2023; Yeo et al., 2023). This model, integrated with deep learning techniques, can analyze EEG or MRI data (Boßelmann et al., 2023), identifying subtle changes, thus assisting physicians in conducting comprehensive risk assessments and developing personalized treatment plans (Abani et al., 2023; Zheng et al., 2024). Furthermore, through interaction with patients, ChatGPT can monitor changes in cognitive status and provide personalized cognitive training and psychological support, offering substantial support for treating cognitive disorders such as dementia (Zheng et al., 2024). Notably, ChatGPT, with its efficient capabilities in parsing and generating human language, has shown proficiency in accurately interpreting medical literature and engaging in empathetic patient communication (Dave et al., 2023; Ferdush et al., 2023) significantly boosting the auxiliary diagnostic and therapeutic potential for MCI.

The use of ChatGPT in diagnosing and treating MCI signifies a significant leap in AI within healthcare, improving diagnostic precision, treatment approaches, and patient interaction (Wójcik et al., 2023). ChatGPT's success points to the vast potential for advanced language models in medicine, especially alongside technologies like BCI and VR. This interdisciplinary approach innovates MCI management and paves the way for efficient, personalized healthcare solutions.

5 Our viewpoint: new perspective on the fusion of ChatGPT and BCI-VR systems in the diagnosis and treatment of MCI

ChatGPT, as a sophisticated language model, is adept at monitoring nuanced changes in linguistic patterns that indicate cognitive decline, such as reduced lexical diversity, simplified syntax, increased speech latency, and disjointed semantic coherence. It interacts with patients by acting as a conversational agent to engage in dialogue, analyze symptom reports, and guide them through cognitive tasks, which are crucial for assessing cognitive functions like memory and executive skills. This ongoing assessment is less invasive and can be conducted in real time, allowing ChatGPT to establish a linguistic baseline and monitor deviations over time. The collected data are analyzed using advanced machine learning techniques to detect trends that may predict cognitive impairment progression, offering a valuable tool for early diagnosis and continuous monitoring in cognitive healthcare.

Integrating ChatGPT into a cognitive disorder diagnosis and treatment system based on BCI and VR technologies showcases a new perspective on technological innovation. This system offers a multifaceted integrated solution by combining the immersive experience of VR, the precise brain activity-capturing capabilities of BCI, and ChatGPT's advanced natural language processing (NLP) functions. Such technological integration not only innovates in treatment methodologies but also represents a qualitative leap at the technological level, bringing cognitive training and rehabilitation activities closer to complex real-world scenarios. Moreover, the application potential of this system spans patient diagnostic assistance, personalized treatment plan formulation, and data analysis for doctors (Ferdush et al., 2023), providing comprehensive support and fresh perspectives for the diagnosis and treatment of cognitive disorders.

As depicted in Figure 1 (Visio, 2021), the system harnesses BCI and VR technologies to immerse patients in realistic virtual environments where cognitive testing is conducted. Patients don a BCI-VR helmet that facilitates the real-time capture and analysis of neural signals associated with specific cognitive functions. This integration allows for precise monitoring of brain activity during VR-based cognitive exercises.

Figure 1.

Figure 1

Open in a new tab

MCI diagnosis and therapeutic platform based on ChatGPT and BCI-VR.

The system utilizes ChatGPT's advanced NLP capabilities to dynamically interpret patients' verbal and behavioral responses. This interaction augments diagnostic accuracy and enhances patient engagement by providing immediate and personalized feedback. For instance, ChatGPT can elucidate test results on the fly and offer tailored medical advice based on the patient's performance and responses during the VR activities.

In addition to diagnostic applications, the platform incorporates therapeutic tools, such as games and music, designed to bolster cognitive function and emotional wellbeing. These tools are accessible within the VR environment, enabling social interactions through multiplayer setups that allow patients to connect with family, friends, or other patients, thereby supporting their social and emotional health.

Furthermore, the platform integrates a telemedicine feature that facilitates direct communication between patients and healthcare providers. This feature supports real-time, face-to-face interactions within the virtual space, enabling physicians to make informed decisions about treatment strategies based on comprehensive data analysis. The system's capacity to merge detailed EEG data analysis with behavioral observations ensures that treatment plans are precise and personalized, reflecting the system's overarching aim to enhance cognitive impairment therapies' accuracy and efficacy.

6 Risks and challenges

The integration of VR and BCI technologies, enhanced by large-scale language models like ChatGPT, epitomizes the forefront of technological innovation in cognitive diagnostics and therapy. While these technologies offer substantial benefits in medical rehabilitation, entertainment, and assistive devices for disabled people and promise to improve the quality of life and rehabilitation outcomes for cognitively impaired patients, their full potential is tempered by significant challenges.

Seamless technical integration demands interdisciplinary collaboration to ensure compatibility across diverse operating systems and hardware platforms (Putze et al., 2020). User acceptance is complicated by adaptation barriers and trust issues, particularly among the elderly, who may also experience discomfort from VR and reservations about BCI technologies.

Data privacy and security are paramount, given the sensitivity of EEG signals and behavioral data captured in VR environments. Robust encryption and stringent security protocols are essential to safeguard patient data (Li et al., 2015). Furthermore, maintaining accuracy and reliability in long-term applications poses challenges that impact the stability and effectiveness of these diagnostic and therapeutic systems.

Regulatory and ethical considerations are crucial, requiring careful management of informed consent, privacy rights, and user autonomy while minimizing the risk of BCI technology misuse (Burwell et al., 2017; Coin, 2020). Addressing these issues necessitates a concerted effort involving technological innovation, user education, and rigorous clinical research, ensuring that the potential of BCI-VR systems in cognitive health care is realized responsibly and effectively.

7 Conclusion

Through a comprehensive literature analysis, this study has uncovered the potential application value of ChatGPT in treating cognitive disorders, mainly through its integration with BCI and VR systems, proposing innovative solutions for diagnosing and treating cognitive disorders. Initially, the study assessed the limitations of traditional diagnostic and treatment methods for cognitive disorders, followed by a detailed analysis of the application of BCI-VR technology and ChatGPT in this field. It explored the feasibility and potential trends of integrating ChatGPT with BCI-VR systems in practical applications. Despite some challenges and gaps in the real-world application of this technological integration, this research suggests that BCI-VR systems incorporating large language models like ChatGPT have broad application prospects.

Author contributions

YY: Writing – original draft, Writing – review & editing. WH: Supervision, Writing – review & editing. WJ: Methodology, Writing – review & editing. XD: Writing – review & editing. HR: Writing – review & editing. NN: Writing – review & editing. DW: Supervision, Writing – review & editing, Funding acquisition.

Funding Statement

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported in part by National Key Research and Development Program of China (2021YFF1200603 and 2023YFF1203702) and National Natural Science Foundation of China (62276022 and 62206014).

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

  1. Abani S., De Decker H. A., Tipold S., Nessler A., Volk J. N. (2023). Can ChatGPT diagnose my collapsing dog? Front. Vet. Sci. 10:1245168. 10.3389/fvets.2023.1245168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allain P., Foloppe D. A., Besnard J., Yamaguchi T., Etcharry-Bouyx F., Gall D. L., et al. (2014). Detecting everyday action deficits in Alzheimer's disease using a nonimmersive virtual reality kitchen. J. Int. Neuropsychol. Soc. 20, 468–477. 10.1017/S1355617714000344 [DOI] [PubMed] [Google Scholar]
  3. Amato L. G., Vergani A. A., Lassi M., Fabbiani C., Mazzeo S., Burali R., et al. (2024). Personalized modeling of Alzheimer's disease progression estimates neurodegeneration severity from EEG recordings. Alzheimer's Dementia Diag. Assessment Dis. Monitoring 16:12526. 10.1002/dad2.12526 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. An J., Kim K., Lim H. J., Kim H. Y., Shin J., Park I., et al. (2024). Early onset diagnosis in Alzheimer's disease patients via amyloid-β oligomers-sensing probe in cerebrospinal fluid. Nat. Commun. 15:18. 10.1038/s41467-024-44818-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Aniwattanapong D. (2021). Effects of cognitive rehabilitation training in elderly with mild cognitive impairment a randomized controlled trial. Eur. Psychiatr. 64, S424–S424. 10.1192/j.eurpsy.2021.1131 [DOI] [Google Scholar]
  6. Arciero P. J., Anderson-Hanley C. A. Y., Arciero J. P., Dunnam M., Panos S. E., Saulnier E., et al. (2020). Neuro-Exergaming for the Prevention and Remediation of Decline Due to Alzheimer's Disease and Related Dementias: A Clinical Trial of the Interactive Physical and Cognitive Exercise System (Ipaces V3). [In English]. Awarded Grant. NIH RePORTER. [Google Scholar]
  7. Au K., Yang W. (2023). Auxiliary use of ChatGPT in surgical diagnosis and treatment. Int. J. Surg. 109, 3940–3943. 10.1097/JS9.0000000000000686 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bellassen V., Iglói K., de Souza L., Dubois L. C., Rondi-Reig B. (2012). Temporal order memory assessed during spatiotemporal navigation as a behavioral cognitive marker for differential Alzheimer's diagnosis. J. Neurosci. 32, 1942–1952. 10.1523/JNEUROSCI.4556-11.2012 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Biswas S. S. (2023). Role of chat GPT in public health. Annal. Biomed. Eng. 51, 868–869. 10.1007/s10439-023-03172-7 [DOI] [PubMed] [Google Scholar]
  10. Boßelmann C. M., Leu C., Lal D. (2023). Are AI language models such as ChatGPT ready to improve the care of individuals with epilepsy? Epilepsia 64, 1195–1199. 10.1111/epi.17570 [DOI] [PubMed] [Google Scholar]
  11. Burwell S., Sample M., Racine E. (2017). Ethical aspects of brain computer interfaces: a scoping review. BMC Med. Ethics 18, 1–11. 10.1186/s12910-017-0220-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Canuet L., Pusil S., López M. E., Bajo R., Pineda-Pardo J. Á., Cuesta P., et al. (2015). Network disruption and cerebrospinal fluid amyloid-beta and phospho-tau levels in mild cognitive impairment. J. Neurosci. 35, 10325–10330. 10.1523/JNEUROSCI.0704-15.2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Carpi M., Fernandes M., Mercuri N. B., Liguori C. (2024). Sleep biomarkers for predicting cognitive decline and Alzheimer's disease: a systematic review of longitudinal studies. JAD 97, 121–143. 10.3233/JAD-230933 [DOI] [PubMed] [Google Scholar]
  14. Carvalho C. M., de Andrade E., Poltronieri K. R., de Oliveira B. C., Ferreira Y. G., Woodruff R. G., et al. (2024). Translation and validation of the audiovisual version of the Montreal cognitive assessment in older adults in Brazil. BMC Geriatrics 24:553. 10.1186/s12877-023-04553-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Chouliaras L., O'Brien J. T. (2023). The use of neuroimaging techniques in the early and differential diagnosis of dementia. Mol. Psychiatr. 28, 4084–4097. 10.1038/s41380-023-02215-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Coin A. (2020). Ethical aspects of BCI technology: what is the state of the art? Philosophies 5:31. 10.3390/philosophies5040031 [DOI] [Google Scholar]
  17. Contador J., Vargas-Martínez A. M., Sánchez-Valle R., Trapero-Bertran M., Lladó A. (2023). Cost-effectiveness of Alzheimer's disease CSF biomarkers and amyloid-PET in early-onset cognitive impairment diagnosis. Eur. Arch. Psychiatr. Clin. Neurosci. 273, 243–252. 10.1007/s00406-022-01439-z [DOI] [PubMed] [Google Scholar]
  18. Cui W., Jeong W., Thölke P., Medani T., Jerbi K., Joshi A. A., et al. (2023). Neuro-GPT: developing a foundation model for EEG. arXiv preprint arXiv:2311.03764. [Google Scholar]
  19. Dave T., Athaluri S. A., Singh S. (2023). ChatGPT in medicine: an overview of its applications, advantages, limitations, future prospects, and ethical considerations. Front. Artif. Int. 6:1169595. 10.3389/frai.2023.1169595 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Delsoz M., Raja H., Madadi Y., Tang A. A., Wirostko B. M., Kahook M. Y., et al. (2023). The use of ChatGPT to assist in diagnosing glaucoma based on clinical case reports. Ophthalmol. Ther. 12, 3121–3132. 10.1007/s40123-023-00805-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Devlin J., Chang M.-W., Lee K., Toutanova K. (2018). BERT: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. [Google Scholar]
  22. dos Santos Moreira N. C., Lima J. E. B. D. F., Marchiori M. F., Carvalho I., Sakamoto-Hojo E. T. (2022). Neuroprotective effects of cholinesterase inhibitors: current scenario in therapies for Alzheimer's disease and future perspectives. J. Alzheimers Dis. Rep. 6, 177–193. 10.3233/ADR-210061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ethayarajh K. (2019). “How contextual are contextualized word representations? Comparing the geometry of BERT, ELMo, and GPT-2 Embeddings,” in Conference on Empirical Methods in Natural Language Processing. 10.18653/v1/D19-100636568019 [DOI] [Google Scholar]
  24. Faghfouri A., Shalchyan V., Toor H. G., Amjad I., Niazi I. K. (2024). A tensor decomposition scheme for EEG-based diagnosis of mild cognitive impairment. Heliyon 10, e26365. 10.1016/j.heliyon.2024.e26365 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ferdush J., Begum M., Hossain S. T. (2023). ChatGPT and clinical decision support: scope, application, and limitations. Annal. Biomed. Eng. 2, 1–6. 10.1007/s10439-023-03329-4 [DOI] [PubMed] [Google Scholar]
  26. Gately M. E., Trudeau S. A., Moo L. R. (2019). In-home video telehealth for dementia management: implications for rehabilitation. Curr. Geriatr. Rep. 8, 239–249. 10.1007/s13670-019-00297-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Gonzales M. M, Insel P. S., Nelson C., Tosun D., Schöll M., Mattsson N., et al. (2018). Chronic depressive symptomatology and CSF amyloid beta and tau levels in mild cognitive impairment. Int. J. Geriatr. Psychiatr. 33, 1305–1311. 10.1002/gps.4926 [DOI] [PubMed] [Google Scholar]
  28. Gosak L., Pruinelli L., Topaz M., Štiglic G. (2024). The ChatGPT effect and transforming nursing education with generative AI: discussion paper. Nurse Educ. Prac 75:103888. 10.1016/j.nepr.2024.103888 [DOI] [PubMed] [Google Scholar]
  29. Haemmerli J., Sveikata L., Nouri A., May A., Egervari K., Freyschlag C., et al. (2023). ChatGPT in glioma adjuvant therapy decision making: ready to assume the role of a doctor in the tumour board? BMJ Health Care Inf. 30:100775. 10.1136/bmjhci-2023-100775 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jiang C. F., Li Y. S. (2007). Virtual hospital - a computer-aided platform to evaluate the sense of direction. 2007 Annual Int. Conf. IEEE Eng. Med. Biol. Soc. 16, 2361–2364. 10.1109/IEMBS.2007.4352801 [DOI] [PubMed] [Google Scholar]
  31. Keneshloo Y., Shi T., Ramakrishnan N., Reddy C. K. (2019). Deep reinforcement learning for sequence-to-sequence models. IEEE Trans. Neural Netw. Learn. Syst. 31, 2469–2489. 10.1109/TNNLS.2019.2929141 [DOI] [PubMed] [Google Scholar]
  32. Khatun S., Morshed B. I., Bidelman G. M. (2024). Monitoring disease severity of mild cognitive impairment from single-channel EEG data using regression analysis. Sensors 24:4. 10.3390/s24041054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kho J., Ioannou A., Mandal A. K. J., Missouris C. G. (2021). Donepezil induces ventricular arrhythmias by delayed repolarisation. Naunyn-Schmiedebergs Arch. Pharmacol. 394, 559–560. 10.1007/s00210-020-02028-4 [DOI] [PubMed] [Google Scholar]
  34. Kim J. W., Alaa A., Bernardo D. (2024). EEG-GPT: exploring capabilities of large language models for EEG classification and interpretation. arXiv preprint arXiv:2401.18006. [Google Scholar]
  35. Kim S., Ma X., Jeon M. J., Song S., Lee J. S., Lee J. U., et al. (2024). Distinct plasma phosphorylated-tau proteins profiling for the differential diagnosis of mild cognitive impairment and Alzheimer's disease by plasmonic asymmetric nanobridge-based biosensor. Biosensors Bioelectr. 250, 116085–116085. 10.1016/j.bios.2024.116085 [DOI] [PubMed] [Google Scholar]
  36. Kim S. K., Kim H., Kim S. H., Kim J. B., Kim L. (2024). Electroencephalography-based classification of Alzheimer's disease spectrum during computer-based cognitive testing. Sci. Rep. 14, 5252–5252. 10.1038/s41598-024-55656-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Koga S., Martin N. B., Dickson D. W. (2023). Evaluating the performance of large language models: ChatGPT and Google Bard in generating differential diagnoses in clinicopathological conferences of neurodegenerative disorders. Brain Pathol. 34:e13207. 10.1111/bpa.13207 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Kostas D., Aroca-Ouellette S., Rudzicz F. (2021). BENDR: using transformers and a contrastive self-supervised learning task to learn from massive amounts of EEG data. Front. Hum. Neurosci. 15:653659. 10.3389/fnhum.2021.653659 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lee T.-S., Goh S. J., Quek S. Y., Phillips R., Guan C., Cheung Y. B., et al. (2013). A brain-computer interface based cognitive training system for healthy elderly: a randomized control pilot study for usability and preliminary efficacy. Plos ONE. 8:419. 10.1371/journal.pone.0079419 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Li Q., Ding D., Conti M. (2015). “Brain-computer interface applications: security and privacy challenges,” in 2015 IEEE conference on Communications and Network Security (CNS). IEEE.30370454 [Google Scholar]
  41. Li Y., Gao W., Luan Z., Zhou Z., Li J. (2023). The impact of chat generative pre-trained transformer (ChatGPT) on oncology: application, expectations, and future prospects. Cureus J. Med. Sci. 15:48670. 10.7759/cureus.48670 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Liu J., Singh A. K., Lin C.-T. (2022). Predicting the quality of spatial learning via virtual global landmarks. IEEE Trans. Neural Syst. Rehab. Eng. 30, 2418–2425. 10.1109/TNSRE.2022.3199713 [DOI] [PubMed] [Google Scholar]
  43. Liu J., Zheng J., Cai X., Wu D., Yin C. (2023). A descriptive study based on the comparison of ChatGPT and evidence-based neurosurgeons. Iscience 26:107590. 10.1016/j.isci.2023.107590 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Luong M.-T., Pham H., Manning C. D. (2015). Effective approaches to attention-based neural machine translation. arXiv preprint arXiv:1508.04025.38165908 [Google Scholar]
  45. Meghdadi A. H., StevanovićKarić M., McConnell M., Rupp G., Richard C., Hamilton J., et al. (2021). Resting state EEG biomarkers of cognitive decline associated with Alzheimer's disease and mild cognitive impairment. Plos ONE 16:e0244180. 10.1371/journal.pone.0244180 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Nguyen K., Hoffman H., Chakkamparambil B., Grossberg G. T. (2021). Evaluation of rivastigmine in Alzheimer's disease. Neurodegen. Dis. Manage. 11, 35–48. 10.2217/nmt-2020-0052 [DOI] [PubMed] [Google Scholar]
  47. Niama Natta D. D., Alagnide E., Kpadonou G. T., Stoquart G. G., Detrembleur C., Lejeune T. M. (2015). Feasibility of a self-rehabilitation program for the upper limb for stroke patients in Benin. Annal. Phys. Rehab. Med. 58, 322–325. 10.1016/j.rehab.2015.08.003 [DOI] [PubMed] [Google Scholar]
  48. OpenAI . (2024). Home. Available online at: https://www.openai.com (accessed May 16, 2024).
  49. Perneczky R., Hansen N., Hofmann A., Laske C., Priller J., Grimmer T., et al. (2024). Blood-based biomarkers for early Alzheimer's disease diagnosis in real-world settings. Methods Mol. Biol. 2785, 3–14. 10.1007/978-1-0716-3774-6_1 [DOI] [PubMed] [Google Scholar]
  50. Plassman B. L., Ford C. B., Smith V. A., DePasquale N., Burke J. R., Korthauer L., et al. (2024). Elevated amyloid-beta pet scan and cognitive and functional decline in mild cognitive impairment and dementia of uncertain etiology. JAD 97, 1161–1171. 10.3233/JAD-230950 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Putze F., Vourvopoulos A., Lécuyer A., Krusienski D., Bermúdez I Badia S., Mullen T., et al. (2020). Editorial: brain-computer interfaces and augmented/virtual reality. Front. Hum. Neurosci. 14:144. 10.3389/fnhum.2020.00144 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Reynolds G. O., Willment K., Gale S. (2021). Mindfulness and cognitive training interventions in mild cognitive impairment: impact on cognition and mood. Am. J. Med. 134, 444–455. 10.1016/j.amjmed.2020.10.041 [DOI] [PubMed] [Google Scholar]
  53. Richard P., Massenot L., Besnard J., Richard E., Le Gall D., Allain P. (2010). “A virtual kitchen to assess the activities of daily life in Alzheimer's disease,” in GRAPP 2010: Proceedings of the International Conference on Computer Graphics Theory and Applications, 378–383. [Google Scholar]
  54. Rodriguez-Santiago M. A., Wojna V., Miranda-Valentin E., Arnold S., Sepulveda-Rivera V. (2024). Diagnosing Alzheimer's disease: Which dementia screening test to use in elderly Puerto Ricans with mild cognitive impairment and early Alzheimer's disease? Alzheimer's Dementia Diagnosis Assessment Dis. Monitoring 16:12554. 10.1002/dad2.12554 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Rogers A., Kovaleva O., Rumshisky A. (2021). A primer in BERTology: what we know about how BERT works. Trans. Assoc. Comput. Ling. 8, 842–866. 10.1162/tacl_a_00349 [DOI] [Google Scholar]
  56. Rutkowski T. M., Komendziński T., Otake-Matsuura M. (2024). Mild cognitive impairment prediction and cognitive score regression in the elderly using EEG topological data analysis and machine learning with awareness assessed in affective reminiscent paradigm. Front. Aging Neurosci. 15:1294139. 10.3389/fnagi.2023.1294139 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sauzéon H., N'Kaoua B., Pala P. A., Taillade M., Auriacombe S., Guitton P. (2016). Everyday-like memory for objects in ageing and A lzheimer's disease assessed in a visually complex environment: the role of executive functioning and episodic memory. J. Neuropsychol. 10, 33–58. 10.1111/jnp.12055 [DOI] [PubMed] [Google Scholar]
  58. Schulte B. (2023). Capacity of ChatGPT to identify guideline-based treatments for advanced solid tumors. Cureus J. Med. Sci. 15:37938. 10.7759/cureus.37938 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Shi T., Keneshloo Y., Ramakrishnan N., Reddy C. K. (2021). Neural abstractive text summarization with sequence-to-sequence models. ACM Trans. Data Sci. 2, 1–37. 10.1145/3419106 [DOI] [Google Scholar]
  60. Sun J., Sun Y., Shen A., Li Y., Gao X., Lu B. (2024). An ensemble learning model for continuous cognition assessment based on resting-state EEG. NPJ Aging 10:129. 10.1038/s41514-023-00129-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Sutskever I., Vinyals Q., Le Q. V. (2024). “Sequence to sequence learning with neural networks,” in Advances in Neural Information Processing Systems, Vol. 27. [Google Scholar]
  62. Tremblay C., Pilote M., Phivilay A., Emond V., Bennett D. A., Calon F. (2007). Biochemical characterization of Aβ and tau pathologies in mild cognitive impairment and Alzheimer's disease. J. Alzheimer's Dis. 12, 377–390. 10.3233/JAD-2007-12411 [DOI] [PubMed] [Google Scholar]
  63. Tsushima W. T., Abe J. R., Siu A. M. (2024). The memory assessment scale: a population-based cognitive impairment screening instrument. Hawai'i J. Health Soci. Welfare 83, 75–80. 10.62547/JSKG9101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Vallejo V., Wyss P., Rampa L., Mitache A. V., Müri R. M., Mosimann U. P., et al. (2017). Evaluation of a novel Serious Game based assessment tool for patients with Alzheimer's disease. Plos ONE 12:e0175999. 10.1371/journal.pone.0175999 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Visio M. (2021). Icon. Redmond, WA: Microsoft Corporation. [Google Scholar]
  66. Wang W., Jiao W., Hao Y., Wang X., Shi S., Tu Z., et al. (2022). Understanding and improving sequence-to-sequence pretraining for neural machine translation. arXiv preprint arXiv:2203.08442. 10.18653/v1/2022.acl-long.18536568019 [DOI] [Google Scholar]
  67. Wójcik S., Rulkiewicz A., Pruszczyk P., Lisik W., Poboży M., Domienik-Karłowicz J. (2023). Beyond ChatGPT: what does GPT-4 add to healthcare? The dawn of a new era. Cardiology J. 30, 1018–1025. 10.5603/cj.97515 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Wu L., Arvai S., Wang S. H. J., Liu A. J., Xu B. (2024). Differential diagnosis of mild cognitive impairment of Alzheimer's disease by Simoa p-tau181 measurements with matching plasma and CSF. Front. Mol. Neurosci. 16:1288930. 10.3389/fnmol.2023.1288930 [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Yamaguchi T., Foloppe D. A., Richard P., Richard E., Allain P. (2012). A dual-modal virtual reality kitchen for (re) learning of everyday cooking activities in Alzheimer's Disease. Pres.-Teleoperators Virtual Environ. 21, 43–57. 10.1162/PRES_a_00080 [DOI] [Google Scholar]
  70. Yang F., Yan D., Wang Z. (2024). Large-Scale assessment of ChatGPT's performance in benign and malignant bone tumors imaging report diagnosis and its potential for clinical applications. J. Bone Oncol. 44:100525. 10.1016/j.jbo.2024.100525 [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Yeo Y. H., Samaan J. S., Ng W. H., Ting P. S., Trivedi H., Vipani A., et al. (2023). Assessing the performance of ChatGPT in answering questions regarding cirrhosis and hepatocellular carcinoma. Clin. Mol. Hepatol. 29:721. 10.3350/cmh.2023.0089 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Yin X., Qiu Y., Zhao C., Zhou Z., Bao J., Qian W. (2021). The role of amyloid-beta and tau in the early pathogenesis of Alzheimer's Disease. Med. Sci. Monitor Int. Med. J. Exp. Clin. Res. 27, e933084–933081. 10.1038/s41392-023-01484-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Yu S. J., Tang H. L., Li W. H., Bin C. L., Liu Z., Tang Z. H., et al. (2023). Donepezil combined with traditional Chinese medicine has promising efficacy on mild cognitive impairment: a systematic review and meta-analysis. Front. Neurosci. 17:1206491. 10.3389/fnins.2023.1206491 [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Zhang X., Lian S., Zhang Y., Zhao Q. (2022). Efficacy and safety of donepezil for mild cognitive impairment: a systematic review and meta-analysis. Clin. Neurol. Neurosurg. 213:107134. 10.1016/j.clineuro.2022.107134 [DOI] [PubMed] [Google Scholar]
  75. Zheng Y., Sun X., Feng B., Kang K., Yang Y., Zhao A., et al. (2024). Rare and complex diseases in focus: ChatGPT's role in improving diagnosis and treatment. Front. Artif. Int. 7:1338433. 10.3389/frai.2024.1338433 [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Zuo L., Dong Y., Liao X., Hu Y., Pan Y., Yan H., et al. (2024). Low HALP (Hemoglobin, Albumin, Lymphocyte, and Platelet) score increases the risk of post-stroke cognitive impairment: a multicenter cohort study. Clin. Interv. Aging 19:885. 10.2147/CIA.S432885 [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Zygouris S., Ntovas K., Giakoumis D., Votis K., Doumpoulakis S., Segkouli S., et al. (2017). A preliminary study on the feasibility of using a virtual reality cognitive training application for remote detection of mild cognitive impairment. J. Alzheimer's Dis. 56, 619–627. 10.3233/JAD-160518 [DOI] [PubMed] [Google Scholar]

Articles from Frontiers in Human Neuroscience are provided here courtesy of Frontiers Media SA

RESOURCES