27-09-2016, 09:09 AM
1456236558-10.1109icstcc.2013.6689056.pdf (Size: 735.61 KB / Downloads: 9)
Abstract—Nanorobots, a yet theoretical but very promising
future concept, is anticipated to introduce powerful solutions
to human health problems. The way that nanorobot systems
are currently envisaged by researchers is such that they exhibit
behaviour that can be uniquely captured with use of multi-agent
based simulations. In this paper, we investigate this assumption
by describing the process of producing a simple agent-based
simulation of a targeted drug delivery system. The simulation
was designed, implemented, and subsequently evaluated against
the original requirement considerations. Experimental data were
retrieved from various simulation runs, with alternating values
for the system’s parameters. The results strengthened our initial
assumption, demonstrating that nanorobotic drug delivery systems
can be effectively simulated by utilizing intelligent agent
technology.
I. INTRODUCTION
Intelligent Agents have proven to be a popular approach
for modern applications that require numerous heterogeneous
resources, which might have individual skills and partial
information about a problem, to work collaboratively into
achieving the problem’s solution. Despite the limited viewpoint
of each agent, a Multi-Agent System (MAS) potentially
exhibits “emergent intelligence” [1], [2] commonly found in
biologically inspired systems.
In parallel, Nanotechnology, currently one of the most
important and fast growing areas in modern science, focuses
on manipulating matter in structures sized from 1 to 100
nanometers. A plethora of nanotechnological applications are
already commercially available; however, the potential of nanotechnological
constructs targeted for medical use is of special
interest. Amongst the various possible medical endeavours,
targeted drug delivery is considered to be a very promising
strategy for overcoming serious barriers in health impairments,
and nowadays, with the rapidly increasing nanotechnology
advances, is regarded as one of the most promising research
topics worldwide.
Combining the aforementioned research areas, our contribution
focuses on the development of a MAS simulation of
nanorobots that deliver an appropriate pharmaceutical substance
to unhealthy human body tissue. Inspired by the fact
that in silico experimentation has established its importance
to the research community over the past decade, with mathematical models, algorithms and computer simulations having
proven to be valuable allies in studying various and diverse
problems in the fields of Biology and Medical Sciences [3],
[4], we opted to simulate a simple targeted drug delivery
system.
The paper is organized as follows. Section II provides a brief
theoretical background on nanotechnology and the challenges
that are yet to be addressed in medical nanorobotics. A description
of the system and its requirements follows in Section
III. Section IV presents the platform that was selected for
the simulation development, and in section V the simulation
results and subsequent analysis follow. Section VI discusses
the contemporary approaches in manufacturing robots in the
microscale. Finally, in section VII we present some brief
conclusions and a short discussion for further work.
II. BACKGROUND
Nanotechnology has already demonstrated its enormous
prospects through a variety of applications in a wide range
of fields, such as energy and environment, information and
communication, etc. including medical nanoscience. Amongst
the various nanotechnological constructs, nanoparticles are
currently of great scientific interest, due to their unique properties.
These are mostly attributed to the far larger surface area
of nanostructured materials compared to masses built from
larger-scale blocks. From a biomedical viewpoint, their size
is what makes them a very attractive option for a variety of
applications, with drug-delivery and chemical sensing being
the most predominant ones. Nanoparticles can carry a payload
that consists of an appropriate pharmaceutical substance or
proteins, genes etc. Due to their small size, nanoparticles
can evade immune system detection and cross the bloodbrain
barrier gaining access to the brain. Nanoparticles suitable
for drug-delivery, such as dendrimers, nanocrystals, polymeric
micelles, lipid nanoparticles and liposomes [5] are already
being manufactured. In in-vitro experiments [6], [7], target
recognition is performed by administering nano-vehicles into
a cell-culture or some other suitable substrate, whereas in invivo
experiments [8], [9] the nanoparticles are injected into the
area of interest, either the bloodstream or specific body sites.
The nanoparticles can only recognize their target and bind to it when they randomly come in contact with it. Additionally, the process of releasing their payload lies purely on the rate
by which their protective surface biodegrades.
Research in this domain is still in its infants, but scientists
already envision nanodevices that will attain additional
navigational capabilities, computational skills and the ability
to communicate. Such nanodevices can be characterized with
the term “nanorobot”. The term may loosely apply to any
nanodevice that exhibits some kind of individual “intelligence”;
however, in the context of this paper, a nanorobot
is defined as a nanotechnological robot nanomachine which
is a mechanical or electromechanical device with dimensions
measured in nanometers. Its purpose in medicine will be to
maintain and protect the human body against pathogens and it
will also have simple computational capabilities. Nanorobots
are envisaged to be vastly applied to medicine, by being
programmed to perform specific biological tasks. The main
characteristics such nanorobots must have include [10]:
• Biocompatibility: The immune system works against any
foreign body introduced in the organism, therefore a
biologically inert or biocompatible coating is considered
essential for immune system evasion;
• Power to function: The nanorobots would consume energy
when performing their various functions, thus power
supply issues must be addressed;
• Communication: Nanorobots would need to communicate
with each other and with external entities;
• Navigation: To achieve targeted actions the nanorobots
should include some form of navigation mechanism;
• Coordination: The nanorobots must be able to coordinate
their actions in a decentralized manner, to behave cooperatively,
and to be programmable and able to process
information.
III. DRUG DELIVERY REQUIREMENTS
The problem was conceptualized as some cancerous tissue
that is located inside a human body with the abnormal lesion
consisting of both normal and abnormal cells, and being
separated from the blood vessel via natural barriers. It is
assumed that the rough location of the tumour is known
via appropriate imaging techniques, therefore a suitable site
of injection is selected. A number of nanorobots are to be
injected into the bloodstream with the purpose of locating the
problematic area and then destroying the abnormal cells, whilst
leaving the healthy tissue unharmed.
One possible approach would include different types of
nanorobots. Some “tracking” nanorobots would locate the
exact position of the cancerous area by using appropriate
chemotactic sensors that have the ability to measure the
concentration of a specific chemical marker [11], [12]. Then,
“barrier degrading” nanorobots would penetrate the natural
barrier (endothelium and basal membrane) to make way for the
“drug delivery nanorobots”. The latter would in turn identify
the unhealthy cells and deliver their payload to subsequently
destroy them.
System Constraints and Limitations
Due to the inherent complexity of biological systems, it is
unavoidable to apply several layers of abstraction, and focus
on a limited number of parameters and functionalities. Several
constraints are introduced and additional limitations emerge
from the selection of tools and methods that are used to
develop the simulation. Some of the identified constraints are:
• Navigation and accessibility of the nanorobots to the
tumour cells. The nanorobots are assumed to be administered
via injection into the bloodstream, in close
proximity to the tumour detected via existing imaging
techniques. However, blood vessels may not necessarily
be directly accessible via injection from the skin. The
actual path to reach the area of interest is not considered
as part of the problem in this simulation.
• Blood flow and obstacles in the bloodstream. The real
environment of nanorobots would contain red and white
blood-cells, free macromolecules etc. Those elements
would introduce moving obstacles of various sizes and
velocities, that would inevitably alter the nanorobot movement.
Furthermore, the actual blood flow within the vessel
is a major factor affecting the trajectory.
• Boundaries of the software world. The tools that were
used for developing the simulation have inherent limitations.
Real life physical quantities such as time, size,
density, viscosity etc. may be simulated; however, those
quantities and their unit analogies would play a fundamental
role in real-life systems. For instance, where
navigation is concerned, the size of a nanorobot compared
to the size of a body cell would be important. Representing
the system using real-life analogies introduces
an extremely high level of complexity difficult to be be
handled.
• Limited resources. In real-world, the nanorobots would
have limited processing capabilities, energy and communication ranges. Those constraints are to be mirrored in
the simulation. However, additional constraints emerge
from the fact that those limited resources will affect
all aspects of the nanorobotic processes, i.e. navigation,
communication and agent cooperation, in a manner that
cannot be predicted.
• Nanorobots are currently a theoretical construct. Although
in many other MAS applications, simulation can
be verified against some real-life scenarios, this is not currently
the case with the specific model. All results will be
purely speculative, and there is no way of determining the
accuracy of the design or the results of the experiments.