Data Science Technologies Research Articles

Advanced semantic segmentation of aircraft main components based on transfer learning and data-driven approach

Abstract The implementation of Smart Airport and Airport 4.0 visions relies on the integration of automation, artificial intelligence, data science, and aviation technology to enhance passenger experiences and operational efficiency. One essential factor in the integration is the semantic segmentation of the aircraft main components (AMC) perception, which is essential to maintenance, repair, and operations in aircraft and airport operations. However, AMC segmentation has challenges from low data availability, high-quality annotation scarcity, and categorical imbalance, which are common in practical applications, including aviation. This study proposes a novel AMC segmentation solution, employing a transfer learning framework based on a sophisticated DeepLabV3 architecture optimized with a custom-designed Focal Dice Loss function. The proposed solution remarkably suppresses the categorical imbalance challenge and increases the dataset variability with manually annotated images and dynamic augmentation strategies to train a robust AMC segmentation model. The model achieved a notable intersection over union of 84.002% and an accuracy of 91.466%, significantly advancing the AMC segmentation performance. These results demonstrate the effectiveness of the proposed AMC segmentation solution in aircraft and airport operation scenarios. This study provides a pioneering solution to the AMC semantic perception problem and contributes a valuable dataset to the community, which is fundamental to future research on aircraft and airport semantic perception. Graphical abstract

A Study on Functional Requirements and Inspection Items for AI System Change Management and Model Improvement on the Web Platform

The rapid adoption of artificial intelligence (AI) on the web platform across multiple sectors has highlighted not only its inherent technical hurdles, such as unpredictability and lack of transparency, but also significant societal concerns. These include the misuse of AI technology, invasions of privacy, discrimination fueled by biased data, and infringements of copyright. Such challenges jeopardize the sustainable growth of AI and risk the erosion of societal trust, industry adoption and financial investment. This analysis explores the AI system’s lifecycle, emphasizing the essential continuous monitoring and the need for creating trustworthy AI technologies. It advocates for an ethically oriented development process to mitigate adverse effects and support sustainable progress. The dynamic and unpredictable nature of AI, compounded by variable data inputs and evolving distributions, requires consistent model updates and retraining to preserve the integrity of services. Addressing the ethical aspects, this paper outlines specific guidelines and evaluation criteria for AI development, proposing an adaptable feedback loop for model improvement. This method aims to detect and rectify performance declines through prompt retraining, thereby cultivating robust, ethically sound AI systems. Such systems are expected to maintain performance while ensuring user trust and adhering to data science and web technology standards. Ultimately, the study seeks to balance AI’s technological advancements with societal ethics and values, ensuring its role as a positive, reliable force across different industries. This balance is crucial for harmonizing innovation with the ethical use of data and science, thereby facilitating a future where AI contributes significantly and responsibly to societal well-being.

Leveraging Data Science Technology for Advancing Credit Risk Assessment

The evaluation of credit risk (CR) has become prominent in recent years, particularly among banks, as default rates are on the rise and economic insecurity remains persistent. Traditional credit scoring techniques oftentimes are inadequate and provide little means for risk estimation, necessitating the development of new models using data science methodologies. In this study, a novel Intelligent Dwarf Mongoose tuned Light Gradient Boosting Machine (IDM-LGBM) model that boosts the accuracy of CR and improves forecasting performance, is introduced. The Light Gradient Boosting Machine model's hyperparameters were optimized using the Intelligent Dwarf Mongoose technique, improving the model's predictive strength. The CR dataset was gathered from the Kaggle platform. The data is then pre-processed using Z-score normalization. To evaluate the efficiency of the suggested IDM-LGBM technique, which has been implemented employing a Python platform. Results show that the IDM-LGBM model performed significantly better than conventional methods in terms of recall (98.1%), accuracy (97.2%), F1-score (97.4%), and precision (96.5%). Subsequent studies could concentrate on addressing real-time data streams and enhancing models to respond to the changing credit environment.

A Combination of Camera and Pitfall Traps: A Method for Monitoring Ground-Dwelling Invertebrates in Farmlands

Ground-dwelling invertebrates are crucial for maintaining biodiversity, soil health, and food production in agricultural ecosystems; however, their diversity is declining globally. Despite advancements in data science and modern monitoring technology, conventional methods remain predominant in the current field investigations of ground-dwelling invertebrates in farmlands. This study aimed to develop simple, convenient, and automated methods to assess ground-dwelling invertebrate biodiversity in farmlands. Both camera and pitfall traps were used to monitor ground-dwelling invertebrates in two farmlands in Ningbo City, China. Over four experimental periods, lasting 34 days in the vineyard farmland and 32 days in the watermelon farmland, the camera traps detected a relatively high density of invertebrates, although the relative dominance of ant density was overestimated in both farmlands. Both methods revealed similar temporal dynamics of the ground-dwelling invertebrates during the four periods in the vineyard farmland. However, only camera traps recorded the temporal dynamics of the invertebrates at daily, hourly, and minute resolutions in both farmlands. Conclusively, this study demonstrates that combining camera and pitfall traps provides an effective method for monitoring the temporal dynamics of ground-dwelling invertebrates in farmlands, offering high-quality data useful for ecological research and biodiversity protection in agricultural management.

Have Your Cake and Eat It Too

As datasets have become a more significant aspect of Open Science, attention has turned to the data transformations that drive their creation. Li and Ludäscher have pointed out the importance of identifying data cleaning workflows as a series of modular transformations that can be extracted for reuse. This modular approach aids reproducibility and allows for transparency in data provenance. However, the constantly evolving nature of data science technology means that even once these modules have been identified and implemented, their functionality must be ported to new platforms as old ones become less applicable or less common in a field of study. When these transformations take place, it is important to consider not only practicality and functionality, but also transparency within a data processing team. Clarity of communication within a team is the first step towards providing clear and transparent documentation to the end user. This case study of an updated workflow process for a long-running longitudinal health and well-being study provides practical examples of these principles.

МЕТОДИ І ПРАКТИКИ ВПРОВАДЖЕННЯ ТА ЗАСТОСУВАННЯ РІЗНИХ ТИПІВ ШТУЧНОГО ІНТЕЛЕКТУ І НАУКИ ПРО ДАНІ ДЛЯ РІШЕННЯ ПРОБЛЕМ У ВОДНІЙ ІНФРАСТРУКТУРІ

The subject of the study in this article was the practice of implementing and applying various types of artificial intelligence and data science to detect leaks from water supply networks, to monitor freshwater conditions and to detect pollution, clean freshwater bodies from waste from industrial and mining enterprises, control freshwater flow and develop more efficient water filtration methods. The article identifies the advantages and disadvantages of the practice of introducing and applying various types of artificial intelligence and data science technologies to automate traditional methods of monitoring, control and related work in the water industry. The article addresses the following objectives: to substantiate the effectiveness of implementation of various types of artificial intelligence technologies and data science methods and their practical application with software and hardware technologies to automate traditional methods of performing work in the water industry. To solve the tasks set, the methodology used was based on general scientific and special research methods, such as theoretical methods (analysis, explanation, generalisation, comparison). The use of this approach allowed us to obtain the following results: the features that affect the accuracy of the analysis of collected data used by technologies of various types of artificial intelligence and data science methods were identified. Practices and methods for more efficient and accurate application of this technology are reflected. Scientific data are analysed. The study allowed identifying the practical opportunities and problems of this technology. Recommendations for the effective use of this technology have been developed. The factors that influence the effective use of this technology in industry are identified.

Mapping cognitive activity from electrocorticography field potentials in humans performing NBack task

Objective. Advancements in data science and assistive technologies have made invasive brain-computer interfaces (iBCIs) increasingly viable for enhancing the quality of life in physically disabled individuals. Intracortical microelectrode implants are a common choice for such a communication system due to their fine temporal and spatial resolution. The small size of these implants makes the implantation plan critical for the successful exfiltration of information, particularly when targeting representations of task goals that lack robust anatomical correlates. Approach. Working memory processes including encoding, retrieval, and maintenance are observed in many areas of the brain. Using human electrocorticography (ECoG) recordings during a working memory experiment, we provide proof that it is possible to localize cognitive activity associated with the task and to identify key locations involved with executive memory functions. Results. From the analysis, we could propose an optimal iBCI implant location with the desired features. The general approach is not limited to working memory but could also be used to map other goal-encoding factors such as movement intentions, decision-making, and visual-spatial attention. Significance. Deciphering the intended action of a BCI user is a complex challenge that involves the extraction and integration of cognitive factors such as movement planning, working memory, visual-spatial attention, and the decision state. Examining field potentials from ECoG electrodes while participants engaged in tailored cognitive tasks can pinpoint location with valuable information related to anticipated actions. This manuscript demonstrates the feasibility of identifying electrodes involved in cognitive activity related to working memory during user engagement in the NBack task. Devoting time in meticulous preparation to identify the optimal brain regions for BCI implant locations will increase the likelihood of rich signal outcomes, thereby improving the overall BCI user experience.

Data Science Techniques for Predictive Analytics in Financial Services

Business intelligence has become one of the crucial technologies in financial services and with the help of predictive analytics; the organisations have been able to improve a lot in various fields. In this paper, the author focuses on the role of prediction in financing, as well as explaining the main data science approaches and technologies that support innovations in this area. In more detail, the specifics of using the predictive model in the estimation of risks and in investment are discussed through case studies, which allow to reveal the advantages and positive results of the application. In addition, the paper a examines the issues that arise in the implementation of financial data analytics including data issues, compliance and growth factors. Through a consideration of these issues, we hope to orchestrate the best understanding of today and tomorrow’s perspectives on prediction analysis for financial services.

An Effective Lung Cancer Diagnosis Model Using the CNN Algorithm

The disease known as lung cancer is a serious condition that may be deadly if it is not diagnosed at an early stage. The diagnosis of lung cancer has to be improved, and there is a need for a cost-effective and user-friendly system that leverages state-of-the-art data science technology. This would help simplify operations, save time and money, and improve diagnosis. This research suggests the use of a convolutional neural network (CNN) architecture for the purpose of categorizing three unique histopathological pictures, namely benign, adenocarcinoma, and squamous cell carcinoma. The purpose of this study is to apply the CNN model to properly classify these three kinds of cancers and to compare the accuracy of the CNN model to the accuracy of other techniques that have been employed in investigations that are comparable to this one. The CNN model was not used in any of the preceding research for the purpose of categorizing these particular histopathological pictures; hence, the relevance of this work cannot be overstated. It is possible to get more positive treatment results by correctly classifying malignant tumors as early as possible. In training, the CNN model obtained an accuracy of 96.11%, and in validation, it earned an accuracy of 97.2%. The suggested method has the potential to improve lung cancer diagnosis in patients by classifying them into subgroups according to the symptoms they exhibit. This approach to machine learning, which makes use of the random forest technique, has the potential to reduce the amount of time, resources, and labor required. Utilizing the CNN model to categorize histopathological pictures may, ultimately, improve the diagnostic accuracy of lung cancer and save lives by allowing early disease identification.

The method of assessing the effectiveness of the security of confidential data of the distributed information system

The paper proposes a method of assessing the effectiveness of the security of confidential data of a distributed information system, based on a model for determining the current threats to the security of confidential data in the information system, on algorithms of fuzzy inference and the theory of fuzzy neural systems, unlike known ones, it uses sufficient and necessary indicators, excludes expert errors , increases the detection of actual threats to the security of confidential information system data by 5%, reduces the cost of purchasing information protection tools from 15 to 30%. It takes into account the following factors: the IT infrastructure of the distributed information system, the capabilities of attackers and their level of motivation in the information system. The proposed approach differs from existing ones by an automated process, the need to involve highly qualified specialists in the field of information security, and low computational complexity; absence of deficiencies in expert assessments; allows you to determine the list of current information security threats in information systems of various classes and types. The task of ensuring the security of confidential data is urgent, which is due to the growth of computer attacks and leaks of information, which are reflected in the statistical data on the commission of crimes in the field of high technologies, the growth of criminal activity using modern communication devices and the Internet. The existing methods of identifying current threats to information security and assessing the effectiveness of the security of confidential data cannot be used at all stages of the life cycle of distributed information systems, they do not take into account the following indicators in the complex: IT infrastructure of distributed systems, current threats to information security, security requirements of confidential data, their cost as important indicators when solving these problems. One of the most important tasks of ensuring the security of confidential data is the evaluation of the effectiveness of the protection system. In this regard, the goal of the research is to improve the quality of the assessment of the effectiveness of the security of confidential data of a distributed information system by determining sufficient and necessary indicators using modern information technologies, which allow the most effective solution of the following tasks: determining the parameters of the adaptive production fuzzy neural systems, which most suitable for solving the tasks, the application of Data Science technologies in data processing, algorithms of fuzzy output.

WiPFIM: A digital platform for interlinking biocollections of wild plants, fruits, associated insects, and their molecular barcodes.

The current knowledge on insects feeding on fruits is limited, and some of the scarce existing data on the fruit-associated insects are secluded within the host institutions. Consequently, their value is not fully realized. Moreover, in countries like Kenya, the integration of biocollections data within a digital framework has not been fully exploited. To address these gaps, this article presents a description of the development of a web-based platform for data sharing and integrating biodiversity historical data of wild plants, fruits, associated insects, and their molecular barcodes (WiPFIM) while leveraging data science technologies. The barcodes corresponding to the biocollections data were retrieved from BOLD database. The platform is an online resource about fruit-insect interactions that can be of interest to a worldwide community of users and can be useful in building innovative tools. The platform is accessible online at https://test-dmmg.icipe.org/wpfhi.

Data science, digital health, big data and smart technologies to support biotechnology – Current perspectives and future challenges

Data science, digital health, big data and smart technologies to support biotechnology – Current perspectives and future challenges

Psychiatryai.com: Real-Time AI Scoping Review (RAISR 4D) in Psychiatry and Mental Health with Live Real-World Evidence and CPD/CME for Psychiatrists

IntroductionPsychiatryai.com was launched in 2021 and initial findings were published at EPA 2023. The portal is an advanced computing science project in Applied Data Science and Evidence-Based Healthcare for my MSc studies at the University of Oxford (Kellogg College). Artificial Intelligence (AI) and Data Science (DS) technology are utilised to analyse live Real-World Evidence (RWE) in Psychiatry and Mental Health from PubMed to provide CPD/CME online. A two-year review of the site and its performance will be presented to EPA 2024.ObjectivesTo develop and study an experimental real-time AI and DS platform in Global Mental Health and Psychiatry, to provide the latest RWE from PubMed for online education and training, and to report findings to EPA 2024 for peer review in Budapest. AI and misinformation are newly identified risks in healthcare (AI Safety Summit 2023). The site also aims to raise awareness about “Aiatrogenesis” to address this problem, with RWE and CPD/CME utilising AI and DS technology for the categorisation and meta-analysis of evidence, rather than the production of possibly misleading or false Generative AI evidence (Monteith et al. BJP 2023; 1-3).MethodsAs reported to EPA 2023 in Paris, a free open-code WordPress site was launched on the 22nd of November 2021 (Psychiatryai.com). The portal has been further developed and now features over 90k pages comprising 7GB of data with Cloudflare security and speed. Live evidence is collected into an open database and research articles are categorised into evidence nodes with AI. The results are presented in a real-time Evidence Matrix and Blueprint, creating 15-minute CPD/CME reflection modules. Data analytics from Psychiatryai.com with Google Analytics (G4A) along with platform insights from two years of development and research will be presented to EPA 2024. The site is conceptualised and designed to be viewed in an interactive VR headset.ResultsLive Citations380000+PubMed Articles Analysed with AI92142CPD/CME23002 hoursAlgorithms/Topics in Psychiatry291Open Data7 GBImage:Image 2:Image 3:ConclusionsPsychiatryai.com has successfully developed a novel AI and DS platform that incorporates the latest research in mental health and psychiatry, providing real-world evidence (RWE) for psychiatrists and healthcare professionals worldwide, along with CPD/CME online. This enhances hypothesis testing in research by presenting a related Evidence Matrix and Blueprint (from the last 365 days) for each evidence node on the site (RAISR 4D). These matrices provide a real-time visual table (8 x 8 / 64) of global research related to the evidence node in the preceding year. The site is VR-ready and has a special focus on AI and Psychiatry, Disaster and Traumatology Sciences, and Youth Mental Health. This project is dedicated to the memory of Dr Denis O’Leary and Dr Navin Venkatraman.Disclosure of InterestNone Declared

A Phase Model of the Bio-Inspired NbOx Local Active Memristor under Weak Coupling Conditions.

For some so-called computationally difficult problems, using the method of Boolean logic is fundamentally inefficient. For example, the vertex coloring problem looks very simple, but the number of possible solutions increases sharply with the increase of graph vertices. This is the difficulty of the problem. This complexity has been widely studied because of its wide applications in the fields of data science, life science, social science, and engineering technology. Consequently, it has inspired the use of alternative and more effective non-Boolean methods for obtaining solutions to similar problems. In this paper, we explore the research on a new generation of computers that use local active memristors coupling. First, we study the dynamics of the memristor coupling network. Then, the simplified system phase model is obtained. This research not only clarifies a physics-based calculation method but also provides a foundation for the construction of customized analog computers to effectively solve NP-hard problems.

Construction of Curriculum System of Data Science and Big Data Technology under the Background of New Engineering

With the rapid development of science and technology, the construction of new engineering has become a hot topic in today s society. The new engineering Department aims to cultivate composite engineering talents with innovative spirit and practical ability to adapt to and promote the rapid economic development and industrial upgrading. Under the background of new engineering, data science and big data technology, as an important discipline direction, have a wide range of application fields and market prospects. Therefore, it is of great significance to build a scientifi c, reasonable and perfect data science and big data technology curriculum system for training high-quality talents to meet the needs of society. This paper will discuss how to build a reasonable curriculum system from the perspective of the job demands of data science and big data technology.

A Machine Learning Model for Prediction of Marine Icing

Abstract Marine icing due to freezing sea spray has been attributed to many safety incidences. Prediction of sea spray icing is necessary for operational safety, design optimization, and structural health. In general, lack of detailed full-scale measurements due to the complexity and costs make validation difficult. The next best option is that of controlled laboratory experiments. The current study is the first study in the field of sea spray icing that investigates the use of new data science technologies like machine learning and feature engineering for the prediction of sea spray icing based on data collected from controlled laboratory experiments. A new prediction model dubbed “Spice” is proposed. Spice is designed “bottom-up” from experimentally collected data, and thus, if the input variables are accurately known, it could be said to be highly accurate within the tested range compared to existing theoretical models. Results from the current study show promising trends; however, more experiments are suggested for increasing the range of confident predictions and reducing the skewness of the training data. Results from spice are compared with five existing models and give icing rates in various conditions in the middle of the spectrum of the other models. It is discussed how validation from two existing full-scale icing measurements from literature proves to be challenging, and more detailed measurements are suggested for the purpose of validation.

Model of current threats to security of confiden-tial data in divisional information systems

The paper proposes a model for determining actual threats to the information security of distributed information systems, based on algorithms of fuzzy inference and the theory of fuzzy neural systems, unlike known ones, uses sufficient and necessary indicators, excludes expert errors, increases the detection of the number of actual threats to information security of distributed systems by 5 %, reduces the cost of purchasing information protection equipment from 15 to 30%. It takes into account the following factors: the IT infrastructure of the distributed information system, the capabilities of attackers and their level of motivation in the distributed information system, the list of existing protection tools in the distributed information system.
 The proposed approach differs from existing ones in the following: lack of involvement of highly qualified specialists in the field of information security; the process is automated, has a low computational complexity; absence of deficiencies in expert assessments; allows you to determine the list of current information security threats in information systems of various classes and types.
 The task of ensuring the security of confidential information becomes the most urgent, which is due to the growth of computer attacks and leaks of information, which are reflected in the statistical data on the commission of crimes in the field of high technologies, the growth of criminal activity using modern communication devices and the Internet.
 Existing methods of modeling current threats to information security and assessing the effectiveness of the information protection system cannot be used at all stages of the life cycle of distributed information systems - they do not take into account the following indicators in the complex: IT infrastructure of distributed information systems, current threats to information security, security requirements for confidential information, a list of means of protecting confidential information and their value as important indicators when solving these problems.
 In order to achieve the goals of ensuring the security of confidential information, it is necessary to: organize the effective creation of an information protection system (information security system), effective modeling (identification of the list) of current threats to information security, identification of the current violator, and also provide the opportunity to conduct a qualitative assessment of the effectiveness of the information security (protection) system.
 One of the most important tasks of ensuring the security of confidential information is the assessment of the effectiveness of the protection (security) system. In this regard, the goal of the work (research) is to improve the quality of the evaluation of the effectiveness of the protection (security) systems of distributed information systems by determining sufficient and necessary evaluation indicators using modern (promising) information technologies that allow the most effective solution of the following tasks: determination operating parameters of adaptive production fuzzy neural systems, which are most suitable for solving the tasks, application of Data Science technologies in data processing, fuzzy output algorithms.

The Intelligent Design of Silicon Photonic Devices

AbstractPhotonic devices based on silicon waveguides are essential to versatile high‐performance and low‐cost photonic integrated systems. Extremely complex silicon photonic devices with hundreds or even thousands of degrees of freedom (DOF) are successfully designed and manufactured based on recent advances in data science and nanofabrication technology. At this level, conventional forward‐reasoning may no longer be suitable for designing high‐performance silicon photonic devices with novel functionalities since the light‐matter interaction is complex and non‐intuitive. Therefore, the timely development of sub‐wavelength silicon photonic devices that can precisely mold the flow of light is a critical and urgent issue requiring joint engineering and scientific efforts. In this paper, an inverse design strategy based on heuristic and gradient descendant algorithms, enabling the realization of large‐scale integrated devices is first introduced. Subsequently, the burgeoning deep learning technology, which offers a promising direction for the automation design of silicon photonics with a data‐driven approach, is discussed. Finally, the obstacles and prospects in this emerging research direction are revealed. Detail discussions from multiple perspectives are provided. This review aims to provide general guidance and a comprehensive reference for scientists developing photonic integrated systems.

Deep-learning framework for optimal selection of soil sampling sites

Soil sampling is one of the most fundamental agricultural processes. The selection of soil sampling locations within a field plays a crucial role in soil-health analysis. The current soil-sampling techniques produce random samples across a field, which are not representative of the field conditions. This work leverages the recent advancements of deep learning in image processing for finding optimal locations that present the important characteristics of a field. The data for training are collected at different fields in local farms with five features: aspect, flow accumulation, slope, NDVI (normalized difference vegetation index), and yield. The soil sampling dataset is challenging because of the unbalance between the number of pixels that represent soil-sampling spots and the number of pixels that represent the background area. In this work, we approach the problem with two methods, the first approach involves utilizing a state-of-the-art model with the convolutional neural network (CNN) backbone, while the second is to innovate a deep-learning design grounded in the concepts of transformer and self-attention. Our framework is constructed with an encoder–decoder architecture with the self-attention mechanism as the backbone. In the encoder, the self-attention mechanism is the key feature extractor, which produces feature maps. In the decoder, we introduce atrous convolution networks to concatenate, fuse the extracted features, and then export the optimal locations for soil sampling. Currently, the model has achieved impressive results on the testing dataset, with a mean accuracy of 99.52%, a mean Intersection over Union (IoU) of 57.35%, and a mean Dice Coefficient of 71.47%, while the performance metrics of the state-of-the-art CNN-based model are 66.08%, 3.85%, and 1.98%, respectively. This indicates that our proposed model outperforms the CNN-based method on the soil-sampling dataset. To the best of our knowledge, our work is the first to provide a soil-sampling dataset with multiple attributes and leverage deep learning techniques to enable the automatic selection of soil-sampling sites. This work lays a foundation for novel applications of data science and machine-learning technologies to solve other emerging agricultural problems.

Digital patient twins for personalized therapeutics and pharmaceutical manufacturing.

Digital twins are virtual models of physical artefacts that may or may not be synchronously connected, and that can be used to simulate their behavior. They are widely used in several domains such as manufacturing and automotive to enable achieving specific quality goals. In the health domain, so-called digital patient twins have been understood as virtual models of patients generated from population data and/or patient data, including, for example, real-time feedback from wearables. Along with the growing impact of data science technologies like artificial intelligence, novel health data ecosystems centered around digital patient twins could be developed. This paves the way for improved health monitoring and facilitation of personalized therapeutics based on management, analysis, and interpretation of medical data via digital patient twins. The utility and feasibility of digital patient twins in routine medical processes are still limited, despite practical endeavors to create digital twins of physiological functions, single organs, or holistic models. Moreover, reliable simulations for the prediction of individual drug responses are still missing. However, these simulations would be one important milestone for truly personalized therapeutics. Another prerequisite for this would be individualized pharmaceutical manufacturing with subsequent obstacles, such as low automation, scalability, and therefore high costs. Additionally, regulatory challenges must be met thus calling for more digitalization in this area. Therefore, this narrative mini-review provides a discussion on the potentials and limitations of digital patient twins, focusing on their potential bridging function for personalized therapeutics and an individualized pharmaceutical manufacturing while also looking at the regulatory impacts.