Journal of

Functional
Biomaterials

Review
Recent Progress in Artificial Neurons for Neuromodulation
Qinkai Jiang 1 and Mengwei Liu 2, *

1 College of Materials Science and Engineering, Sichuan University, Chengdu 610065, China;
jiangqinkai@[Link]
2 School of Communication and Electronic Engineering, East China Normal University, Shanghai 200241, China
* Correspondence: mwliu@[Link]

Abstract: Driven by the rapid advancement and practical implementation of biomaterials, fabrica-
tion technologies, and artificial intelligence, artificial neuron devices and systems have emerged
as a promising technology for interpreting and transmitting neurological signals. These systems
are equipped with multi-modal bio-integrable sensing capabilities, and can facilitate the benefits
of neurological monitoring and modulation through accurate physiological recognition. In this
article, we provide an overview of recent progress in artificial neuron technology, with a particular
focus on the high-tech applications made possible by innovations in material engineering, new
designs and technologies, and potential application areas. As a rapidly expanding field, these
advancements have a promising potential to revolutionize personalized healthcare, human enhance-
ment, and a wide range of other applications, making artificial neuron devices the future of brain-
machine interfaces.

Keywords: artificial neurons; bio-integrated system; multi-modal sensing; implantable devices;

neuromodulation

1. Introduction
As one of the main organs in the human body, the brain plays a significant role in
people’s daily activities. It consists of billions of neurons interconnected in a sophisticated
Citation: Jiang, Q.; Liu, M. Recent
neural network. Brain–machine interfaces (BMIs) have the ability to bridge the gap between
Progress in Artificial Neurons for
humans and machines by interpreting and transmitting neurological information [1]. This is
Neuromodulation. J. Funct. Biomater. crucial in areas such as neuron rehabilitation, brain signal decoding, and body control [2–4].
2024, 15, 214. [Link] Over the past few decades, people have been trying to understand the complex operation
10.3390/jfb15080214 of the brain, which involves hundreds of millions of neurons working closely. These
efforts have led to a number of exciting applications, yet not enough to acquire a full
Academic Editor: Emad
understanding of the brain’s functional wiring diagram, which hinders the development of
Moeendarbary
all related fields.
Received: 6 June 2024 Currently, the artificial neuron has recently gained significant attention in the field
Revised: 12 July 2024 of BMIs, as it allows for the sensing and interpretation of neural activities, mimicking the
Accepted: 22 July 2024 working functions of the brain and biological nervous systems [5]. Designing interactive
Published: 30 July 2024 artificial neuron systems that integrate sensing, storage, and processing capabilities has
become a key focus in this research area. These interactive artificial neuron devices and
systems are considered highly important in endowing BMIs with neuromorphic sensing
and interactive characteristics, enabling them to solve more complex problems. Current
Copyright: © 2024 by the authors.
advancements in artificial neurons primarily rely on electrophysiological signals, such
Licensee MDPI, Basel, Switzerland.
as microelectrode arrays (MEA), to interpret and transmit neurological information by
This article is an open access article
recording action potentials of individual cells and local field potentials [6]. For example,
distributed under the terms and
Professor Michael Brecht and his research team reported a single neuron stimulation
conditions of the Creative Commons
Attribution (CC BY) license (https://
method to precisely achieve evoked spike pattern control and measurement [7]. Such
[Link]/licenses/by/
methods could make contributions to insights into the neuronal working principles and
4.0/). “decoding” of cortical spikes [8].

J. Funct. Biomater. 2024, 15, 214. [Link] [Link]

J. Funct. Biomater. 2024, 15, 214 2 of 14

However, performing communication with neurons using the natural language—neuro

transmitters—and developing chemically based devices are still major challenges ahead [9].
Neurotransmitters are endogenous chemicals that mediate chemical synaptic transmission,
enabling brain function and shaping our everyday life and activity. Imbalance of neuro-
transmitters would cause multiple neurological disorders, such as Parkinson’s disease, major
depressive disorder, and epilepsy [10]. Thus, a closed-loop chemical-to-electrical-to-chemical
signal transduction system will provide us an alternative method for better communication
with biological neural networks and modulation of their activities.
In contrast to conventional computer processors, artificial neurons have the ability to
emulate the functions and operational patterns of neuronal activity in the human brain,
enabling them to tackle more complex tasks. This field of research is interdisciplinary,
involving material science, neuromorphic engineering, electronics/iontronics, electrochem-
istry, and neuroscience [5,8,11]. However, despite significant progress, these neuromorphic
devices are still relatively new and face numerous challenges when it comes to establishing
direct communication with biological neural networks. One of the key challenges in the
development of artificial neurons is finding a way to achieve both bio-integrity and the
ability to replicate living neural functions simultaneously. Addressing the requirements for
materials, fabrication methods, and influential applications is of utmost importance. In this
paper, we will outline these requirements for artificial neurons and provide an overview of
the recent state-of-the-art advancements, along with their various applications.
We have organized this article into several sections, each addressing different aspects
and requirements of artificial neurons. We start by discussing the necessary prerequi-
sites and materials for these devices, followed by an exploration of their novel designs
and potential uses. We then present a comprehensive conclusion summarizing the find-
ings and advancements in the field. Additionally, we provide an outlook on future pos-
sibilities and the potential impact of these developments. Throughout the review, we
emphasize the importance of supporting materials and unique fabrication methods that
contribute to the success of these devices. The progress made in the development of such
devices and systems has paved the way for the emerging flexible bioelectronic technology,
which has the potential to expand the scope of current technologies in human–machine
interfaces (HMI).

2. Materials and Fabrication Technology

As an implantable bio-integrated devices, the material selection and fabrication process
of artificial neurons need to be chosen carefully in order to reach the high standard for
long-term working inside human body. In this Section, we are going to discuss the material
requirements and suitable fabrication technologies for artificial neuron devices and systems.

2.1. Material Requirements for Artificial Neurons

The design and fabrication of artificial neurons draw inspiration from biological
synapses and neural networks. To create a functional artificial neuron device, various
components are required, including a data acquisition system that utilizes multi-modal
bionic receptors for sensing, an artificial synapse for data processing and analysis, and an
output unit for controlling biological neural networks. For illustration purposes, Figure 1
provides a visual representation of a basic artificial neuron structure.
J. Funct. Biomater. 2024, 15, 214 3 of 14

Figure 1. Schematic illustration of the artificial neuron key components. (a) Schematic of structure
and working mechanism of biohybrid neurons, showing the interface between biological and artificial
neurons. Such design could achieve neuromorphic communication and form a closed loop system.
(b) Illustration demonstrating the detailed design of the artificial neuron and showing different
components, including an electrochemical sensor, a signal processing unit, and a hydrogel-based
molecular releasing part. Such design enables the device to detect and release neurotransmitters.
Reprinted with permission from Ref. [12] Copyright 2022 Springer Nature. (c) Conformal brain
electrodes. (i) Image of the brain electrode fabricated on silk film and the Au electrical probe array.
(ii) Schematic of device configuration, showing the detailed structure of each layer in the device.
(iii) Image of the device during the animal experiment when the brain electrode is implanted in the
rat brain. The insect image shows the electrocorticography (ECoG) signals of different statuses, such
as resting, epileptic seizure onset, as well as epilepsy inhibition statuses. The device also includes
a silk-based diffractive optical elements (DOEs), as shown in the green inset picture. This optical
component aims to achieve real-time monitor-like drug release processes and conformal mounting
status. Reprinted with permission from Ref. [13] Copyright 2019 John Wiley and Sons.
J. Funct. Biomater. 2024, 15, 214 4 of 14

To ensure the safe and effective long-term interaction with the human body, particu-
larly with brain tissues, the materials used in artificial neurons need to have low mechanical
mismatch and robust biocompatibility to minimize potential damage such as inflamma-
tion, infection, tissue intrusion, and allergic reactions [14]. Additionally, the multi-modal
sensing capabilities of artificial neuron devices require materials that can be modified in
various ways to capture multiple functions and achieve high sensitivity for accurate phys-
iological monitoring [15,16]. Currently, there are several intrinsically biocompatible soft
materials available that can serve as stretchable sensing components to record or respond
to external stimuli. These materials include natural materials [17–20], polymers [21–23],
hydrogels [24–27], and nanocomposite-based elastomers [28–32]. These materials can act as
the fundamental building blocks of bionic receptors. In the following sections, we discuss
several examples of these materials and highlight their key features that make them suitable
for artificial neuron functions and applications.
In recent decades, there has been a growing need to develop materials that possess
both robust bio-compatibility and flexibility for use in implantable bio-integrated artificial
neurons, driven by advancements in technology and the desire to interface with diverse
biological environments within the brain. As an example, silk materials have emerged as a
versatile option for fabricating bio-integrated electronics, with numerous applications in
tissue engineering and implantable medical devices [33–36]. Recent research has shown
that silk-based materials exhibit strong adhesion to brain tissues, thanks to their unique
mechanical properties and the ability to undergo chemical modifications [37,38]. Brain
electrodes and devices incorporating silk-based materials have demonstrated excellent
bio-compatibility and precise sensing capabilities.
Despite the promising qualities of silk protein and other natural materials, their usage
is still in the early stages and presents several technical and commercial challenges that
need to be addressed. These challenges include difficulties in reliable raw material pro-
cessing and storage at a large scale, resulting in limited applicability [39,40]. Consequently,
researchers frequently turn to synthetic materials that have already achieved large-scale pro-
duction. One such example is the modified poly(3,4-ethylenedioxythiophene): polystyrene
sulfonate (PEDOT:PSS) hydrogel, which has found extensive use in the development of
bionic receptors and artificial synapses [41,42]. Organic neuromorphic devices that utilize
PEDOT:PSS as the active material have successfully demonstrated synaptic functions such
as short-term potentiation (STP) and global regulation of synaptic behavior. Furthermore,
PEDOT:PSS can serve as a suitable surface for in vitro monitoring, and can also be utilized
for the electrochemical detection of neurotransmitters through redox reactions. For instance,
when dopamine is introduced into the electrolyte solution during the detection process,
the conductance change is enhanced due to the oxidation of dopamine at the postsynaptic
gate electrode.

2.2. Fabrication Methods for Artificial Neurons

In addition to the choice of high-quality materials, the selection of fabrication methods
is a crucial step in the preparation of artificial neuron devices. Recent advancements in var-
ious fabrication techniques have greatly enhanced the capabilities of bio-integrated sensors.
To effectively summarize the fabrication requirements for artificial neurons, it is important
to differentiate the various components based on their specific functions. As illustrated
in Figure 1, the structural components typically include the bionic receptors, the artificial
synapse, and the output unit. Currently, in the case of implantable devices, soft materials
still dominate as a major proportion of the fabrication process, as they provide enhanced
flexibility and expanded functionalities compared to conventional sensing technologies.
Furthermore, during the device fabrication process, careful consideration must be given
to designing sensors that exhibit high sensitivity, fast response times, good stability, and
tunable synaptic properties. Addressing these factors and incorporating them into the
fabrication process will contribute to the overall performance and functionality of artificial
neuron devices.
J. Funct. Biomater. 2024, 15, 214 5 of 14

Bionic receptors, which are responsible for sensing chemical molecules in the sur-
rounding environment, serve as a critical component of artificial neurons. Therefore, the
development of new nanofabrication methods plays a vital role in enhancing the overall
performance of these devices [39,43]. Extensive research efforts have been devoted to
advancements in nanomaterial synthesis and device fabrication procedures, encompassing
techniques such as inkjet printing, laser printing, optical microlithography, and hot-pressing
modeling. For instance, a recent device fabrication method involving a chemically me-
diated artificial neuron consisting of three essential building blocks: a dopamine (DA)
electrochemical sensor, a resistive switching memristor, and a heat-induced DA-releasing
hydrogel [12]. A complete process flow was established, including the fabrication of a
flexible substrate using a polyethylene terephthalate (PET) film, surface modification, and
the incorporation of a polyvinyl alcohol (PVA)/silicon dioxide (SiO2)/DA hydrogel.
By employing these advanced nanofabrication techniques, researchers are able to
enhance the performance and functionality of artificial neuron devices, ultimately leading
to more precise and efficient chemical sensing capabilities.
Besides the bionic receptor, the memristor is also a crucial component in the construc-
tion of artificial neurons, responsible for information processing. Recent advancements in
nanoscale material synthesis and corresponding nanofabrication techniques have enabled
the development of more precise and advanced memristors. Typically, memristors utilize a
metal–insulator–metal (MIM) structure, where the insulator layer is sandwiched between
two metal electrodes.
To achieve multimodal sensing and computing capabilities, further improvements and
developments in various aspects are necessary. These include techniques for decoupling
multiple signals, designs for innovative device structures, and explanations for new sensing
mechanisms. By continuously refining and enhancing these aspects, the performance and
functionality of memristors can be improved, leading to more efficient and versatile artificial
neuron systems.

3. Neuron Stimulation Technology

Apart from neuronal activity recording, neural stimulation is another fundamental
component to construct an integral closed-loop feedback system for biological neural sys-
tem regulation. Currently, the vast majority of treatments for neurological disorders are
mainly based on stimulation technology, which involve oral administration of pharmaceu-
ticals, electrical stimulation, and chemical stimulation.
Among them, the first two are widely used in the clinical treatment of neurological
diseases, including epilepsy, Parkinson’s disease, depression, etc. However, the important
factor is that they both can lead to various side effects, which can then affect physiological
functions such as memory, learning, etc. In addition, drug resistance and dependence
are also prevalent consequences that occur among neurological disease patients under
certain prescription drug treatment. In comparison, local chemical stimulation, especially
endogenous molecules such as neurotransmitters, is an ideal solution for these above-
mentioned problems. Multiple devices have been successfully designed to achieve local
chemical stimulation for neural network regulation. In this Section, we will demonstrate
several methods that have been reported in recent research, including brain electrodes,
microfluidic systems, a hydrogel-based releasing unit, and an organic electronic ion pump.

3.1. Brain Electrodes

Implanted brain electrodes are efficient methods to achieve electrical stimulation for
neuronal activity modulation. Many groups have reported different brain electrode designs,
which can stimulate the inactive biological neuronal network and modulate the brain
functions. Some of them are now used in the clinical treatment for neurological disorder
diseases like major depressive disorder, epilepsy, and Parkinson’s disease. However, such
treatment sometimes can cause side effects, sometimes severely influencing patients’ daily
activities. For example, deep brain stimulation and electroconvulsive therapy, which have
 J. Funct. Biomater. 2024, 15, 214 6 of 14

been used as treatment methods for major depression, were associated with a risk of
cognitive side effects, especially memory impairment. Although such side effects have
been lowered due to technique improvement, they have not been successfully eliminated
so far [44,45].
Furthermore, since the biological neurons’ activity is always based on chemical com-
munications, electrical signals will cause a mismatch between living neurons and the
artificial device and system, and find it hard to form an integral closed-loop feedback
system. Thus, recent studies are more likely to focus on the chemical communication,
especially for the endogenous molecules’ release methods for the artificial neurons, as
shown below in the following sections.

3.2. Hydrogel-Based Releasing Unit

Recently, several research groups focus on the study of chemically mediated hydrogel
to achieve various drug molecules released for the therapeutic treatment of multiple
diseases. This technique can also be used in artificial neural networks for biological neuron
stimulation. For example, the research group of Xiaodong Chen has fabricated a heat-
responsive hydrogel for the stimulation of dopamine (DA) release, forming a chemical
communication loop between artificial neurons and biological neural networks, as shown
in Figure 2a. In the experiment results part, they reported robust heat-responsive properties
of this selected hydrogel, which can emulate controllable DA-releasing behavior in the
axon vesicles [12]. Connected in series with a memristor, the deformation of the DA-
encapsulated hydrogel can be triggered using a small potential (~0.5 V), allowing for the
neurotransmitters to diffuse out of the PVA hydrogel and then fire the afferent nerve.
Therefore, the system can achieve mediated DA release by adjusting the resistance value of
the memristor.
Although this chemical-mediated hydrogel-based artificial neuron could enable bidi-
rectional communication between artificial and biological neurons, there are still some
features that need to be further improved, such as response time, power consumption,
etc. In addition, this hydrogel-releasing design still has several unresolved problems that
need to be fixed for practical applications. For instance, the heating process might cause
damage to living cells and neurons when the temperature exceeds a certain value. Also,
downscaling of the system, especially the hydrogel-releasing unit, and reducing the energy
consumption need to be considered for long-term use. Lastly, as an implantable device
. Biomater. 2024, 15, x FOR PEER REVIEW system, the material’s and device’s stability need to be continually improved
7 of 14 for integration
with a complex neuron network.

Figure 2. Cont.
J. Funct. Biomater. 2024, 15, 214 7 of 14

Figure 2. ExamplesFigureof 2.
neuron stimulation
Examples technologies.
of neuron stimulation(a)technologies.
Hydrogel-releasing unit. (i) Schematic
(a) Hydrogel-releasing unit. (i) Schematic
illustration of a temperature-sensitive dopamine release component, fabricated using
illustration of a temperature-sensitive dopamine release component, fabricated PVA/SiO2/DA
using PVA/SiO2/DA
hydrogel. (ii) Results of dopamine release by the hydrogel-based unit. Reprinted with permission
hydrogel. (ii) Results of dopamine release by the hydrogel-based unit. Reprinted with permis-
from Ref. [12] Copyright 2022 Springer Nature. (b) Optical image of an implantable drug delivery
sion from Ref. [12] Copyright 2022 Springer Nature. (b) Optical image of an implantable drug
system fabricated on an elastomeric substrate and scanning electron micrographs of the platinum–
delivery
silicone composite. system
Reprinted fabricated
with on an
permission elastomeric
from substrate 2015
Ref. [46] Copyright and scanning electron
The American micrographs of the
Associ-
ation for the Advancement of Science. (c) Illustration of an organic electronic ion pump based on2015 The Ameri-
platinum–silicone composite. Reprinted with permission from Ref. [46] Copyright
can Association
PEDOT:PSS recording electrodefor the
and itsAdvancement of Science.
working principle. (c) Illustration
The device is able toof an organic
record electronic ion pump
the neuro-
based on PEDOT:PSS recording electrode and its working principle.
transmitters and release molecules from the reservoir (left). Reprinted from Ref. [47]. The device is able to record the
neurotransmitters and release molecules from the reservoir (left). Reprinted from Ref. [47].
Although this chemical-mediated hydrogel-based artificial neuron could enable bi-
3.3. Microfluidic
directional communication Systemsartificial
between Releasing Unit
and biological neurons, there are still some
features that need toAnother way
be further to enablesuch
improved, electrical-to-chemical signal consumption,
as response time, power transduction between
etc. artificial
and
In addition, this biological neuronsdesign
hydrogel-releasing is microfluidic systems.
still has several Many studies
unresolved havethat
problems demonstrated
need the local
release of endogenous molecules using microfluidics. Their basic design
to be fixed for practical applications. For instance, the heating process might cause dam- and structures are
as shown in Figure 2b.
age to living cells and neurons when the temperature exceeds a certain value. Also,
downscaling of the It is true that
system, microfluidics
especially is capable of delivering
the hydrogel-releasing unit,any
andsoluble compound,
reducing the providing
large possibility for different molecules, drugs, and endogenous chemicals. Unfortunately,
this system always requires the delivery of the target compound in a carrier fluid and will
likely to induce convection, and might risk disrupting the fragile biochemical macrocondi-
tion. Although some groups have demonstrated a convection-free delivery system based
on redox switching of the conducting polymer, the number of applicable molecules and
compounds are still limited. In addition, complex and bulky set ups, such as pumps and
valves, are usually inevitable, making the practical application restricted and inconvenient.
Finally, the poor on–off rate ratio, release rate, and reaction speed are also disadvantages of
this system which need improvements.
J. Funct. Biomater. 2024, 15, 214 8 of 14

3.4. Organic Electronic Ion Pump

So far, a new technology, organic electronic ion pumps (OEIP), has recently started
to be used in the design of bioelectronic neuron for the delivery of different endogenous
neurotransmitters. Professor Daniel T. Simon and his research team reported a bioelectronic
neuron pixel, which could achieve an inhibitory neurotransmitter, γ-aminobutyric acid
(GABA) delivery, and stop epileptiform activity [42,47]. The OEIP can mimic the working
principle of exocytotic release of neurotransmitters in biological neurons. In the design
of OEIP, organic electrodes, normally made of conducting polymers, and permselective
membrane are two key components, enabling precise neurotransmitters release without
involving liquid flow, which is always necessary in microfluidic systems. In detail, as shown
in Figure 2c, when a potential is applied between the two electrodes, one in reservoir and
one in medium, an electric field will be established and the migration of neurotransmitters
in the cation-conduction channel will be aroused. In the end, the process of record and
release of GABA at the same site was demonstrated, and epileptiform activity of the
downstream neurons can be suppressed successfully.
It is worth noting that, compared to the former mentioned types of neurotransmitter
release methods, the structure of OEIP is quite compact and simple, avoiding any cumber-
some and complex experimental set up, and can achieve recording and stimulating at the
same site with high resolution. However, these systems are always limited in turn-on speed,
since device dynamics are largely governed by the travel distance of delivered molecules.
Thus, reducing the effective channel length and developing device with significantly faster
turn-on are the areas that need more attention in the near future in order to improve the
performance of OEIP for a better control of the release of neurotransmitters and neuronal
activity modulation.

4. Synaptic Devices for Neural Signal Analysis

Apart from neural stimulation, analysis of signals after collection is also a critical part
in order to form a closed-loop system used for human–machine interface applications.
By mimicking the functions of biological synapses, properties like computing efficiency
and speed can be greatly increased for a better application in bio-integrated intelligent
systems. As shown in Figure 3, we provide an overview of synaptic devices used for neural
signal analysis.
For example, Liu et al. has demonstrated successful seizure detection by using
memristor-based brain-machine interfaces. The illustration in Figure 3a shows the con-
cept of their system design and its detecting flow. In their report, the power efficiency
of this synaptic-like memristor-based sensing system is 400 times than traditional CMOS
circuits. It proves that in-memory computations on the memristor crossbar can help with
the reduction in energy consumption [48]. Additionally, the research group of Prof. George
G. Malliaras reported a neuromorphic device, which can behave like the homeoplasticity
phenomena in the neural environment (Figure 3b). By using electrolyte gating to build
intricate connections between devices, they also reveal the advantages of networks of
neuromorphic devices with less hardwired connectivity [49]. In addition, Boyuan Mu et al.
have established near-infrared artificial synapses to achieve biological synaptic functions,
such as short-term and long-term plasticity and spike rate dependence based on electrical
and optical properties’ modulation. The results showed in their work indicate that the
development of artificial synapses could contribute to the areas of neural robots and neural
computing [50].
J. Funct.
J. Funct. Biomater.
Biomater. 2024,
2024, 15,
15, 214
x FOR PEER REVIEW 9 9of
of 14

Figure
Figure 3.3. Synaptic
Synaptic devices
devices forfor neural signal analysis.
neural signal analysis. (a)
(a) Memristor-based
Memristor-based neural
neural signal analysis
signal analysis
system for brain–machine interfaces (BMIs). Conceptual diagram shows the processes of recording
system for brain–machine interfaces (BMIs). Conceptual diagram shows the processes of record-
neural activities and signal analyses based on a memristor array. Reprinted with permission from
ing neural activities and signal analyses based on a memristor array. Reprinted with permission
Ref. [48] Copyright 2020 Springer Nature. (b) Schematic of the configuration of the array of neuro-
from [Link].
morphic [48] Copyright
Channel2020 Springer
of each deviceNature.
is made(b) Schematic
of PEDOT: of the
PSS. configuration
Reprinted of the array
with permission of
from
neuromorphic devices. Channel of each device is made of PEDOT: PSS. Reprinted with
Ref. [49] Copyright 2017 Springer Nature. (c) Schematic illustration and SEM image of synaptic tran- permission
from Ref.
sistor. [49] Copyright
Application 2017 sensory
of artificial Springerneuron
[Link]
(c) Schematic
based onillustration
an array ofand SEM image
artificial of synaptic
synaptic devices.
Reprinted
transistor. with permission
Application from Ref.
of artificial [50] Copyright
sensory 2021 John
neuron system Wiley
based on and [Link] artificial synaptic
an array
devices. Reprinted with permission from Ref. [50] Copyright 2021 John Wiley and Sons.
For example, Liu et al. has demonstrated successful seizure detection by using
5. Potential Applications
memristor-based brain-machine interfaces. The illustration in Figure 3a shows the concept
Artificial
of their systemneurons
design have
and itsthedetecting
ability toflow.
mimicInthe fundamental
their report, the capabilities of biological
power efficiency of this
neurons, such as sensing, processing, and regulating neuronal activities.
synaptic-like memristor-based sensing system is 400 times than traditional CMOS [Link],
because
It provesof their
that ability forcomputations
in-memory neuromodulation,
on the they hold immense
memristor potential
crossbar can for various
help with the re-
applications, including neuron rehabilitation, brain signal decoding,
duction in energy consumption [48]. Additionally, the research group of Prof. George and human body
G.
control. Specifically,
Malliaras reported a artificial neurondevice,
neuromorphic devices can be
which canintegrated withthe
behave like biological efferent
homeoplasticity
nerves, or their
phenomena engineered
in the counterparts,(Figure
neural environment to establish
3b). complete
By using artificial or hybrid
electrolyte gating synaptic
to build
reflex arcs, as depicted in Figure 4.
intricate connections between devices, they also reveal the advantages of networks of
 neurons, such as sensing, processing, and regulating neuronal activities. Consequently,
because of their ability for neuromodulation, they hold immense potential for various ap-
plications, including neuron rehabilitation, brain signal decoding, and human body con-
trol. Specifically, artificial neuron devices can be integrated with biological efferent nerves,
J. Funct. Biomater. 2024, 15, 214
or their engineered counterparts, to establish complete artificial or hybrid synaptic10reflex
of 14

arcs, as depicted in Figure 4.

Figure 4. Different applications of artificial neuron. (a) (i) Schematic illustration of the potential
interface between artificial neuron and biological neural network systems. (ii) Long-term potentiation
of postsynaptic current of artificial neurons. Reprinted with permission from Ref. [51] Copyright
2020 Springer Nature. (b) (i) Application of artificial neurons for re-establishing the connection of a
mouse sciatic nerve. (ii) When exposed to a dopamine stimulus, the artificial neuron can trigger the
movement of the mouse leg and control the degree of such movement with different molecular con-
centrations. (iii) The response angle of the mouse leg under DA stimulus with different concentrations.
Reprinted with permission from Ref. [12] Copyright 2022 Springer Nature. (c) (i) Device structures
and working principles: positive GABA ions being delivered. (ii) Suppression of epileptiform activity
in the hippocampus with OEIPs. Reprinted with permission from Ref. [52] Copyright 2015 John
Wiley and Sons.

As an illustration of the applications, the NTS (neurotransmitter-triggered synapse)

can be seamlessly integrated with a motor/control unit, such as an insect leg, actuator,
or pneumatic robot, to create a hybrid bioelectronic reflex arc that emulates the muscle
J. Funct. Biomater. 2024, 15, 214 11 of 14

activation process [53–55]. These advancements present significant potential in the fields
of neurorobotics and neural prosthetics. In addition to interneurons, this system can also
replicate the function of neurotransmitter-triggered motor neurons, enabling control over
muscle contraction feedback. To exemplify this functionality, a dopamine (DA) stimulus is
employed to initiate the movement of either a robotic hand or a mouse leg. Notably, the
concentration of DA directly impacts the potential drop across the resistance (connected in
series with the memristor), which in turn regulates the motion of both the robotic hand and
the mouse leg. The experimental results demonstrate that in the absence of DA stimulus,
neither the robotic hand nor the mouse leg can provide any feedback due to low input
signals [12].
Furthermore, the team led by Christophe Bernard has made significant progress in
utilizing three distinct models to induce epileptiform activity. They have successfully
demonstrated that the administration of gamma-aminobutyric acid (GABA) leads to rapid
and targeted suppression of this activity. This groundbreaking research suggests that these
devices hold immense potential in terms of drug delivery within the brain, particularly for
the administration of antiepileptic agents [52].

6. Conclusions and Outlook

To summarize, this review provides an introduction to the concept and fundamental
structure of artificial neurons, offering an overview of recent advancements and emphasiz-
ing the key requirements and applications. We made a thorough summary of the recent
development of artificial neurons, including the new materials, device architectures, fab-
rication techniques, and potential applications. To meet the criteria of artificial neurons,
two crucial elements must be considered: the careful selection of appropriate materials,
and the implementation of fabrication methods ensuring bio-integrity. Furthermore, the
optimization of the paired materials’ preparation and fabrication processes is essential to
seamlessly integrate diverse sensors and synaptic devices.
However, despite significant progress, there are still several challenges that need to
be addressed in the near future. Firstly, the current detection limits are insufficient for
accurately detecting sub-micromolar concentration changes in different neurotransmitters.
Secondly, the bulky size of the sensing component hampers achieving precise spatial
accuracy, thereby limiting the ability to achieve high-resolution molecule sensing. Thirdly,
the level of device integration has not yet reached the stage of enabling simultaneous
multichannel/multimodal sensing and processing. Fourthly, there is a lack of exploration
in the areas of neuron adhesion, high flexibility, tissue compatibility, and their potential
applications in understanding and treating depression mechanisms. These are all crucial
aspects that require further research and improvement.

Author Contributions: Conceptualization, M.L. and Q.J.; methodology, M.L.; software, M.L. and
Q.J.; validation, M.L. and Q.J.; formal analysis, M.L.; investigation, M.L. and Q.J.; resources, M.L.;
data curation, M.L. and Q.J.; writing—original draft preparation, M.L. and Q.J.; writing—review and
editing, M.L.; visualization, M.L. and Q.J.; supervision, M.L.; project administration, M.L.; funding
acquisition, M.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The original contributions presented in the study are included in the
article, further inquiries can be directed to the corresponding authors.
Conflicts of Interest: The authors declare no conflict of interest.
J. Funct. Biomater. 2024, 15, 214 12 of 14

References
1. Fu, T.-M.; Hong, G.; Zhou, T.; Schuhmann, T.-G.; Viveros, R.-D.; Lieber, C.-M. Stable long-term chronic brain mapping at the
single-neuron level. Nat. Methods 2016, 13, 875–882. [CrossRef]
2. Park, S.W.; Kim, J.; Kang, M.; Lee, W.; Park, B.-S.; Kim, H.; Choi, S.-Y.; Yang, S.; Ahn, J.-H.; Yang, S. Epidural electrotherapy for
epilepsy. Small 2018, 14, 1801732. [CrossRef]
3. Gutruf, P.; Krishnamurthi, V.; Vázquez-Guardado, A.; Xie, Z.; Banks, A.; Su, C.-J.; Xu, Y.; Haney, C.R.; Waters, E.A.; Kandela, I.;
et al. Fully implantable optoelectronic systems for battery-free, multimodal operation in neuroscience research. Nat. Electron.
2018, 1, 652–660. [CrossRef]
4. Kozielski, K.L.; Jahanshahi, A.; Gilbert, H.B.; Yu, Y.; Erin, Ö.; Francisco, D.; Alosaimi, F.; Temel, Y.; Sitti, M.l. Nonresonant
powering of injectable nanoelectrodes enables wireless deep brain stimulation in freely moving mice. Sci. Adv. 2021, 7, eabc4189.
[CrossRef]
5. Kim, Y.; Chortos, A.; Xu, W.; Liu, Y.; Oh, J.Y.; Son, D.; Kang, J.; Foudeh, A.M.; Zhu, C.; Lee, Y.; et al. A bioinspired flexible organic
artificial afferent nerve. Science 2018, 360, 998–1003. [CrossRef]
6. Huang, Q.; Tang, B.; Romero, J.C.; Yang, Y.; Elsayed, S.K.; Pahapale, G.; Lee, T.-J.; Morales Pantoja, I.E.; Han, F.; Berlinicke, C.;
et al. Shell microelectrode arrays (MEAs) for brain organoids. Sci. Adv. 2022, 8, eabq5031. [CrossRef]
7. Doron, G.; Brecht, M. What single-cell stimulation has told us about neural coding. Philos. Trans. R. Soc. B Biol. Sci. 2015, 370,
20140204. [CrossRef]
8. Loynachan, C.N.; Soleimany, A.P.; Dudani, J.S.; Lin, Y.; Najer, A.; Bekdemir, A.; Chen, Q.; Bhatia, S.N.; Stevens, M.M. Renal
clearable catalytic gold nanoclusters for in vivo disease monitoring. Nat. Nanotechnol. 2019, 14, 883–890. [CrossRef]
9. Zhang, Y.; Riexinger, J.; Yang, X.; Mikhailova, E.; Jin, Y.; Zhou, L.; Bayley, H. A microscale soft ionic power source modulates
neuronal network activity. Nature 2023, 620, 1001–1006. [CrossRef]
10. Zhang, Y.; Zhou, Z.; Fan, Z.; Zhang, S.; Zheng, F.; Liu, K.; Zhang, Y.; Shi, Z.; Chen, L.; Li, X.; et al. Self-powered multifunctional
transient bioelectronics. Small 2018, 14, 1802050. [CrossRef]
11. von Erlach, T.C.; Bertazzo, S.; Wozniak, M.A.; Horejs, C.-M.; Maynard, S.A.; Attwood, S.; Robinson, B.K.; Autefage, H.; Kallepitis,
C.; del Río Hernández, A.; et al. Cell-geometry-dependent changes in plasma membrane order direct stem cell signalling and fate.
Nat. Mater. 2018, 17, 237–242. [CrossRef]
12. Wang, T.; Wang, M.; Wang, J.; Yang, L.; Ren, X.; Song, G.; Chen, S.; Yuan, Y.; Liu, R.; Pan, L.; et al. A chemically mediated artificial
neuron. Nat. Electron. 2022, 5, 586–595. [CrossRef]
13. Shi, Z.; Zheng, F.; Zhou, Z.; Li, M.; Fan, Z.; Ye, H.; Zhang, S.; Xiao, T.; Chen, L.; Tao, T.H.; et al. Silk-Enabled Conformal Multifunc-
tional Bioelectronics for Investigation of Spatiotemporal Epileptiform Activities and Multimodal Neural Encoding/Decoding.
Adv. Sci. 2019, 6, 1801617. [CrossRef]
14. Liu, M.; Zhang, Y.; Tao, T.H. Recent progress in bio-integrated intelligent sensing system. Adv. Intell. Syst. 2022, 4, 2100280.
[CrossRef]
15. Liu, M.; Zhang, Y.; Liu, K.; Zhang, G.; Mao, Y.; Chen, L.; Peng, Y.; Tao, T.H. Biomimicking antibacterial opto-electro sensing
sutures made of regenerated silk proteins. Adv. Mater. 2021, 33, 2004733. [CrossRef]
16. Liu, M.; Zhang, Y.; Wang, J.; Qin, N.; Yang, H.; Sun, K.; Hao, J.; Shu, L.; Liu, J.; Chen, Q.; et al. A star-nose-like tactile-olfactory
bionic sensing array for robust object recognition in non-visual environments. Nat. Commun. 2022, 13, 79. [CrossRef]
17. Ma, X.; Sun, X.; Chen, J.; Lei, Y. Natural or natural-synthetic hybrid polymer-based fluorescent polymeric materials for bio-
imaging-related applications. Appl. Biochem. Biotechnol. 2017, 183, 461–487. [CrossRef]
18. Gyarmati, B.; Pukánszky, B. Natural polymers and bio-inspired macromolecular materials. Eur. Polym. J. 2017, 93, 612–617.
[CrossRef]
19. Yang, H.; Fan, F.R.; Xi, Y.; Wu, W. Bio-derived natural materials based Triboelectric devices for self-powered ubiquitous wearable
and implantable intelligent devices. Adv. Sustain. Syst. 2020, 4, 2000108. [CrossRef]
20. Wani, S.; Gautam, S.; Qadrie, Z.; Gangadharappa, H. Silk fibroin as a natural polymeric based bio-material for tissue engineering
and drug delivery systems—A review. Int. J. Biol. Macromol. 2020, 163, 2145–2161. [CrossRef]
21. Hsiao, Y.-S.; Ho, B.-C.; Yan, H.-X.; Kuo, C.; Chueh, D.-Y.; Yu, H.-h.; Chen, P. Integrated 3D conducting polymer-based bioelectronics
for capture and release of circulating tumor cells. J. Mater. Chem. B 2015, 3, 5103–5110. [CrossRef]
22. Xu, W.; Zhong, L.; Xu, F.; Song, W.; Wang, J.; Zhu, J.; Chou, S. Ultraflexible transparent bio-based polymer conductive films based
on Ag nanowires. Small 2019, 15, 1805094. [CrossRef] [PubMed]
23. Lin, C.-H.; Luo, S.-C. Zwitterionic conducting polymers: From molecular design, surface modification, and interfacial phe-
nomenon to biomedical applications. Langmuir 2022, 38, 7383–7399. [CrossRef] [PubMed]
24. Han, L.; Lu, X.; Wang, M.; Gan, D.; Deng, W.; Wang, K.; Fang, L.; Liu, K.; Chan, C.W.; Tang, Y.; et al. A mussel-inspired conductive,
self-adhesive, and self-healable tough hydrogel as cell stimulators and implantable bioelectronics. Small 2017, 13, 1601916.
[CrossRef]
25. Noshadi, I.; Walker, B.W.; Portillo-Lara, R.; Shirzaei Sani, E.; Gomes, N.; Aziziyan, M.R.; Annabi, N. Engineering biodegradable
and biocompatible bio-ionic liquid conjugated hydrogels with tunable conductivity and mechanical properties. Sci. Rep. 2017, 7,
4345. [CrossRef]
26. Lan, S.; Gao, Z.; Barbosa, M.; Santato, C. Flexible ion-gated transistors making use of poly-3-hexylthiophene (P3HT): Effect of the
molecular weight on the effectiveness of gating and device performance. J. Electron. Mater. 2020, 49, 5302–5307. [CrossRef]
J. Funct. Biomater. 2024, 15, 214 13 of 14

27. Zhang, P.; Guo, W.; Guo, Z.H.; Ma, Y.; Gao, L.; Cong, Z.; Zhao, X.J.; Qiao, L.; Pu, X.; Wang, Z.L. Dynamically crosslinked dry
ion-conducting elastomers for soft iontronics. Adv. Mater. 2021, 33, 2101396. [CrossRef] [PubMed]
28. Choi, S.; Han, S.I.; Jung, D.; Hwang, H.J.; Lim, C.; Bae, S.; Park, O.K.; Tschabrunn, C.M.; Lee, M.; Bae, S.Y.; et al. Highly
conductive, stretchable and biocompatible Ag–Au core–sheath nanowire composite for wearable and implantable bioelectronics.
Nat. Nanotechnol. 2018, 13, 1048–1056. [CrossRef]
29. Sun, H.; Zhou, K.; Yu, Y.; Yue, X.; Dai, K.; Zheng, G.; Liu, C.; Shen, C. Highly stretchable, transparent, and bio-friendly strain
sensor based on self-recovery ionic-covalent hydrogels for human motion monitoring. Macromol. Mater. Eng. 2019, 304, 1900227.
[CrossRef]
30. Xu, Y.; Sun, B.; Ling, Y.; Fei, Q.; Chen, Z.; Li, X.; Guo, P.; Jeon, N.; Goswami, S.; Liao, Y.; et al. Multiscale porous elastomer
substrates for multifunctional on-skin electronics with passive-cooling capabilities. Proc. Natl. Acad. Sci. USA 2020, 117, 205–213.
[CrossRef]
31. Bahena, A.; Magana Rivera, I.; Lopez Gonzalez, H.R.; Handa, R.; Enríquez-Medrano, F.; Kumar, S.; Carrizales, R.; Fernández, S.;
Valencia, L.; Díaz de León, R. Bio-elastomer nanocomposites reinforced with surface-modified graphene oxide prepared via in
situ coordination polymerization. RSC Adv. 2020, 10, 36531–36538. [CrossRef]
32. Magana Rivera, I.; Georgouvelas, D.; Handa, R.; Neira-Velazquez, M.; Héctor, R.; Lopez Gonzalez, H.R.; Javier, F.; Medrano,
E.; Díaz de León, R.; Valencia, L. Fully bio-based elastomer nanocomposites comprising polyfarnesene reinforced with plasma-
modified cellulose nanocrystals. Polymers 2021, 13, 2810. [CrossRef]
33. Zhang, Y.; Zhou, Z.; Sun, L.; Liu, Z.; Xia, X.; Tao, T.H. “Genetically engineered” biofunctional triboelectric nanogenerators using
recombinant spider silk. Adv. Mater. 2018, 30, 1805722. [CrossRef]
34. Huang, T.; Zhang, Y.; He, P.; Wang, G.; Xia, X.; Ding, G.; Tao, T.H. “Self-matched” tribo/piezoelectric nanogenerators using
vapor-induced phase-separated poly (vinylidene fluoride) and recombinant spider silk. Adv. Mater. 2020, 32, 1907336. [CrossRef]
35. Liu, M.; Tao, T.H.; Zhang, Y. Silk Materials Light Up the Green Society. Adv. Energy Sustain. Res. 2021, 2, 2100035. [CrossRef]
36. Liu, M.; Zhang, Y.; Zhang, Y.; Zhou, Z.; Qin, N.; Tao, T.H. Robotic Manipulation under Harsh Conditions Using Self-Healing
Silk-Based Iontronics. Adv. Sci. 2022, 9, 2102596. [CrossRef]
37. Zhang, Y.; Tao, T.H. Skin-friendly electronics for acquiring human physiological signatures. Adv. Mater. 2019, 31, 1905767.
[CrossRef] [PubMed]
38. Zhang, Y.; Tao, T.H. A bioinspired wireless epidermal photoreceptor for artificial skin vision. Adv. Funct. Mater. 2020, 30, 2000381.
[CrossRef]
39. Zhou, Z.; Zhang, S.; Cao, Y.; Marelli, B.; Xia, X.; Tao, T.H. Engineering the Future of Silk Materials through Advanced Manufactur-
ing. Adv. Mater. 2018, 30, 1706983. [CrossRef]
40. Sahoo, J.K.; Hasturk, O.; Falcucci, T.; Kaplan, D.L. Silk chemistry and biomedical material designs. Nat. Rev. Chem. 2023, 7,
302–318. [CrossRef]
41. Isaksson, J.; Kjäll, P.; Nilsson, D.; Robinson, N.; Berggren, M.; Richter-Dahlfors, A. Electronic control of Ca2+ signalling in
neuronal cells using an organic electronic ion pump. Nat. Mater. 2007, 6, 673–679. [CrossRef]
42. Simon, D.T.; Kurup, S.; Larsson, K.C.; Hori, R.; Tybrandt, K.; Goiny, M.; Jager, E.W.H.; Berggren, M.; Canlon, B.; Richter-Dahlfors,
A. Organic electronics for precise delivery of neurotransmitters to modulate mammalian sensory function. Nat. Mater. 2009, 8,
742–746. [CrossRef] [PubMed]
43. Zhang, Y.; Chen, X.; Chen, D.; Yao, Z.; Xu, S.; McArdle, P.; Qazilbash, M.; Liu, M. Partially metal-coated tips for near-field
nanospectroscopy. Phys. Rev. Appl. 2021, 15, 014048. [CrossRef]
44. Lisanby, S.H. The effects of electroconvulsive therapy on memory of autobiographical and public events. Arch. Gen. Psychiatry
2000, 57, 581–590. [CrossRef]
45. Hu, Y.P.; Rakesh, G.; Lisanby, S.H. Recent developments in noninvasive neuromodulation for mood and anxiety disorders. Curr.
Behav. Neurosci. Rep. 2015, 2, 173–185. [CrossRef]
46. Minev, I.R.; Musienko, P.; Hirsch, A.; Barraud, Q.; Wenger, N.; Moraud, E.M.; Gandar, J.; Capogrosso, M.; Milekovic, T.; Asboth,
L.; et al. Electronic dura mater for long-term multimodal neural interfaces. Science 2015, 347, 159–163. [CrossRef]
47. Jonsson, A.; Inal, S.; Uguz, I.; Williamson, A.; Kergoat, L.; Rivnay, J.; Khodagholy, D.; Berggren, M.; Bernard, C.; Malliaras, G.;
et al. Bioelectronic neural pixel: Chemical stimulation and electrical sensing at the same site. Proc. Natl. Acad. Sci. USA 2016, 113,
9440–9445. [CrossRef]
48. Liu, Z.; Tang, J.; Gao, B.; Yao, P.; Li, X.; Liu, D.; Zhou, Y.; Qian, H.; Hong, B.; Wu, H. Neural signal analysis with memristor arrays
towards high-efficiency brain–machine interfaces. Nat. Commun. 2020, 11, 4234. [CrossRef]
49. Gkoupidenis, P.; Koutsouras, D.A.; Malliaras, G.G. Neuromorphic device architectures with global connectivity through electrolyte
gating. Nat. Commun. 2017, 8, 15448. [CrossRef]
50. Mu, B.; Guo, L.; Liao, J.; Xie, P.; Ding, G.; Lv, Z.; Zhou, Y.; Han, S.-T.; Yan, Y. Near-Infrared Artificial Synapses for Artificial Sensory
Neuron System. Small 2021, 17, 2103837. [CrossRef]
51. Keene, S.T.; Lubrano, C.; Kazemzadeh, S.; Melianas, A.; Tuchman, Y.; Polino, G.; Scognamiglio, P.; Cinà, L.; Salleo, A.; Van der
Burg, Y.; et al. A biohybrid synapse with neurotransmitter-mediated plasticity. Nat. Mater. 2020, 19, 969–973. [CrossRef] [PubMed]
52. Williamson, A.; Rivnay, J.; Kergoat, L.; Jonsson, A.; Inal, S.; Uguz, I.; Ferro, M.; Ivanov, A.; Sjöström, T.A.; Simon, D.T.; et al.
Controlling epileptiform activity with organic electronic ion pumps. Adv. Mater. 2015, 27, 3138–3144. [CrossRef] [PubMed]
J. Funct. Biomater. 2024, 15, 214 14 of 14

53. Lee, Y.; Oh, J.Y.; Xu, W.; Kim, O.; Kim, T.R.; Kang, J.; Kim, Y.; Son, D.; Tok, J.B.-H.; Park, M.J.; et al. Stretchable organic
optoelectronic sensorimotor synapse. Sci. Adv. 2018, 4, eaat7387. [CrossRef] [PubMed]
54. Shim, H.; Sim, K.; Ershad, F.; Yang, P.; Thukral, A.; Rao, Z.; Kim, H.-J.; Liu, Y.; Wang, X.; Gu, G.; et al. Stretchable elastic synaptic
transistors for neurologically integrated soft engineering systems. Sci. Adv. 2019, 5, eaax4961. [CrossRef]
55. He, K.; Liu, Y.; Wang, M.; Chen, G.; Jiang, Y.; Yu, J.; Wan, C.; Qi, D.; Xiao, M.; Leow, W.R.; et al. An artificial somatic reflex arc. Adv.
Mater. 2020, 32, 1905399. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.