Integrating Brain-Computer Interface and Neuromorphic Computing for Human Digital Twins

Chen Shang; Dinh Thai Hoang; Jiadong Yu

arxiv: 2410.23639 · v3 · pith:6SL4L5P7new · submitted 2024-10-31 · 💻 cs.HC · cs.NI

Integrating Brain-Computer Interface and Neuromorphic Computing for Human Digital Twins

Chen Shang , Jiadong Yu , Dinh Thai Hoang This is my paper

Pith reviewed 2026-05-23 19:15 UTC · model grok-4.3

classification 💻 cs.HC cs.NI

keywords Human Digital TwinsBrain-Computer InterfaceSpiking Neural NetworkNeuromorphic ComputingFederated LearningBio-inspired FrameworkData Privacy

0 comments

The pith

A bio-inspired framework uses brain signals from BCI, spiking neural networks, and federated learning to build human digital twins with reduced device heterogeneity and energy use.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper introduces a framework for human digital twins that collects data primarily from brain signals via brain-computer interfaces rather than varied sensors. The approach employs spiking neural networks to process data in a brain-like manner for lower energy consumption and incorporates federated learning to enhance privacy. By doing so, it seeks to address challenges in data heterogeneity, efficiency, and security for creating personalized digital replicas. The authors support their proposal with a case study and discuss future directions.

Core claim

The paper claims that integrating BCI sensor technology to capture brain signals as the data source for HDTs, combined with an SNN-based neuromorphic computing learning model and a FL strategy, minimizes device heterogeneity, enhances data collection efficiency, provides richer physiological and psychological data, reduces energy consumption, and strengthens data privacy.

What carries the argument

The twofold bio-inspired scheme consisting of BCI for brain signal capture and SNN-based neuromorphic computing with federated learning for processing.

If this is right

Device heterogeneity is minimized by relying on a uniform BCI data source.
Data collection becomes more efficient while yielding richer physiological and psychological information.
Energy consumption decreases due to the discrete spike processing in SNNs.
Data privacy improves through the integration of federated learning.
The framework's performance is validated through a case study.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

If the framework holds, it could reduce the need for multiple heterogeneous devices in personal health monitoring systems.
Challenges in BCI signal quality might require additional validation not covered in the case study.
Extensions to other domains like real-time psychological state tracking could follow from the richer data claim.
Scalability to large user bases would depend on the FL strategy's effectiveness in practice.

Load-bearing premise

Brain signals captured via BCI can serve as a sufficient and superior primary data source for constructing personalized HDTs without introducing substantial new problems such as signal quality or user acceptance.

What would settle it

A direct comparison experiment where traditional multi-device HDT construction outperforms the BCI-based approach in data richness, energy efficiency, or privacy metrics.

Figures

Figures reproduced from arXiv: 2410.23639 by Chen Shang, Dinh Thai Hoang, Jiadong Yu.

**Figure 2.** Figure 2: An illustration of the proposed framework for the biologically-inspired human digital twin (Bio-HDT). This process begins with the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: The proposed network architecture based on SNNs. (a) Brain signal extraction is performed by HMD equipped with EEG electrodes [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

**Figure 4.** Figure 4: Performance on various methods. (a) Energy consumption of [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗

**Figure 5.** Figure 5: The achievable WSP of various methods. V. CHALLENGES AND FUTURE RESEARCH DIRECTIONS In this section, we discuss the practical deployment challenges faced by Bio-HDT from two perspectives: brain signal processing and SNN-driven FL. Additionally, we explore future research directions based on the aforementioned analysis. A. Challenges on Processing Brain Signals The primary challenge in deploying the Bio-H… view at source ↗

read the original abstract

The integration of immersive communication into a human-centric ecosystem has intensified the demand for sophisticated Human Digital Twins (HDTs) driven by multifaceted human data. However, the effective construction of HDTs faces significant challenges due to the heterogeneity of data collection devices, the high energy demands associated with processing intricate data, and concerns over the privacy of sensitive information. This work introduces a novel biologically-inspired (bio-inspired) HDT framework that leverages Brain-Computer Interface (BCI) sensor technology to capture brain signals as the data source for constructing HDT. By collecting and analyzing these signals, the framework not only minimizes device heterogeneity and enhances data collection efficiency, but also provides richer and more nuanced physiological and psychological data for constructing personalized HDTs. To this end, we further propose a bio-inspired neuromorphic computing learning model based on the Spiking Neural Network (SNN). This model utilizes discrete neural spikes to emulate the way of human brain processes information, thereby enhancing the system's ability to process data effectively while reducing energy consumption. Additionally, we integrate a Federated Learning (FL) strategy within the model to strengthen data privacy. We then conduct a case study to demonstrate the performance of our proposed twofold bio-inspired scheme. Finally, we present several challenges and promising directions for future research of HDTs driven by bio-inspired technologies.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This is a conceptual framework proposal for BCI-SNN-FL human digital twins that asserts efficiency and privacy gains without any metrics, baselines, or implementation details.

read the letter

The core of this paper is a high-level sketch for using brain signals from BCI as the main input to build human digital twins, processed via spiking neural networks and kept private with federated learning. It flags real issues like device heterogeneity, energy use, and data sensitivity, then suggests the bio-inspired stack could address them by giving richer physiological and psychological signals while cutting power draw. That combination is the main thing on offer, and the authors do a clean job laying out why each piece might fit together for this application. The SNN angle for low-energy spike-based processing and the FL privacy layer are standard moves in their fields, but tying them specifically to HDT construction via BCI is at least a coherent synthesis. The case study is mentioned but not shown, so nothing is demonstrated. The claims about minimized heterogeneity, richer data, lower energy, and better privacy are stated as outcomes of the framework, yet no numbers, comparisons to existing HDT pipelines, or even basic signal quality checks appear. That leaves the argument resting on the premise that BCI will be sufficient and superior without evidence that signal noise, user comfort, or integration overhead won't create new problems. The paper is aimed at researchers already working on neuromorphic hardware or HCI applications for digital twins who might want a starting point for discussion. It does not contain new measurements or derivations, so it is not the kind of work that needs referee time right now. I would not recommend sending it out for review until there is at least a small implemented prototype with some concrete results.

Referee Report

2 major / 1 minor

Summary. The paper proposes a bio-inspired Human Digital Twin (HDT) framework that uses Brain-Computer Interface (BCI) sensors to capture brain signals as the primary data source for personalized HDT construction. It combines this with a Spiking Neural Network (SNN)-based neuromorphic computing model to emulate brain-like processing for reduced energy use and integrates Federated Learning (FL) to enhance privacy. The framework is claimed to minimize device heterogeneity, improve data collection efficiency, and provide richer physiological/psychological data; a case study is mentioned to demonstrate performance, followed by discussion of future challenges.

Significance. If the claimed benefits were quantitatively validated, the integration of BCI with SNN and FL could represent a meaningful advance for HDTs in human-centric ecosystems by addressing heterogeneity, energy, and privacy issues through a biologically inspired approach, potentially enabling more nuanced and efficient digital representations of human states.

major comments (2)

[Case study] Case study section: No quantitative metrics, baselines, ablation studies, error bars, datasets, or comparisons to conventional HDT pipelines are supplied to support the central claims that the BCI-SNN-FL scheme 'minimizes device heterogeneity,' 'enhances data collection efficiency,' 'reduces energy consumption,' or 'strengthens data privacy.' The performance demonstration therefore rests entirely on an undescribed high-level proposal rather than verifiable evidence.
[Introduction / Abstract] Framework introduction and abstract: The assumption that BCI-captured brain signals can serve as a sufficient primary data source that overcomes heterogeneity and efficiency issues without introducing new problems (e.g., signal quality, user acceptance, or integration overhead) is stated but not tested or modeled; no analysis of these trade-offs is provided to substantiate superiority over existing sensor-based HDT approaches.

minor comments (1)

[Abstract] The abstract and framework description use several high-level benefit statements (e.g., 'richer and more nuanced' data) without defining measurable criteria or relating them to specific SNN or FL mechanisms.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. Our manuscript proposes a conceptual bio-inspired HDT framework integrating BCI, SNN, and FL; the case study is illustrative rather than empirical. We address the comments by clarifying scope and adding discussion of limitations.

read point-by-point responses

Referee: [Case study] Case study section: No quantitative metrics, baselines, ablation studies, error bars, datasets, or comparisons to conventional HDT pipelines are supplied to support the central claims that the BCI-SNN-FL scheme 'minimizes device heterogeneity,' 'enhances data collection efficiency,' 'reduces energy consumption,' or 'strengthens data privacy.' The performance demonstration therefore rests entirely on an undescribed high-level proposal rather than verifiable evidence.

Authors: The manuscript is a framework proposal paper. The case study illustrates the architecture and potential benefits at a conceptual level without claiming empirical validation. We will revise the case study section to explicitly label it as illustrative, remove or qualify unsupported performance claims, and state that quantitative evaluation with metrics, baselines, and comparisons is reserved for future work. This clarifies the paper's scope without misrepresenting the contribution. revision: yes
Referee: [Introduction / Abstract] Framework introduction and abstract: The assumption that BCI-captured brain signals can serve as a sufficient primary data source that overcomes heterogeneity and efficiency issues without introducing new problems (e.g., signal quality, user acceptance, or integration overhead) is stated but not tested or modeled; no analysis of these trade-offs is provided to substantiate superiority over existing sensor-based HDT approaches.

Authors: We agree that trade-offs of BCI as primary source (signal quality, acceptance, overhead) are not analyzed. We will add a dedicated paragraph or subsection discussing these limitations and potential issues, referencing BCI literature, and noting that superiority is hypothesized based on design rationale rather than tested. This provides balance while retaining the core proposal. revision: yes

Circularity Check

0 steps flagged

No circularity: conceptual framework proposal with no derivations or fitted predictions

full rationale

The paper presents a high-level bio-inspired HDT architecture using BCI, SNN, and FL. It contains no equations, parameter fits, predictions, or derivation chains. Claims about minimizing heterogeneity or reducing energy are stated as expected benefits of the proposed scheme rather than results derived from internal models or self-citations. No load-bearing steps reduce to inputs by construction. The work is self-contained as an architectural proposal and case-study outline without internal circular reasoning.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The paper introduces no fitted numerical parameters. It relies on the domain assumption that BCI signals can replace heterogeneous sensor data and postulates the bio-inspired framework itself without independent evidence.

axioms (1)

domain assumption Brain signals from BCI provide richer and more nuanced physiological and psychological data than conventional heterogeneous devices
Invoked when describing the benefits of the proposed data source in the abstract.

invented entities (1)

Bio-inspired HDT framework no independent evidence
purpose: To integrate BCI, SNN, and FL for addressing HDT construction challenges
The framework is introduced as the central contribution but carries no independent falsifiable evidence outside the proposal.

pith-pipeline@v0.9.0 · 5769 in / 1387 out tokens · 36033 ms · 2026-05-23T19:15:05.307340+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

SNNs utilize discrete neural spikes to emulate the way of human brain processes information, thereby enhancing the system’s ability to process data effectively while reducing energy consumption.
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We then conduct a case study to demonstrate the performance of our proposed twofold bio-inspired scheme.

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Spiking Personalized Federated Learning for Brain-Computer Interface-Enabled Immersive Communication
cs.LG 2026-03 unverdicted novelty 6.0

Develops an SNN-integrated personalized federated learning model for BCI brain-signal analysis in immersive communication, reporting highest identification accuracy and 6.46x lower inference energy than ANN baselines.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages · cited by 1 Pith paper

[1]

Networking architecture and key supporting technologies for human digital twin in personalized healthcare: A comprehensive survey,

J. Chen, C. Yi, S. D. Okegbile, J. Cai, and X. Shen, “Networking architecture and key supporting technologies for human digital twin in personalized healthcare: A comprehensive survey,”IEEE Commun. Surv. Tutor., vol. 26, no. 1, pp. 706–746, 2024

work page 2024
[2]

Digital twin in industry: State-of-the-art,

F. Tao, H. Zhang, A. Liu, and A. Y . C. Nee, “Digital twin in industry: State-of-the-art,” IEEE Trans. Ind. Informat. , vol. 15, no. 4, pp. 2405– 2415, 2019

work page 2019
[3]

Human digital twin for personalized healthcare: Vision, architecture and future directions,

S. D. Okegbile, J. Cai, D. Niyato, and C. Yi, “Human digital twin for personalized healthcare: Vision, architecture and future directions,”IEEE Netw., vol. 37, no. 2, pp. 262–269, 2023

work page 2023
[4]

Human digital twin for fitness management,

B. R. Barricelli, E. Casiraghi, J. Gliozzo, A. Petrini, and S. Valtolina, “Human digital twin for fitness management,” IEEE Access, vol. 8, pp. 26 637–26 664, 2020

work page 2020
[5]

Development and application of Artificial Neural Network,

Y . Wu and J. Feng, “Development and application of Artificial Neural Network,” Wireless Personal Communications, vol. 102, pp. 1645–1656, 2018

work page 2018
[6]

Cognixion One: The world’s first brain computer interface with augmented reality wearable speech generating device,

“Cognixion One: The world’s first brain computer interface with augmented reality wearable speech generating device,” Cognixion ONE. [Online]. Available: https://one.cognixion.com/

work page
[7]

Neuralink,

Neuralink, “Neuralink,” neuralink.com, 2024. [Online]. Available: https://neuralink.com/

work page 2024
[8]

Early diagnosis of Alzheimer’s disease based on resting-state brain networks and deep learning,

R. Ju, C. Hu, p. zhou, and Q. Li, “Early diagnosis of Alzheimer’s disease based on resting-state brain networks and deep learning,” IEEE/ACM Trans. Comput. Biol. Bioinf. , vol. 16, no. 1, pp. 244–257, 2019

work page 2019
[9]

Toward on-demand deep brain stimulation using online Parkinson’s disease prediction driven by dynamic detection,

A. Mohammed, M. Zamani, R. Bayford, and A. Demosthenous, “Toward on-demand deep brain stimulation using online Parkinson’s disease prediction driven by dynamic detection,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 25, no. 12, pp. 2441–2452, 2017

work page 2017
[10]

A human-centric Metaverse enabled by Brain-Computer Interface: A survey,

H. Y . Zhu, N. Q. Hieu, D. T. Hoang, D. N. Nguyen, and C.-T. Lin, “A human-centric Metaverse enabled by Brain-Computer Interface: A survey,” IEEE Commun. Surv. Tutor. , vol. 26, no. 3, pp. 2120–2145, 2024. 9

work page 2024
[11]

Progressive tandem learning for pattern recognition with deep spiking neural networks,

J. Wu, C. Xu, X. Han, D. Zhou, M. Zhang, H. Li, and K. C. Tan, “Progressive tandem learning for pattern recognition with deep spiking neural networks,” IEEE Trans. Pattern Anal. Mach. Intell. , vol. 44, no. 11, pp. 7824–7840, 2022

work page 2022
[12]

Energy-efficient decentralized federated learning for UA V swarm with spiking neural networks and leader election mechanism,

C. Shang, D. T. Hoang, M. Hao, D. Niyato, and J. Yu, “Energy-efficient decentralized federated learning for UA V swarm with spiking neural networks and leader election mechanism,”IEEE Wireless Commun. Lett., pp. 1–1, 2024

work page 2024
[13]

Energy-efficient and intelligent ISAC in V2X networks with spiking neural networks-driven DRL,

C. Shang, J. Yu, and D. T. Hoang, “Energy-efficient and intelligent ISAC in V2X networks with spiking neural networks-driven DRL,” arXiv preprint arXiv:2501.01038, 2025

work page arXiv 2025
[14]

Communication-efficient learning of deep networks from decentralized data,

H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Proc. Int. Conf. Artif. Intell. Stat. , 2016. [Online]. Available: https://api.semanticscholar.org/CorpusID:14955348

work page 2016
[15]

Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals,

A. L. Goldberger, L. A. Amaral, L. Glass, J. M. Hausdorff, P. C. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, and H. E. Stanley, “Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals,” Circulation, vol. 101, no. 23, pp. e215–e220, 2000

work page 2000

[1] [1]

Networking architecture and key supporting technologies for human digital twin in personalized healthcare: A comprehensive survey,

J. Chen, C. Yi, S. D. Okegbile, J. Cai, and X. Shen, “Networking architecture and key supporting technologies for human digital twin in personalized healthcare: A comprehensive survey,”IEEE Commun. Surv. Tutor., vol. 26, no. 1, pp. 706–746, 2024

work page 2024

[2] [2]

Digital twin in industry: State-of-the-art,

F. Tao, H. Zhang, A. Liu, and A. Y . C. Nee, “Digital twin in industry: State-of-the-art,” IEEE Trans. Ind. Informat. , vol. 15, no. 4, pp. 2405– 2415, 2019

work page 2019

[3] [3]

Human digital twin for personalized healthcare: Vision, architecture and future directions,

S. D. Okegbile, J. Cai, D. Niyato, and C. Yi, “Human digital twin for personalized healthcare: Vision, architecture and future directions,”IEEE Netw., vol. 37, no. 2, pp. 262–269, 2023

work page 2023

[4] [4]

Human digital twin for fitness management,

B. R. Barricelli, E. Casiraghi, J. Gliozzo, A. Petrini, and S. Valtolina, “Human digital twin for fitness management,” IEEE Access, vol. 8, pp. 26 637–26 664, 2020

work page 2020

[5] [5]

Development and application of Artificial Neural Network,

Y . Wu and J. Feng, “Development and application of Artificial Neural Network,” Wireless Personal Communications, vol. 102, pp. 1645–1656, 2018

work page 2018

[6] [6]

Cognixion One: The world’s first brain computer interface with augmented reality wearable speech generating device,

“Cognixion One: The world’s first brain computer interface with augmented reality wearable speech generating device,” Cognixion ONE. [Online]. Available: https://one.cognixion.com/

work page

[7] [7]

Neuralink,

Neuralink, “Neuralink,” neuralink.com, 2024. [Online]. Available: https://neuralink.com/

work page 2024

[8] [8]

Early diagnosis of Alzheimer’s disease based on resting-state brain networks and deep learning,

R. Ju, C. Hu, p. zhou, and Q. Li, “Early diagnosis of Alzheimer’s disease based on resting-state brain networks and deep learning,” IEEE/ACM Trans. Comput. Biol. Bioinf. , vol. 16, no. 1, pp. 244–257, 2019

work page 2019

[9] [9]

Toward on-demand deep brain stimulation using online Parkinson’s disease prediction driven by dynamic detection,

A. Mohammed, M. Zamani, R. Bayford, and A. Demosthenous, “Toward on-demand deep brain stimulation using online Parkinson’s disease prediction driven by dynamic detection,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 25, no. 12, pp. 2441–2452, 2017

work page 2017

[10] [10]

A human-centric Metaverse enabled by Brain-Computer Interface: A survey,

H. Y . Zhu, N. Q. Hieu, D. T. Hoang, D. N. Nguyen, and C.-T. Lin, “A human-centric Metaverse enabled by Brain-Computer Interface: A survey,” IEEE Commun. Surv. Tutor. , vol. 26, no. 3, pp. 2120–2145, 2024. 9

work page 2024

[11] [11]

Progressive tandem learning for pattern recognition with deep spiking neural networks,

J. Wu, C. Xu, X. Han, D. Zhou, M. Zhang, H. Li, and K. C. Tan, “Progressive tandem learning for pattern recognition with deep spiking neural networks,” IEEE Trans. Pattern Anal. Mach. Intell. , vol. 44, no. 11, pp. 7824–7840, 2022

work page 2022

[12] [12]

Energy-efficient decentralized federated learning for UA V swarm with spiking neural networks and leader election mechanism,

C. Shang, D. T. Hoang, M. Hao, D. Niyato, and J. Yu, “Energy-efficient decentralized federated learning for UA V swarm with spiking neural networks and leader election mechanism,”IEEE Wireless Commun. Lett., pp. 1–1, 2024

work page 2024

[13] [13]

Energy-efficient and intelligent ISAC in V2X networks with spiking neural networks-driven DRL,

C. Shang, J. Yu, and D. T. Hoang, “Energy-efficient and intelligent ISAC in V2X networks with spiking neural networks-driven DRL,” arXiv preprint arXiv:2501.01038, 2025

work page arXiv 2025

[14] [14]

Communication-efficient learning of deep networks from decentralized data,

H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Proc. Int. Conf. Artif. Intell. Stat. , 2016. [Online]. Available: https://api.semanticscholar.org/CorpusID:14955348

work page 2016

[15] [15]

Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals,

A. L. Goldberger, L. A. Amaral, L. Glass, J. M. Hausdorff, P. C. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, and H. E. Stanley, “Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals,” Circulation, vol. 101, no. 23, pp. e215–e220, 2000

work page 2000