NORA: Neuro-headstage for Online Recording and Actuation in Freely Behaving Animal Models

arXiv:2611.15936 · eess.SP · Submitted 2026-11-11 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Electrical Engineering. The day in electrical engineering: signals, systems and the hardware that carries them.

Rosa: We're looking at the paper: "NORA: Neuro-headstage for Online Recording and Actuation in Freely Behaving Animal Models".

Dev: Here's the gist of the paper. NORA presents a compact wireless neuro-headstage designed for simultaneous neural recording and stimulation in freely behaving animal models, which establishes a practical platform for artifact-reduced neuromodulation research System Architecture The system comprises two main functional blocks: an acquisition block and a stimulation block Each of the two functional blocks is supplied with its own power and ground rails, provided by a single 60 mAh LiPo battery To minimize the overall form factor, the system is packaged in a compact 3D-printed case and a folded rigid-flex PCB composed of two 15 mm x 18 mm boards The folded assembly stacks the neural recorder/stimulator circuitry, MCU, Omnetics connector, and battery within the case, weighing 8.22 g altogether Two nRF5340 MCUs coordinate data acquisition from the RHD2216 and stimulation delivery by the RHS2116 Communication Protocol Design The software framework is designed to support upstream data transfer as well as downstream stimulation and configuration commands for two or more separate BLE connections This is accomplished via a custom data transfer protocol, handled on each side of the transfer by an abstract “manager” layer responsible for coordinating the reception and transmission of data The MCU manager layer sits between the BLE and sensor drivers This manager handles initialization, sampling, and response to received packets Artifact Mitigation Strategy The system employs a power-rail separation strategy to minimize stimulation artifacts The recording and stimulation modules operate from separate power supplies to minimize electrical coupling and stimulation artifacts This separation allows for simultaneous recording and stimulation without sample-dropping or stimulation waveform distortion The difference in average artifact amplitude between configurations is consequential for post-hoc artifact removal In Vivo Validation Results The electrical performance of NORA was first characterized on the bench, showing an averaged input-referred noise across the five channels is 1.92 µV over 300 Hz–7 kHz The second experiment served to demonstrate the acquisition reliability of the NORA system by capturing single-neuron activity at a maximum sampling rate of 20 kHz Recordings from an A53T group showed average beta power elevation consistent with prior findings, confirming that the NORA captures signal features comparable to a commercial headstage Experimental Capabilities The primary contributions of this work include: (1) a compact, fully head-mounted wireless neuro-headstage for simultaneous neural recording and stimulation in freely behaving animal models (2) a duplex wireless architecture with isolated recording and stimulation domains for artifact-reduced bidirectional neural interfacing (3) compatibility with standard Omnetics-based neural probes to support experimental integration (4) in vivo validation demonstrating reliable acquisition of neural features, including single-unit spiking activity and pathological power signatures NORA is a compact, wireless neural interface supporting up to 5 bidirectional channels and an aggregate 20 kHz sampling rate It integrates seamlessly into existing experimental workflows through a standard 36-pin Omnetics connector Future directions include leveraging the platform to develop closed-loop devices with diverse sensing and stimulation modalities ACKNOWLEDGMENT We thank Xiao Wu, Yinfei Li, and Xinyu Song for their technical contributions and assistance in testing REFERENCES

1: Y.

Rosa: Yang et al., “Wireless multilateral devices for optogenetic studies of individual and social behaviors,” Nature Neurosci., vol.

Dev: 24, pp.

Rosa: 1035– 1045, 2021

2: J.

Paper summary: Dev: So we've seen how NORA achieves this simultaneous recording and stimulation using isolated power domains and separate wireless paths. It’s a compact package that manages data transfer through a custom manager layer on each side, coordinating the rhd twenty-two one six and rhs twenty-one one six microcontrollers >

Rosa: The paper by Savanna Blade, Yuhan Hou, He Jia, Laura Kondrataviciute, Runtian Yang, Suneil K. Kalia, and Xilin Liu titled "NORA: Neuro-headstage for Online Recording and Actuation in Freely Behaving Animal Models" presents this device as a way to get bidirectional neural interfacing with minimal artifacts >

Taro: It seems the real contribution here is showing that you can integrate recording and stimulation in a freely behaving model without sample dropping or waveform distortion because of that electrical separation >

Dev: And when we look at the title, "Neuro-headstage for Online Recording and Actuation in Freely Behaving Animal Models," it highlights the dual function and the context of free movement which is key for behavioral studies >

Rosa: The implications are that this provides a platform that supports up to five bidirectional channels with an aggregate two zero kilohertz sampling rate, integrating easily with standard omnetics based neural probes >

Taro: Future work suggested by the authors involves leveraging this platform to develop closed loop devices with diverse sensing and stimulation modalities, which opens up a lot of possibilities for real time feedback systems >

Dev: So NORA is essentially a compact wireless interface that bridges the gap between recording and stimulation in complex animal behavior experiments, relying on that power rail separation for its artifact reduction strategy >

Conclusion: Rosa: So we're wrapping up on NORA, this compact wireless neuro-headstage for recording and stimulating in freely moving animals that we just talked about >

Dev: Yeah, looking at the title again, "NORA: Neuro-headstage for Online Recording and Actuation in Freely Behaving Animal Models," it just sounds like a really practical tool for getting real data when the animal isn't sitting still >

Taro: It's about putting recording and stimulation into one small thing that moves with the animal, which is a big step beyond just having two separate pieces of equipment >

Rosa: Exactly, and the fact that it works wirelessly means you can actually do this during naturalistic behavior studies, not just in a lab setting where everything has to be wired up >

Dev: The authors are showing that they managed to get both recording and stimulation happening on one platform without messing up the signal or causing noise from the stimulation itself >

Taro: And they did that by separating the power rails, which is smart because electrical coupling between recording and stimulation modules can totally mess up your results >

Rosa: So what does this mean for us in terms of using this thing? It seems like it fits right into existing experimental workflows with standard stuff like omnetics connectors >

Dev: It really simplifies things by making the system smaller and more integrated, which usually means less hassle when you're trying to run long experiments >

Taro: But we have to remember that they validated it in vivo, so the real test is how it handles the messy, unpredictable world of a living animal >

Rosa: And that validation showed they captured things like single-unit spiking and even those pathological power signatures we talked about before >

Dev: So while the engineering looks solid with its battery and chip coordination, the real win here seems to be in making this complex bidirectional interface actually work reliably outside a controlled bench setting >

Taro: It's a compact package that supports up to five bidirectional channels at twenty kilohertz sampling rate, which gives you a lot of data for behavioral analysis >

Rosa: Right, so the big picture here is moving from bulky setups to something you can actually wear on an animal and get reliable feedback from >

Dev: And it opens the door for more complex things down the road, like developing closed loop devices that use this recording and stimulation to give real-time control >

Savanna Blade, Yuhan Hou, He Jia, Laura Kondrataviciute, Runtian Yang, Suneil K. Kalia, Xilin Liu

Department of Electrical and Computer Engineering, University of Toronto Department of Surgery, Krembil Research Institute, Toronto Western Hospital

eess.SP

Submitted: 2026-11-11

Updated: 2026-11-11

Comments: 5 pages, 7 figures. Accepted for publication in IEEE BioCAS 2026. Author accepted manuscript

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 87/100

The gist: NORA presents a compact wireless neuro-headstage designed for simultaneous neural recording and stimulation in freely behaving animal models…

Key concepts

Acquisition Block
This block is responsible for capturing neural signals from the animal's brain. It works in tandem with the stimulation block, coordinating data collection and managing the flow of information to ensure reliable signal capture during experiments.
Stimulation Block
This component delivers electrical stimulation to the brain. Crucially, it operates on its own power supply rails, completely separate from the recording circuitry. This isolation minimizes electrical noise and prevents stimulation artifacts from corrupting the neural recordings.
Power-Rail Separation Strategy
NORA uses distinct power supplies for recording and stimulation modules. This physical separation is a key design feature intended to reduce electrical coupling between the two functions, thereby minimizing artifacts that would otherwise interfere with accurate neural measurements.
Custom Data Transfer Protocol
The system uses a unique protocol managed by an abstract 'manager' layer to handle communication. This layer coordinates upstream data transfer and downstream commands for multiple BLE connections, ensuring efficient and organized data exchange between the animal setup and the external computer.

Terminology

Summary

NORA presents a compact wireless neuro-headstage designed for simultaneous neural recording and stimulation in freely behaving animal models, which establishes a practical platform for artifact-reduced neuromodulation research

System Architecture

The system comprises two main functional blocks: an acquisition block and a stimulation block Each of the two functional blocks is supplied with its own power and ground rails, provided by a single 60 mAh LiPo battery To minimize the overall form factor, the system is packaged in a compact 3D-printed case and a folded rigid-flex PCB composed of two 15 mm x 18 mm boards The folded assembly stacks the neural recorder/stimulator circuitry, MCU, Omnetics connector, and battery within the case, weighing 8.22 g altogether Two nRF5340 MCUs coordinate data acquisition from the RHD2216 and stimulation delivery by the RHS2116

Communication Protocol Design

The software framework is designed to support upstream data transfer as well as downstream stimulation and configuration commands for two or more separate BLE connections This is accomplished via a custom data transfer protocol, handled on each side of the transfer by an abstract “manager” layer responsible for coordinating the reception and transmission of data The MCU manager layer sits between the BLE and sensor drivers This manager handles initialization, sampling, and response to received packets

Artifact Mitigation Strategy

The system employs a power-rail separation strategy to minimize stimulation artifacts The recording and stimulation modules operate from separate power supplies to minimize electrical coupling and stimulation artifacts This separation allows for simultaneous recording and stimulation without sample-dropping or stimulation waveform distortion The difference in average artifact amplitude between configurations is consequential for post-hoc artifact removal

In Vivo Validation Results

The electrical performance of NORA was first characterized on the bench, showing an averaged input-referred noise across the five channels is 1.92 µV over 300 Hz–7 kHz The second experiment served to demonstrate the acquisition reliability of the NORA system by capturing single-neuron activity at a maximum sampling rate of 20 kHz Recordings from an A53T group showed average beta power elevation consistent with prior findings, confirming that the NORA captures signal features comparable to a commercial headstage

Experimental Capabilities

The primary contributions of this work include:

(1) a compact, fully head-mounted wireless neuro-headstage for simultaneous neural recording and stimulation in freely behaving animal models

(2) a duplex wireless architecture with isolated recording and stimulation domains for artifact-reduced bidirectional neural interfacing

(3) compatibility with standard Omnetics-based neural probes to support experimental integration

(4) in vivo validation demonstrating reliable acquisition of neural features, including single-unit spiking activity and pathological power signatures

Conclusion

NORA is a compact, wireless neural interface supporting up to 5 bidirectional channels and an aggregate 20 kHz sampling rate It integrates seamlessly into existing experimental workflows through a standard 36-pin Omnetics connector Future directions include leveraging the platform to develop closed-loop devices with diverse sensing and stimulation modalities

ACKNOWLEDGMENT

We thank Xiao Wu, Yinfei Li, and Xinyu Song for their technical contributions and assistance in testing

REFERENCES

[1] Y. Yang et al. p. 0, “Wireless multilateral devices for optogenetic studies of individual and social behaviors,” Nature Neurosci. p. 4, vol. 24, pp. 1035– 1045, 2021

[2] J. P. Newman et al. p. 4, “ONIX: a unified open-source platform for multimodal neural recording and perturbation during naturalistic behavior,” Nature Methods, vol. p. 4 22, pp. 187–192, 2025

[3] D. J. Pederson, C. p. 4 J. Quinkert, M. p. 4 A. Arafat, J. P. Somann, J. D. Williams, R. p. 4 A Bercich, Z Wang, G O Albors, J G R Jefferys and P P Irazoqui, “The bionode: A closed-loop neuromodulation implant,” ACM Trans Embed Comput Syst. p. 4, vol. 18, no 1 Feb 2019

[4] W. Ouyang et al. p. 4, “A wireless and battery-less implant for multimodal closed-loop neuromodulation in small animals,” Nature Biomed Eng. p. 4, vol. 7, pp. 1252–1269, 2023

[5] M. D Hoang, W Kang, M Koh, and S-M Park, “Fully wireless implantable device capable of multichannel neural spike recording and stimulation for long-term freely moving rodent study,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. p. 4 33, p 1621–1632, 2025

[6] W Lee, S Blade, Y Hou, A McIver, R Yang, X Liu and A G Richardson, “A miniature wireless device for closed-loop auditory sleep modulation,” in IEEE EMBS Conference on Neural Engineering (NER), Nov 2025

[7] R Yang, Y Hou, S Blade, Y Li, V Tyagi, G-E Boudreault-Morales, J Zariffa and X Liu, “A multi-modal system featuring wireless flexible sensor patches and a depth-sensing imager for home-based monitoring of rehabilitation exercises,” Integrated Circuits and Systems, vol. 1, no 4 pp 227–238 Dec 2024

[8] A Zhou, S R Santacruz, B C Johnson, G Alexandrov, A Moin, F L Burghardt, J M Rabaey, J M Carmena and R Muller, “A wireless and artefact-free 128-channel neuromodulation device for closed-loop stimulation and recording in non-human primates,” Nature Biomedical Engineering, vol. 3, no 1 p 15–26 Jan 2019

[9] D R Merrill, M Bikson, and J G R Jefferys, “Electrical stimulation of excitable tissue: design of efficacious and safe protocols,” Journal of Neuroscience Methods, vol. 141, no 2 p 171–198 Feb 2005

[10] B Lee, M K Koripalli, Y Jia, J Acosta, M S E Sendi, Y Choi and M Ghovanloo, “An implantable peripheral nerve recording and stimulation system for experiments on freely moving animal subjects,” Scientific Reports, vol. 8, no 1 p 6115 Apr 2018

[11] L Kondrataviciute et al., “Cortical and basal ganglia beta oscillations and frequency-dependent DBS effects in the A53T Parkinson’s disease rat model,” npj Parkinsons Dis. p. 4, vol.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI systems can perform simultaneous neural recording and stimulation in freely behaving animal models because NORA features Separate recording and stimulation interfaces, each managed with a dedicated MCU and wireless module. This separation allows for simultaneous recording and stimulation without sample-dropping or stimulation waveform distortion.

  2. Improved AI systems benefit from reduced artifacts through electrical isolation, as the system minimizes interference by employing independent power domains and dedicated wireless communication paths for the recording and stimulation subsystems. This design helps in achieving artifact-reduced bidirectional neural interfacing.

  3. Improved AI systems can integrate into existing experimental workflows because NORA features a 36-pin Omnetics interface compatible with a wide variety of commercial neural probes. This compatibility supports experimental integration by allowing researchers to use established infrastructure without needing custom probe interfaces.

  4. Improved AI systems gain reliable signal acquisition capabilities through the system's ability to capture single-unit spiking activity and pathological power signatures. Validation demonstrated that NORA captures features with results comparable to a commercial wireless headstage, which is crucial for studying complex phenomena like Parkinson’s disease signatures.

  5. Improved AI systems can manage complex data streams efficiently via a modular firmware framework where the MCU Manager handles data transfer, and the GUI Application Manager manages board instances. This allows for straightforward modification of common operations and control over multiple devices from a single interface using an intuitive and repeatable workflow.

Related papers