Digital Twins for Drug Design: Importance, Applications and Future Perspectives

Introduction

The pharmaceutical field is one of the rapidly evolving industries that continually searches for new drugs to counteract newly emergent diseases as well as strengthen the fight against the old ones. This desire for new drugs calls for a costly, complex, and time-consuming drug design process. The traditional evolutionary pattern of drugs involves a series of steps, including target identification, discovery, and investigational new drug application (IND), which dictate multiple planning, experimentation, and analysis. However, throughout the complex process, new emerging technology emerges to hasten and smoothen the drug discovery and design process. One of the most burgeoning technological applications in drug design is the use of Digital Twins. Drug design is a multifacial scientific field, involving target identification, small molecules, and drug candidates’ design. This method often starts with target identification, where specific biological molecules, like proteins or enzymes, lead to drug intervention. After identifying the targets, the researchers utilize computer-aided drug design to design and create specific small molecule capable of regulating the target. Molecular modeling and virtual screening and medicinal chemistry optimization of the lead compound occurs to enhance developing various drug candidates designed to possess the desired pharmacological properties. Lastly, selected candidates pass through rigorous test before getting approval for market release. Is the use of computer models of a real-life process to optimize how it is maintained and updated, ideally on the network but at a rate limited by the requirement for human involvement.^{1, 2}

The Digital Twins in drug design are novel models to induce behaviour with computational precision and predict high-level models’ behaviour explicitly. It creates high-net possible models using computational molecular dynamics and quantum mechanics models integrating the multiple complex models using data from a range to sources and levels of structural biology, genomics, and pharmacology. In addition, digital twins enable research to access complex relationships between tens or hundreds target of a million lead, and feedback to initiate tests allows patient feedback, such being able to test patient specific. It holds paramount beneficial application in personalized medicine. Implication on the resultant drug is that it considerably improves the first cycle time, save time, cost by predicting candidate drugs and removing failure drug trials. It also advances first cycle drug development, forecasting patient trials, and overall drug trials. Ultimately the drug has an improved efficacy and safety profile. In conclusion, digital utilization in drug discovery and design is a perfect ideal in drug design. Digital twins make virtual clinical trials in patients a possible alternative. It reduces multiple target lead, it removes certain drugs from animal trials, and tests patient specificity. In addition, it eliminates clinical drug failures from approved drugs. There is massively mitigated risk sponsored trials. It highly expedites the drug design time process.^3-5

What distinguishes this review from current literature is its wide-angle view of how digital twins merge with molecular simulations—including molecular dynamics and quantum-mechanical methods—and artificial intelligence within drug discovery. Earlier papers have examined digital twins, but mostly in separate silos; in contrast, this article knits together their uses in molecular modelling, pharmacokinetics and pharmacodynamics, synthetic chemistry, and even the evolving regulatory landscape. It also highlights how digital twins could drive personalized medicine and virtual clinical trials, moving patient-specific simulations from theory to practice. By mapping present uses alongside emerging trends, pairing them with diagrams, and benchmarking them against classical workflows, the review sketches a clear road ahead for adopting these technologies throughout pharmaceutical research and development. 

Understanding Digital Twins

Digital twins have recently appeared as a novel approach reinventing traditional practices of design, monitoring, and optimization across multiple domains. At its core, a digital twin is a virtual digital representation of a live physical actor or process that is synchronously fed with real-time information to keep the digital model in sync with its real-world analogue. The living digital counterpart enables the in-depth analysis, forecasting, and optimization of performance in quantifiable terms, contributes to increased efficiencies and systems reliability, as well as potential innovation. In the context of drug design, they provide a completely new approach to modelling of biologic systems, which enables the investigation of molecular interactions, forecasting of the reaction of drugs, and the accelerated production of innovative therapeutics. A digital representation refers to a live physical actor’s digital counterpart in the form of a virtual model. It includes the physical characteristics forming the geometry and appearance of the physical counterpart and its characteristics such as dynamism, and how it interacts with its environment and other physical actors. Digital replicas of physical entities and processes in chemical system design contain the information obtained from sensors, simulations, data feed, or any other data sources concerning the monitored physical systems.^6-8

Biological digital twins are virtual representations of living organisms, cells, tissues, or entire ecosystems. These highly detailed computational models are designed to mimic the behavior, characteristics, and responses of their physical counterparts with remarkable accuracy. By leveraging vast amounts of biological data, advanced computational algorithms, and powerful simulation capabilities, researchers can create digital replicas that capture the intricate complexities of biological systems.^{9, 10}

Digital twins are developed across several steps, starting with the acquisition of data from physical sensors, measurements, or simulations. Subsequently, the obtained data is processed, integrated, and transformed into a digital form, based on sophisticated modeling and simulation methods. Machine learning strategies and artificial intelligence are commonly used for analysis and interpretation, helping personalize the digital twin model and increase its efficiency in the long term. Digital twins have been already applied to numerous fields, including manufacturing and aerospace, as well as healthcare and urban planning. For example, in manufacturing, they are used for predictive maintenance studies, process optimization, and quality control, as the system can analyze the expected equipment behaviour and predict malfunctions. In aerospace, the digital twins are helpful for design validation, performance monitoring, and fleet management, improving the level of safety and sustainability. The digital twins related specifically to healthcare are used for personalized medicine purposes, disease modeling, and drug development. So, how do the digital twins work in drug designing? Three important components are essential for such a system to simulate, analyze, and optimize the biological systems and drug candidates. ^{11, 12}

Digital twins utilize predictive analytics algorithms for predicting drug response, pharmacokinetics and pharmacodynamics and optimizing drug treatment regimens . Especially when researchers input the patient data and design scenarios of drug functioning within a virtual twin, it is possible to tailor treatment to an individual patient profile’s needs and maximize drug efficacy while minimizing its side effects. Optimizing using feedback loops. Digital twin is an excellent tool for the iterative optimization of drug candidates using feedback loops for obtaining experimental results and simulating computational predictions. By simultaneously refining the models based on the new knowledge obtained from parallel systems and laboratory experiments, researchers can more quickly identify promising drug candidates. Thus, digital twin is a powerful tool for modeling, analyzing, and optimizing biological systems and drug effects on these systems. These systems offer a possibility to make a virtual replica, predict interactions, and reduce the time of the creation of new therapeutics. Consequently, digital twin as a tool and a concept is promising for introducing a new level of drug research and development. Thus, it is quite promising to expect the breakthrough in the development of new drugs, and a universal use of digital twins by the pharmaceutical companies, mainly if they digitize their processes.^{5, 13, 14} 

The Role of Digital Twins in Drug Discovery

Drug discovery occupies a central place in the pursuit of novel therapeutics for the treatment of diseases. Over the years, the research and development of new therapeutic interventions have been characterized by long timelines, overuse of resources, and high attrition rates. The traditional drug discovery paradigm has, however, undergone a paradigm shift with the emergence of the digital twin concept, with significant transformational effects on the drug development landscape. When traditional drug discovery methods are described in more detail, the drug outweighs the activity phase. Target identification of the original target involves targeting chemical-biological molecules, including, but not limited to, proteins and enzymes, participating in the development and manifestation of the disease. Once the target is validated, the phenotypic model or chemical library is screened to select potential modulators – the lead. These leads have optimal efficacy and pharmacokinetic characteristics. The lead is then multi-optimized by syntheses, structure-activity studies, and pharmacological profiling. The observed body is transformed into the clinic resulting in slow cycle times and unnecessary increases in body utilization. Before they enter the clinic, these selected leads go through various body tests.  The traditional approach to drug discovery, however, has several limitations. First, it relies on a trial-and-error approach, which does not result in a significant trial and error approach and attrition rates. The biological system is too complex and the structure strains too much to predict it very absolutely for constitutional and temporal reasons. Additionally, traditional drug discovery is geographically and financially expensive. Syndrome development cycles and six to seven weeks between prostate cancer studies can be as expensive as hundreds of billions of dollars.^{1, 15, 16}

To address the challenges of traditional drug discovery methods, digital twins pose a viable solution with immense potential. A digital twin provides a virtual replica of biological systems and offers the ability to formulate digital twins of drug candidates and molecules, allowing researchers to simulate and predict the behavior of molecules with a high degree of accuracy. The computational-based method uses molecular dynamics simulation, quantum mechanics calculations, and artificial intelligence algorithms and model techniques to simulate molecular interactions, drug response prediction, and optimization of clinical drug interventions. An advantage of digital twins during drug discovery is the ability to offer insights through drug-target engagement. Unlike empirical methods that this latter achieves results from the overall molecular testing environments, the digital twin helps to explore how the drug factor and the target come into effect to achieve binding. They explain the structural explanation of binding affinity, the determinant, selectivity, and efficacy. By integrating experimental data from structural biology X-ray and Cryo-EM, and pharmacology digital twin facilitates the generation of rational design stagers leading to the selection of optimized compound candidates. Furthermore, digital twins offer the opportunity to cut the duration of the rigorous drug discovery timeline by making lead optimization happen faster and reducing the heavy clinical trial process duration. The iterative model of the twin enabling the computation to be refined over time while basing the refined averages on the current and new experimental data availing from the basic research making clinical changeover become easy and faster. The digital also offers the possibility of creating a class of A digital Twin 2.0 which would allow for the prediction and explorations of drug combinations to reach a definable efficacy and personalized medicine establishment. Digital twins also empower researchers to tackle the conceptual paradigm of research completely. In that, when the twin avails data, analysis of the testing strong outs, and developed confirmation through simulation. Digital simulated models, then the simulated models become a powerful platform to test hypothesis plans.^17-19

Figure 1: Digital Twin for Drug Discovery – This illustration represents the integration of digital twin technology in drug discovery, showcasing virtual replicas of biological systems, real-time data feedback, and AI-driven molecular simulations. Click here to View Figure

Digital Twins for Drug Molecule Synthesis

The application of digital twin technology in drug molecule synthesis is transforming the chemical manufacturing industry by permitting virtual tests and model-based predictive synthesis. In terms of chemical developments, a digital twin represents a sophisticated simulation copy of all components and stages of a given reaction for an entire molecular process, including selection of precursors, up to the isolation of products. A digital twin is capable of sophisticated simulations at the intersection of quantum chemistry, reaction kinetics, thermodynamics, machine learning which enables accurate prediction of complex chemical reactions and optimal pathways to novel drug chemistries important for modern medicine. Such systems can be used to virtually try out different synthetic routes alongside reagents, solvents and catalysts in addition to other conditions greatly streamlining the resources and time traditionally required in labs focusing on physical experimentation. For example, computers aided retrosynthetic analysis has empirically been proven smartly integrated through AI’s powerful intelligence embedded within contemporaneous cheminformatics enabling connections to physicochemical databases employing AI algorithms for predictive intuition during data-driven design across broad horizons using tightly woven rule boxes packed with extremely smart search-and-consider mechanisms weighing not only reaction going forward but retroactively assessing yield vs stereoselectivity along metrication towards greenness envy whilst evaluating emittance straining.The Synthia/Chematica/IBMrsnXfC emerging consortia digital encyclopedias dished up computational alchemy primed alchemic parsing system when sponteously allied intertwined via nodes ors paving paths earning bond cross linked computations coupled precomposing minds!^{20, 21}

Furthermore, the utilization of digital twins allows for the continuous monitoring and optimizing drug synthesis in batch or continuous flow – on both types of chemical systems. Digital twins have feedback control systems that incorporate sensor data, reaction kinetics, as well as thermodynamic parameters which enable real-time forecasting and optimization of reactions. This ability profoundly impacts enhancing the efficiency and yield while reducing waste and energy consumption. To illustrate, during large-scale pharmaceutical synthesis conducted in continuous-flow reactors, digital twins could manage the temperature, pressure, flow rate, and mixing for streamlined control which ensures product uniformity and compliance to regulations instantaneously. In addition to these real-time adjustments, other advanced computational models within the twin can simulate actionable lifecycles of catalysts as well as depict patterns concerning their degradation along with byproduct formation; thus helping chemists adjust parameters or switch routes if needed.The modeling extends to performing risk analysis assessing possible side reactions, thermal runaways or incompatibilities vital for safeguarding safety during industrial-grade synthesis. Moreover integrating high-performance computing with quantum chemical approaches like DFT or Hartree-Fock enables.^{22, 23}

Figure 2: AI-generated molecular structures demonstrate potential pathways for retrosynthetic analysis in drug design. Click here to View Figure

Building Digital Twins for Molecular Simulations

Digital twins represent a unique technology for replicating and simulating real-world processes, making inroads in drug design for the simulation and investigation of molecular interactions. Central to the creation and use of digital twins is molecular simulation, which mimics the behaviour of molecules and predicts interaction with biological targets. Advanced computational tools such as molecular dynamics simulations and quantum mechanics calculations allow for the production of high-fidelity digital twins that help rationalize drug design and optimization. Molecular dynamics is the principal source of information in digital twin production due to its dynamic nature and detailed representation of molecular movement. Molecular dynamics is a simulation method that uses Newton’s equations of motion to chart the progress of atoms and molecules over a period. The positions and speeds of atoms are updated dynamically based on of interatomic forces derived from potential energy functions. MD simulations enable investigation of conformational dynamics and optima, flexibility, and binding modes of biomolecule interactions due to their high spatial and temporal resolution.^{24, 25}

MD simulations are incredibly useful in drug discovery since they help to describe the structural dynamics of proteins, nucleic acids, and other biological macromolecules involved in disease-related biochemical processes. These simulations can be used to predict the binding of small molecules to their target proteins, predict the stability of the corresponding drug-receptor complexes, estimate the rate of repressor-activator interactions, and find vital components of those interactions. Due to MD simulations, researchers can also analyse ligand-protein fields, study allosteric effects and other ways of fitting the ligands into the proteins, and compare different types of enzymes, protein distributions, and other features that can be used during the optimization of a drug candidate. MD also allows people to study the behaviour of lipid bilayers, experts, and other parts of the biological systems, which can be used to understand how a drug candidate may expand membrane or dissolve in water. MD also makes it possible to calculate the way drug candidate interacts with the cell lines membrane, thus giving the necessary data to predict absorption, biodistribution, and excretion rates, facilitate drug administration, or provide a difference between the same drugs. Finally, QM calculations serve as another type of complimentary to MD analysis. While the former depicts the behaviour of chromosomes as a collection of particles, the latter does so as a collection of wave-particles. Hence, QM provides people with exact geometrical and electronic data.^{26, 27}

Regarding drug design, QM simulations are specifically useful for the investigation of chemical reactions, ligand binding, and electronic properties of drug candidates. Utilizing electronic energies, molecular orbitals, and charge distributions, QM simulations help researchers access the stability and reactivity of drug molecules, predict reaction pathways, and optimize synthesis routes by accurate calculation. Another main application of QM simulations is the calculation of binding free energies, electrostatic interactions, and dispersion forces, which provide quantitative analysis of ligand-protein interactions and binding affinity. Thus, the accuracy and reliability of the digital twin model are crucial to drug discovery and design. If the molecular representation, force field, or simulation parameterization, simplification, or inaccuracy occurs, it would cause inappropriate prediction and misleading conclusion. Thus, it is important to use rigorous validation, benchmark, and quality control measures for quality control of the drug’s accuracy. Furthermore, accurate process modeling is important when the system interaction is inadequate, such as hydrogen bond, van der Waals interaction, hydrophobic effect. Researchers can develop force fields, parameterization scheme, and sampling method for QM by improving the force field, parameterization scheme, and sampling methods to make them suitable for drug discovery and development. Experimental data, e.g., X-ray crystallography, nuclear magnetic resonance spectroscopy, and biophysical assays data, are also critical for validating the digital twin model. Simulation result can be checked based on experimental observation, so a reiterative process to validate the digital twin model and re-parametrize the force field.^{28, 29}

Molecular simulations are pivotal in developing digital twins for drug design. They offer the researcher a framework to not only simulate molecular interaction but also analyze and optimize them with higher precision and resolution. Using computational frameworks such as molecular dynamics simulations and complex quantum mechanics calculations empowers the researcher to develop more accurate digital twin that improve the rational design and optimization of drug candidates. Nevertheless, there is a need for validation, integration of experimental results, and simultaneous upgradation of multiple simulations methodologies to ensure that twin simulations are validated and accurate. The advent of digital innovation in the pharmaceutical domain has opened doors to multiple opportunities where digital twins are likely to revolutionize the discovery, design, and delivery of drugs. This would facilitate better outcomes and good health for a patient.^{30, 31} 

Figure 3: This figure illustrates the concept of a digital twin used in molecular simulations, highlighting how virtual models replicate real-world molecular behaviors. Click here to View Figure

Applications of Digital Twins in Pharmacokinetics and Pharmacodynamics

In the rapidly changing world of drug development, knowing the pharmacokinetic and pharmacodynamic characteristics of possible therapeutics is critical for improving efficiency, reducing adverse effects, and guiding clinical decisions. A digital twin is a “computerized model of the structure and operation of a method, product, or a device that can be utilized for different purposes” which can refine our supposition of vital pharmacokinetic and pharmacodynamic measures. Before we dive into how this new technology has changed our understanding of the pharmacokinetic and pharmacodynamic bases of drug behavior and the trajectory of drug discovery, a few words on what digital twins are and how they function. Drug interactions-primary, secondary with one or multiple drugs or with curated networks of proteins affect pharmacological outcomes. Digital twins are utilized in predicting and monitoring these phenomena. Digital twins are essential to evaluating the performance of drug-drug interactions numerically. A digital twin is a computer program created based on the structure and operating principles of an object or biological system. It allows specialists to simulate the physiological processes similar to bodily functions. In the area of drug design, digital twins are used to estimate the effect of a drug on the human body. It may be a priori to the phase of laboratory research on living organisms or may be based on the results of previous research. A digital twin can be used to predict the effect of a drug on the metabolism of molecules or cells. Performance is assessed by comparing digital twin output to monitoring outcomes. If they correspond, the result is labelled as credible. This can be used before and after the drug is accepted.^{32, 33}

The fate and efficacy of a drug within the body are entirely determined by drug metabolism. Therefore, studying the metabolic pathways, enzyme structures, and kinetics is essential. In addition, it helps predict pharmacokinetic parameters, diagnose potential drug-drug interactions, and refine drug design. Digital twins are a valuable technology for creating metabolic models of drug metabolism. They use structural biology data, enzyme kinetics, and computational techniques to model the phenomenon. In drug development, digital twins help replicate metabolic pathways of drug metabolism and predict the rate of metabolite formation, enzyme kinetics, and metabolic stability profiles. By evaluating drugs’ effects on metabolic enzymes such as cytochrome P450s, UDP-glucuronosyltransferases, and sulfotransferases, researchers may determine whether a given drug will be metabolized, detoxified, or bioactivated . In turn, this will aid in identifying drug candidates with acceptable metabolic profiles in the development pipeline. The concept of digital twins also aids in assessing metabolic diversity among people and population groups. For example, scientists can conduct individualized drug metabolism predictions based on genetic polymorphisms, drug-drug interaction, environmental factors. By incorporating genomics, pharmacogenetics, and system biology models data, digital twin development enables patients to tailor drug administration directions for optimum performance.^{34, 35}

Pharmacological effects and therapeutic outcomes of drugs are based on their interactions with molecular targets. Thus, the structural basis of drugs-target interactions, their binding kinetics, and mechanisms of action are crucial for rational drug design and optimization. Digital twins are a promising framework for simulation and analysis of drugs-target interactions. Digital twins integrate the data from structural biology, computational modeling, and experimental sources to build the model of the drugs-target interaction. Among the applications of digital twins is drug design. Researchers may use them to simulate the drug binding to target proteins, predict the binding affinity, and estimate the structure-activity relationship with high accuracy and spatial detail. Researchers also may use molecular docking tools, molecular dynamics simulation, and quantum mechanics calculation approach to consider the conformational dynamics, binding modes, and the energy of drugs-target binding. In this way, digital twins may guide the design of highly potent and selective drugs. Besides, digital twins may help to consider drugs-target interactions in real physiological conditions. Researchers can predict the drug distribution, transport, and accumulation in cellular membranes, organelles, or tissues. In such a way, it is possible to predict the concentration of drugs in the target structures, estimate the potential level of their target’s engagement, and optimize the delivery route to maximize efficacy and minimize toxicity. In addition, digital twins provide a possibility to consider drugs-target interactions in a broader biological network. Researchers can simulate the interaction between drugs and complex biological networks such as signal pathways, protein-protein interaction, or genetic networks. By integration of the data from omics and networks analysis tool and systems biology model, it is possible to predict the drug mode of action, find the novel target for drugs, and develop efficient drug combinations. Digital twins are a flexible framework for simulation and prediction of PK/PD properties of drug design. By integration of computation models and structural biology and experimental, digital twins enable to explore the drug interaction, model drug metabolism, and simulate drug-target interaction. Therefore, under the umbrella of digital innovation at the pharmaceutical market, digital twins are expected to revolutionize the process of drug discovery, design, and delivery and improve the quality of patient care.^36-39 

Figure 4: Advantages of Using Digital Twins in Drug Design This figure illustrates the key benefits of implementing digital twin technology in pharmaceutical research, including accelerated drug development, reduced costs, enhanced prediction of drug efficacy and safety, improved personalization, and minimized clinical trial failures. Click here to View Figure

Advantages of Using Digital Twins in Drug Design

The search for novel pharmaceuticals necessitates a high degree of novelty, accuracy, and efficiency. Digital twins, as a result, have proven to be a game-changing technology in this field, with several benefits for drug design and development. Digital twins provide a virtual copy of biological systems and drug candidates, thus significantly changing the landscape of traditional drug discovery approaches. Given the value of digital twins to drug design, this essay focuses on its benefits, which are the ability to boost drug development timelines, reduce expenses and resource commitments, and enhance drug safety and efficacy. There are several benefits to utilizing digital twins in drug design, including accelerating drug development timelines. Drug discovery is traditionally a time-consuming process, with several procedures, testing iterations, and regulatory barriers before a drug can be brought to market. Still, digital twins can make the process much more accessible by providing a virtual chemistry for testing and improving the process. Digital twins are based on complex computer models like molecular dynamics simulations and machine learning that aid in predicting how a drug would function, communicate, and be effective. Researchers can investigate large chemical space, evaluate hundreds of potential drugs, and choose promising ones in a fraction of the time spent on conventional work by using these models. Not only does it save time, but it also makes it simpler and quicker to locate good prospects. It also transforms how clinical tests are done since digital twins allow researchers to simulate clinical tests. Building a digital twin network promotes communication and information sharing between researchers, allowing for faster dissemination of fascinating projects. Allowing scientists to use digital twins as a foundation for cooperation encourages research and allows persons to learn from one another, resulting in quicker drug release and a more effective drug discovery process. In addition to improving the drug development timetable, this approach also lowers costs. Traditional drug development is expensive: many possible drugs are developed and experimentally validated using the time-consuming and expensive process. However, digital twins help scientists do computational experiments, limiting the need for experiments in laboratory settings.^{40, 41}

Another critical advantage of digital twins in drug design is their ability to enhance drug safety and efficacy by predicting and optimizing drug properties and interactions in silico. More often than not, traditional drug discovery processes face various safety incidents, drug reactions and interactions, adverse drug reactions and other safety-related human consequences. Despite drug adaptation and wastage processes, the impact is severe, particularly in the drug discovery and development process. Thus, the use of digital twins enables researchers to simulate and predict conceivable safety occurrences, the reaction to the drug, and human responses more accurately and explicitly. Integrations of digital twins with pharmacokinetic and pharmacodynamic data, coupled with contextual associations among such data using structural information and additional computational approaches, assess the chemical compound’s safety profile, therapeutic index, and off-target attributes before embarking on a clinical trial. It further facilitates the optimization of chemical compound approaches, including dosage, formulation approaches, and delivery mechanisms, to lessen off-target attributes and optimize the design concept. Moreover, digital twins promote the discovery of personalized medicine by simulating drug reaction on virtual patients. This enabled researchers to acquire electronic health records, omics, and available clinical and genetic data to tailor drug therapy to specific patient needs. Thus, the use of digital twins offers multidimensional benefits to drug design, including saving time and costs and enhanced the drug’s safety and efficacy profile. Finally, the use of a digital twin variant will eliminate the need for numerous illicit drug testing by providing clear replicas of biological systems.^42-44 

Figure 5: This illustration depicts the role of digital twins in pharmacokinetics and pharmacodynamics, showcasing how virtual models simulate drug absorption, distribution, metabolism, and excretion, as well as drug-target interactions. Click here to View Figure

Figure 6: This Figure illustrates how digital twin models simulate and analyze complex interactions between drug molecules and biological systems, including target binding, metabolic pathways, and physiological responses. Click here to View Figure

Challenges and Limitations

Digital twins have become one of the most exciting technologies with enormous potential to transform drug design in an age of digital innovation. While the potential of biological digital twins is immense, there are also significant challenges to overcome. Ensuring the accuracy and reliability of these virtual models requires access to vast amounts of high-quality biological data, as well as continuous validation against real-world observations. Below we discuss the critical challenges that may prevent the widespread adoption and implementation of digital twins in drug design, including data quality and availability, computational performance and resources, and validation and regulatory acceptance. For instance, one of the significant challenges in the application of digital twins for drug design concerns data quality and availability; that is, due to the dependence on vast numbers of data sources, including structural biology data, genomics, pharmacology, and several clinical data types to develop accurate models of biological systems and drug candidates. The access to high-quality, comprehensive, and well-curated datasets commonly represents a bottleneck in several drug discovery efforts because the availability of datasets is rather restricted by factors like data silos, proprietary datasets, and privacy concerns. Also, the quality of the source datasets has been questioned, for example, large gaps, records, errors, and other standardization issues imply that many well-curated datasets are shallow, inconsistent, and irritatingly heterogeneous. Such factors have made it extremely hard to build digital twins and validate them. Therefore, the paper suggests that a lot more needs to be done to address such limitations; for example, entities should drastically improve the way they share, standardize, and curate the information they collect. Additionally, public-private initiatives and government funds have gathered substantial support to set up massive data consortia and open-access databases. Moreover, enhanced machine learning systems in data analytics have advanced enough to correct errors and gaps and improve the quality of data significantly.^{45, 46}

Apart from that, another challenge related to the use of digital twins in drug design is the need for vast computational power and resources necessary to create and run simulations. Digital twins rely on robust computational strategies, including molecular dynamics simulations, quantum mechanics calculations, and machine learning algorithms, to accurately and comprehensively model biological systems and predict drug properties. However, most of these methods consumed enormous resources comprising high-performance computing infrastructure, specialized software, and human expertise. HPC, for instance, is limited by various factors, including high costs, limited availability and scalability. In addition, digital twin simulations are large and complex, necessitating challenges in algorithm development, parallelization, and optimization. Moreover, the rate of technological development is rapid and requires continuous investment in hardware, software and education. Addressing these challenges required strategic investment in HPC, cloud and parallel processing technologies. Finally, collaborative activities like joint access to advanced computing resources, distributed computing networks and cloud providers should be utilized. Moreover, the new field of algorithm development and the optimization of parallel architectures must be used to address challenges associated with digital twin simulations.^{47, 48}

In conclusion, a major bottleneck to the adoption of digital twins in drug design is the validation and regulatory acceptance of virtual models for making decisions in drug development. Regulatory bodies, such as the U.S. Food and Drug Administration and the European Medicines Agency, require rigorous proof of the accuracy, reliability, and predictivity of digital twin models to approve decision-making in drug discovery and development. Model validation involves demonstrating the model’s ability to accurately simulate and predict biological processes, drug properties, and clinical outcomes. This includes rigorous testing against experimental data, benchmarking studies, and sensitivity analyses under diverse conditions and perturbations. Furthermore, validation should address and quantify the uncertainties, variabilities, and limitations present in digital twin simulations with full transparency and accountability. To achieve regulatory acceptance, researchers, regulatory bodies, and industry stakeholders should work together to develop guidelines, standards, and best practices for model validation and qualification. Additionally, transparency, reproducibility, and thorough documentation of digital twin models are critical for building the confidence and trust of regulatory agencies. Finally, awareness creation and capacity-building efforts, such as education and training programs, will help equip researchers and regulators with the knowledge and skills to navigate the complexities of digital twin regulatory approval processes. While digital twins present a promising paradigm for drug design, they do not come without their fair share of challenges, limitations. Resolving these bottlenecks is a priority to unlock the full potential of digital twins in disrupting the pharmaceutical industry for accelerated drug discovery, improved patient outcomes, and a more efficient healthcare system in the 21^st  century.^49-51

Figure 7: It illustrates key challenges and limitations associated with implementing digital twin technology in drug design, including data quality issues, high computational demands, validation complexities, and regulatory barriers. Click here to View Figure

Future Perspectives and Emerging Trends

With the pharmaceutical sector’s continuous evolution, digital twins are likely to be at the forefront of drug design and synthesis revolution. In this article, we present the future perspectives of digital twin technology with respect to drug design. We will consider emerging trends on the use of digital twin technology in drug design, potentials, possible improvements, possible combination with artificial intelligence and machine learning, regulatory concerns, and challenges of adoption. The future outlook for digital twin technology in drug design is full of potential, aimed at bringing new changes to the pharmaceutical industry. Firstly, computational models involved in digital twins such as quantum mechanics calculations, molecular dynamics simulation, and machine learning are expected to be more accurate and become better than before hence projection of complex biological processes will henceforth be with a high level of fidelity. In addition, the rate of predictions and the accuracy will also improve as predictive factors that make use of multiomic data and real-world data will be considered. This will enable the researchers to create process-based images of disease mechanism as well as pharmacological physiological models which will ultimately help in personalized medicine. Lastly, there will be advances in high-performance computing and allowance to new architecture such cloud computing, and high-performance computing that will enable additional complexity challenge to be completed.^{52, 53}

As a result, artificial intelligence and machine learning are promising to revolutionize digital twin technology for drug design. In particular, AI and ML algorithms improve the ability to analyze massive amounts of data, discover patterns, and derive actionable inferences. Consequently, digital twins gain better predictive power and efficiency in pharmaceutical R&D, enabling multiple new use cases. In the future, AI and ML will help to optimize drug design with improved potency, selectivity, and pharmacokinetic properties. For example, AI and ML algorithms learn from chemical structures, as well as biological and clinical data to recommend the right lead molecules, predict their pharmacological properties, and inspire the design of novel therapeutics. Furthermore, AI and ML systems can provide new, previously unknown information on the relationships, correlations, and causalities between variables in complex computer models of disease and drug action. AI and ML algorithms will help optimize patient selection through the use of large-scale omics-data, network analysis, and systems biology to identify novel biomarkers, targets, and drug combinations for personalized medicine. Lastly, AI and ML algorithms will also address the experimental design and various computational tasks to improve dramatically and reduce errors in drug research and development.^{49, 54, 55} 

Figure 8: The future perspectives of digital twins in drug design, highlighting advancements in AI integration, personalized medicine, real-time simulations, and regulatory adaptation. Click here to View Figure

Given the promising technological advancements, integration with AI and ML, and regulatory considerations discussed in this paper, digital twins have a tremendous potential to transform the pharmaceutical industry and revolutionize scientific research and drug development in the 21 st century. However, several regulatory considerations and adoption challenges need to be addressed to realize the full impact of digital twins on the pharmaceutical sector and botany drug discovery and development. Regulatory agencies such as the U.S. Food and Drug Administration and the European Medicine agency require rigorous evidence of the accuracy, reliability, and predictive validity of digital twin models used to design and develop new drugs. Validation of digital twins involves the demonstration of their ability to simulate accurately and predict relevant biological processes, drug properties, and clinical outcomes against experimental data, benchmarking studies, and sensitivity analyses to assess model performance under various conditions and generalizations. Additionally, validation efforts must address uncertainties, variability limitations, and other sources of errors common in digital twin simulations make the developers accountable and transparent in the model validation and analysis process. Furthermore, data privacy, security, and interoperability adoption challenges must be addressed to promote the mass adoption of the digital twin target solution for drug design. Pharmaceutical companies and research institutions should develop a data governance framework, secure data sharing agreements, and standardized data formats, and implement the infrastructure and expertise necessary to meet regulatory requirements and protect. Additionally, collaboration between research institutions, pharmaceutical companies, and government regulatory agencies is essential to address the technical barriers to adoption, share best practices, and conduct further research on digital twin technology.^56-58

Conclusion

In the realm of drug design, digital twins stand as a beacon of innovation, offering transformative solutions to age-old challenges. Throughout this article, we have explored the multifaceted applications of digital twins in pharmaceutical research and development, from accelerating drug discovery to personalizing medicine. As we conclude, let us recap the key points discussed, reflect on the importance of digital twins in shaping the future of drug design, and ponder the potential impact of this technology in healthcare.

Digital twins have emerged as virtual replicas of biological systems and drug candidates, enabling researchers to simulate, analyze, and optimize drug properties with unprecedented accuracy and efficiency. Advantages of digital twins in drug design include accelerating drug development timelines, reducing costs and resource requirements, and improving drug safety and efficacy. Challenges and limitations in adopting digital twins for drug design include data quality and availability, computational power and resources, and validation and regulatory acceptance. Case studies and examples highlight successful applications of digital twins in drug design, ranging from predicting protein-ligand interactions to personalizing cancer therapies and repurposing existing drugs for new indications. Future perspectives and emerging trends in digital twin technology for drug design include potential advancements, integration with artificial intelligence and machine learning, and regulatory considerations and adoption challenges. Importance of Digital Twins in Shaping the Future of Drug Design.

Digital twins hold immense promise in shaping the future of drug design by revolutionizing traditional approaches and unlocking new possibilities for innovation. By providing virtual replicas of biological systems and drug candidates, digital twins offer researchers unprecedented insights into the complexities of disease mechanisms, drug interactions, and patient responses. The integration of digital twins with advanced computational techniques, such as molecular dynamics simulations, quantum mechanics calculations, and machine learning algorithms, empowers researchers to explore vast chemical spaces, optimize drug design parameters, and predict clinical outcomes with greater accuracy and efficiency. Moreover, digital twins enable researchers to embrace personalized medicine approaches by tailoring drug treatments to individual patient profiles, maximizing efficacy and minimizing adverse effects. By leveraging multiomic data, electronic health records, and real-world evidence, digital twins facilitate the identification of biomarkers, therapeutic targets, and drug combinations for precision medicine interventions. Ultimately, digital twins have the potential to accelerate the pace of drug discovery, improve patient outcomes, and advance healthcare in the 21st century. As the pharmaceutical industry continues to embrace digital innovation, digital twins will play an increasingly pivotal role in shaping the future of drug design and personalized medicine.

In conclusion, digital twins hold the promise of revolutionizing healthcare by providing researchers, clinicians, and patients with powerful tools for understanding, predicting, and optimizing disease mechanisms and therapeutic interventions. By offering virtual replicas of biological systems and drug candidates, digital twins empower stakeholders to make informed decisions, prioritize research priorities, and improve patient outcomes. As digital twins continue to evolve and mature, their impact on healthcare is expected to grow exponentially, shaping the future of medicine and transforming the way diseases are diagnosed, treated, and managed. By fostering collaboration, innovation, and transparency, digital twins have the potential to usher in a new era of precision medicine, where treatments are tailored to the individual needs of patients, leading to improved health outcomes and enhanced quality of life for all.

Acknowledgement

The authors thank Institute of Engineering & Management Kolkata, University of Engineering & Management Kolkata for providing Financial Support and all necessary facilities required for carrying out the research successfully. The Authors specially thank Dr. Satyajit Chakraborti, Director, IEM Kolkata for motivating towards working on Digital Twins.

Funding Sources

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Conflict of Interest

The author(s) do not have any conflict of interest.

Data Availability Statement

The images used in this paper are generated using DALL-E 3 AI Image generator using specific prompts.

References

1. Mariam, Z., S.K. Niazi, and M. Magoola, Unlocking the Future of Drug Development: Generative AI, Digital Twins, and Beyond. BioMedInformatics, 2024. 4(2): p. 1441-1456.DOI: https://doi.org/10.3390/biomedinformatics4020079.
   CrossRef
2. Bordukova, M., et al., Generative artificial intelligence empowers digital twins in drug discovery and clinical trials. Expert Opinion on Drug Discovery, 2024. 19(1): p. 33-42.DOI: https://doi.org/10.1080/17460441.2023.2273839.
   CrossRef
3. Moingeon, P., et al., Virtual patients, digital twins and causal disease models: Paving the ground for in silico clinical trials. Drug discovery today, 2023: p. 103605.DOI: https://doi.org/10.1016/j.drudis.2023.103605.
   CrossRef
4. Katsoulakis, E., et al., Digital twins for health: a scoping review. NPJ Digital Medicine, 2024. 7(1): p. 77.DOI: https://doi.org/10.1038/s41746-024-01073-0.
   CrossRef
5. Venkatesh, K.P., G. Brito, and M.N. Kamel Boulos, Health digital twins in life science and health care innovation. Annual Review of Pharmacology and Toxicology, 2024. 64: p. 159-170.DOI: https://doi.org/10.1146/annurev-pharmtox-022123-022046.
   CrossRef
6. Grieves, M.W., Digital twins: past, present, and future, in The Digital Twin. 2023, Springer. p. 97-121.
   CrossRef
7. Javaid, M., A. Haleem, and R. Suman, Digital twin applications toward industry 4.0: A review. Cognitive Robotics, 2023. 3: p. 71-92.DOI: https://doi.org/10.1016/j.cogr.2023.04.003.
   CrossRef
8. Juarez, M.G., V.J. Botti, and A.S. Giret, Digital twins: Review and challenges. Journal of Computing and Information Science in Engineering, 2021. 21(3): p. 030802.DOI: https://doi.org/10.1115/1.4050244.
   CrossRef
9. Fonseca, L.L., L. Böttcher, B. Mehrad, and R.C. Laubenbacher, Metamodeling and Control of Medical Digital Twins. arXiv preprint arXiv:2402.05750, 2024.
10. Boetttcher, L., L.L. Fonseca, and R. Laubenbacher, Control of medical digital twins with artificial neural networks. bioRxiv, 2024: p. 2024.03. 18.585589.DOI: https://doi.org/10.1098/rsta.2024.0228.
    CrossRef
11. Zheng, H., T. Liu, J. Liu, and J. Bao, Visual analytics for digital twins: a conceptual framework and case study. Journal of Intelligent Manufacturing, 2024. 35(4): p. 1671-1686.DOI: https://doi.org/10.1007/s10845-023-02135-y.
    CrossRef
12. Tao, F., H. Zhang, and C. Zhang, Advancements and challenges of digital twins in industry. Nature Computational Science, 2024. 4(3): p. 169-177.DOI: https://doi.org/10.1038/s43588-024-00603-w.
    CrossRef
13. Chen, Y., et al., Digital twins in pharmaceutical and biopharmaceutical manufacturing: a literature review. Processes, 2020. 8(9): p. 1088.DOI: https://doi.org/10.3390/pr8091088.
    CrossRef
14. Subasi, A. and M.E. Subasi, Digital twins in healthcare and biomedicine. Artificial Intelligence, Big Data, Blockchain and 5G for the Digital Transformation of the Healthcare Industry, 2024: p. 365-401.DOI: https://doi.org/10.1016/B978-0-443-21598-8.00011-7.
    CrossRef
15. Vallée, A., Envisioning the Future of Personalized Medicine: Role and Realities of Digital Twins. Journal of Medical Internet Research, 2024. 26: p. e50204.DOI: 10.2196/50204.
    CrossRef
16. Subramanian, K., Digital twin for drug discovery and development—the virtual liver. Journal of the Indian Institute of Science, 2020. 100(4): p. 653-662.DOI: https://doi.org/10.1007/s41745-020-00185-2.
    CrossRef
17. Laubenbacher, R., et al., Building digital twins of the human immune system: toward a roadmap. NPJ digital medicine, 2022. 5(1): p. 64.DOI: https://doi.org/10.1038/s41746-022-00610-z.
    CrossRef
18. Kamel Boulos, M.N. and P. Zhang, Digital twins: from personalised medicine to precision public health. Journal of personalized medicine, 2021. 11(8): p. 745.DOI: https://doi.org/10.3390/jpm11080745.
    CrossRef
19. Eftimie, R., A. Mavrodin, and S.P. Bordas, From digital control to digital twins in medicine: A brief review and future perspectives. Advances in Applied Mechanics, 2023. 56: p. 323-368.DOI: https://doi.org/10.1016/bs.aams.2022.09.001.
    CrossRef
20. Ali, R.S.A.E., J. Meng, M.E.I. Khan, and X. Jiang, Machine learning advancements in organic synthesis: A focused exploration of artificial intelligence applications in chemistry. Artificial Intelligence Chemistry, 2024. 2(1): p. 100049.DOI: https://doi.org/10.1016/j.aichem.2024.100049.
    CrossRef
21. Struble, T.J., et al., Current and future roles of artificial intelligence in medicinal chemistry synthesis. Journal of medicinal chemistry, 2020. 63(16): p. 8667-8682.DOI: https://doi.org/10.1021/acs.jmedchem.9b02120.
    CrossRef
22. Strieth-Kalthoff, F., et al., Artificial intelligence for retrosynthetic planning needs both data and expert knowledge. Journal of the American Chemical Society, 2024. 146(16): p. 11005-11017.DOI: https://doi.org/10.1021/jacs.4c00338.
    CrossRef
23. Loeffler, H.H., et al., Reinvent 4: Modern AI–driven generative molecule design. Journal of Cheminformatics, 2024. 16(1): p. 20.DOI: https://doi.org/10.1186/s13321-024-00812-5.
    CrossRef
24. Behle, E., J.M. Herold, and A.H. Schug, Towards cellular digital twins of in vivo tumors. Biophysical Journal, 2023. 122(3): p. 301a-302a.DOI: 10.1016/j.bpj.2022.11.1700.
    CrossRef
25. Cellina, M., et al., Digital Twins: The New Frontier for Personalized Medicine? Applied Sciences, 2023. 13(13): p. 7940.DOI: https://doi.org/10.3390/app13137940.
    CrossRef
26. Antipas, G., N. Ntallis, K. Karalis, and L. Miroslaw, Digital twin on the cloud: system-agnostic prediction of pharmaceutical placebo stability via computing-as-a-service and experimental validation. 2023.DOI: https://doi.org/10.21203/rs.3.rs-3001054/v1.
    CrossRef
27. Yang, P.C., et al., A multiscale predictive digital twin for neurocardiac modulation. The Journal of physiology, 2023. 601(17): p. 3789-3812.DOI: https://doi.org/10.1113/JP284391.
    CrossRef
28. Sarantinoudis, N., G. Tsinarakis, P. Dedousis, and G. Arampatzis, Model-Based Simulation Framework for Digital Twins in the Process Industry. IEEE Access, 2023.DOI: 10.1109/ACCESS.2023.3322926.
    CrossRef
29. Silber, K., et al., Accelerating reaction modeling using dynamic flow experiments, part 2: development of a digital twin. Reaction Chemistry & Engineering, 2023. 8(11): p. 2849-2855.DOI: 10.1039/D3RE00244F.
    CrossRef
30. Guo, Y., A. Klink, P. Bartolo, and W.G. Guo, Digital twins for electro-physical, chemical, and photonic processes. CIRP Annals, 2023. 72(2): p. 593-619.
    CrossRef
31. Khalyutin, S., I. Starostin, and I. Agafonkina, Generalized method of mathematical prototyping of energy processes for digital twins development. Energies, 2023. 16(4): p. 1933.DOI: https://doi.org/10.3390/en16041933.
    CrossRef
32. Reali, F., et al. Digital Twins and Virtual Populations: Applications in Quantitative Systems Pharmacology. in BUILD-IT Workshop 2023–BUILding a DIgital Twin: requirements, methods, and applications. 2023. Sl.
33. Möller, J. and R. Pörtner, Digital twins for tissue culture techniques—concepts, expectations, and state of the art. Processes, 2021. 9(3): p. 447.DOI: https://doi.org/10.3390/pr9030447.
    CrossRef
34. Sinisi, S., Artificial intelligence and model checking methods for in silico clinical trials. 2020.
35. Chasseloup, E., A.C. Hooker, and M.O. Karlsson, Generation and application of avatars in pharmacometric modelling. Journal of Pharmacokinetics and Pharmacodynamics, 2023. 50(5): p. 411-423.DOI: https://doi.org/10.1007/s10928-023-09873-9.
    CrossRef
36. Liang, W.S., et al., Emerging therapeutic drug monitoring technologies: considerations and opportunities in precision medicine. Frontiers in Pharmacology, 2024. 15: p. 1348112.DOI: https://doi.org/10.3389/fphar.2024.1348112.
    CrossRef
37. Anthwal, A., et al., Industry 4.0 Technologies Adoption for Digital Transition in Drug Discovery and Development: A Review. Journal of Industrial Information Integration, 2024: p. 100562.DOI: https://doi.org/10.1016/j.jii.2024.100562.
    CrossRef
38. Venkatapurapu, S.P., M.T. DeWitt, M. Behar, and P.M. D’Alessandro, Digital Twins for Proactive and Personalized Healthcare–Challenges and Opportunities. Handbook of Digital Twins: p. 888-902.DOI: https://doi.org/10.1201/9781003425724.
    CrossRef
39. Aravindakshan, M.R., C. Mandal, A. Pothen, and C. Maass, DigiLoCS: A Leap Forward in Predictive Organ-on-Chip Simulations. bioRxiv, 2024: p. 2024.03. 28.587123.DOI: https://doi.org/10.1371/journal.pone.0314083.
    CrossRef
40. Zhang, Y., et al., A framework towards digital twins for type 2 diabetes. Frontiers in Digital Health, 2024. 6: p. 1336050.DOI: https://doi.org/10.3389/fdgth.2024.1336050.
    CrossRef
41. Balasubhramanyam, A., R. Ramesh, R. Sudheer, and P.B. Honnavalli, Revolutionizing Healthcare: A Review Unveiling the Transformative Power of Digital Twins. IEEE Access, 2024.DOI: 10.1109/ACCESS.2024.3399744.
    CrossRef
42. Tiwari, S. and A. Shukla, Digital Twins in Industry: Real-World Applications and Innovations, in Transforming Industry using Digital Twin Technology. 2024, Springer. p. 1-18.
    CrossRef
43. De Domenico, M., et al., Challenges and opportunities for digital twins in precision medicine: a complex systems perspective. arXiv preprint arXiv:2405.09649, 2024.DOI: https://doi.org/10.1038/s41746-024-01402-3.
    CrossRef
44. Jamshidi, M.B., M. Ebadpour, and M.M. Moghani. Cancer digital twins in metaverse. in 2022 20th International Conference on Mechatronics-Mechatronika (ME). 2022. IEEE.
    CrossRef
45. Sharma, A., et al., Digital twins: State of the art theory and practice, challenges, and open research questions. Journal of Industrial Information Integration, 2022. 30: p. 100383.DOI: https://doi.org/10.1016/j.jii.2022.100383.
    CrossRef
46. Tzachor, A., et al., Potential and limitations of digital twins to achieve the sustainable development goals. Nature Sustainability, 2022. 5(10): p. 822-829.DOI: https://doi.org/10.1038/s41893-022-00923-7.
    CrossRef
47. AlBalkhy, W., et al., Digital twins in the built environment: Definition, applications, and challenges. Automation in Construction, 2024. 162: p. 105368.DOI: https://doi.org/10.1016/j.autcon.2024.105368.
    CrossRef
48. Helbing, D. and J.A. Sánchez-Vaquerizo, Digital twins: potentials, ethical issues and limitations, in Handbook on the politics and governance of Big Data and Artificial Intelligence. 2023, Edward Elgar Publishing. p. 64-104.
    CrossRef
49. of Engineering, N.A., et al., Opportunities and Challenges for Digital Twins in Biomedical Research. 2023.DOI: https://doi.org/10.17226/26922.
    CrossRef
50. Tang, C., et al., A roadmap for the development of human body digital twins. Nature Reviews Electrical Engineering, 2024: p. 1-9.DOI: https://doi.org/10.1038/s44287-024-00025-w.
    CrossRef
51. Talukder, A.K., S.N. Sarbadhikari, E. Selg, and R.E. Haas, Digital Twins, Digital Triplets, and eXplainable AI, in Precision Health, in Roles and Challenges of Semantic Intelligence in Healthcare Cognitive Computing. 2024, IOS Press. p. 1-30.
    CrossRef
52. Askr, H., et al., Deep learning in drug discovery: an integrative review and future challenges. Artificial Intelligence Review, 2023. 56(7): p. 5975-6037.DOI: https://doi.org/10.1007/s10462-022-10306-1.
    CrossRef
53. Sun, T., et al., The digital twin in medicine: a key to the future of healthcare? Frontiers in Medicine, 2022. 9: p. 907066.DOI: https://doi.org/10.3389/fmed.2022.907066.
    CrossRef
54. Sayam, A.G., M. Pradhan, and A.K. Choudhury, Artificial Intelligence the Futuristic Technology in the Drug Discovery Process: A Review. Journal of Young Pharmacists, 2023. 15(3): p. 390-396.DOI: 10.5530/jyp.2023.15.54.
    CrossRef
55. Sahal, R., S.H. Alsamhi, and K.N. Brown, Personal digital twin: a close look into the present and a step towards the future of personalised healthcare industry. Sensors, 2022. 22(15): p. 5918.DOI: https://doi.org/10.3390/s22155918.
    CrossRef
56. Armeni, P., et al., Digital Twins for Health: Opportunities, Barriers and a Path Forward. 2023.DOI: 10.5772/intechopen.112490.
    CrossRef
57. Rasheed, A., O. San, and T. Kvamsdal, Digital twin: Values, challenges and enablers from a modeling perspective. IEEE access, 2020. 8: p. 21980-22012.DOI: 10.1109/ACCESS.2020.2970143.
    CrossRef
58. Gnanamalar, R.H., Human Digital Twin Processes and their Future Check for updates. Transforming Industry using Digital Twin Technology: p. 187.DOI: https://doi.org/10.1007/978-3-031-58523-4_10.
    CrossRef