Opportunities and challenges of 5G network technology toward precision medicine

Abstract

Moving away from traditional “one‐size‐fits‐all” treatment to precision‐based medicine has tremendously improved disease prognosis, accuracy of diagnosis, disease progression prediction, and targeted‐treatment. The current cutting‐edge of 5G network technology is enabling a growing trend in precision medicine to extend its utility and value to the smart healthcare system. The 5G network technology will bring together big data, artificial intelligence, and machine learning to provide essential levels of connectivity to enable a new health ecosystem toward precision medicine. In the 5G‐enabled health ecosystem, its applications involve predictive and preventative measurements which enable advances in patient personalization. This review aims to discuss the opportunities, challenges, and prospects posed to 5G network technology in moving forward to deliver personalized treatments and patient‐centric care via a precision medicine approach.

INTRODUCTION

In recent years, there is a dire need to change the “one‐size‐fits‐all” approach toward precision medicine because this approach may not be realistic as individual cells may harbor particular genetic variants. ¹ Precision medicine is a state‐of‐the‐art approach to tailor disease prevention and treatment by tapping into an individual's genomic data, lifestyle, and environment‐related information to provide the proper treatment to the right person at the right time. Precision medicine also includes pharmacogenomics, the key component of precision medicine, that optimizes medication selection by considering patients' genetic profiles. The genetic make‐up of individual patients may result in different responses to the same dose of a similar medication. Genetic variations affect the body's response to medications due to different drug‐metabolizing enzymes, drug transporters, drug targets, and drug receptors. ² Thus, precision medicine offers a personalized approach in enhancing the prediction of therapeutic response and outcomes, reducing costs and adverse effects of medications, while improving the efficacy of treatments sustainability and efficiency of healthcare systems.

Precision medicine has revolutionized and become more inter‐related with digitalization and big data. ³ Precision medicine relies on vast amounts of data, such as genomic and clinical data which contain information regarding sociodemographics, medical conditions, genetics, and treatments. The integration of diverse datasets from various sources, including electronic health records (EHRs) and wearable devices, has been facilitated by the advent of 5G network technology. This makes data aggregation more convenient, allowing healthcare providers to identify patterns, biomarkers, and potential treatment strategies. Thus, the digital transformation of 5G network technology usage brings us one step closer to the promise of precision medicine.

The term “5G” herein refers to the “fifth generation” of wireless transmission technology. The rollout of 5G communication networks began in various countries around the world starting in 2019. ⁴ The first wireless system, which is the first‐generation (1G), has the speed of only 2.4 Kbps (1 kilobit = 1000 bits) was introduced in 1981. Over the past 30 years of the ongoing evolutions, the latest 5G wireless network technology has achieved remarkable speeds ranging from 1–10 Gbps (1 Gbit = 1000 Mbit) in comparison to the previous fourth‐generation (4G) wireless networks which reached the speed up to 150 Mbps (1 Mbit = 1000 kbit). ⁴

The 5G network technology offers high‐speed transmission, lower latency, near real‐time connectivity, high bandwidth, and durability for massive data streams. ⁵ Compared with 4G network technology, the use of Ultra‐Wide Band networks in 5G network technology provides a band breadth of 4000 Mbps, resulting in 400 times faster speed at low energy levels. ⁴ The new generation of wireless 5G network technology provides the speed and accessibility for generating, integrating, and analyzing big data, which is helpful for precision medicine. In the event of 5G failure, a network redundancy connection can be utilized to ensure continuous operations and data analysis. This involves using alternative network options, such as Wireless Fidelity (WiFi), wired connections, or a secondary 4G or 3G network.

In addition, leveraging EHRs across various clinical settings has catalyzed the advancement of precision medicine by integrating off‐site communications via 5G networking. ⁶ , ⁷ Besides that, precision medicine utilizes different tools, such as omics technologies (genomic, transcriptomic, proteomic, metabolomic, and others), big data, artificial intelligence (AI), and machine learning (ML). ⁸ Harnessing the power of these analysis and diagnostic tools together with high‐performance computing technologies using 5G network technology poise the development of precision medicine strategies in a way that it was not before. This review entails the opportunities, challenges, and prospects of 5G network technology in the precision medicine landscape.

LEVERAGING THE EMERGENCE OF 5G NETWORK TECHNOLOGY FOR ADVANCING PRECISION MEDICINE

Disease diagnosis through next‐generation sequencing

In October 1990, the generation of the first human reference genome was begun and accomplished in April 2003 under The Human Genome Project (HGP). ⁹ This project was based on first‐generation sequencing (i.e., Sanger sequencing which adopts the chain termination method to sequence all 3 billion base pairs of the human genome). At the beginning of the 21st century, the next‐generation sequencing (NGS) era began to accelerate the sequencing process beyond the reach of traditional Sanger sequencing in terms of cost and time spent to sequence an entire genome. ¹⁰

With the completion of HGP, it has provided the fundamental information about human blueprint with the first sequence of human genome. ¹¹ The information from gene sequencing technology has become a powerful predictive tool in preventing diseases based on the reference human genome sequence and the complete sequences of key model organisms. By investigating an individual's genome, it could be used to identify the risk of disease by looking at the trends or patterns to enable healthcare professionals to intervene at earlier stages. Thus, this has made precision medicine possible for personalized treatments for individuals who share genetic markers, traits, or conditions, coupled with the 5G technology.

The 5G technology is the forefront service provider that involves data science workloads on high‐performance computing platforms in the cloud or on‐premises. ¹² It has since accelerated the growth of big data analytics and the health information sector which require a 5G network that provides high transmission speed for various techniques in precision medicine including the NGS. The NGS is a massively parallel sequencing technology that can sequence an entire human genome within a single day and offers ultra‐high throughput and speed ¹³ with the support of 5G network technology in contrast to the Sanger sequencing which involved sequencing only one DNA fragment at a time. NGS has enabled a much deeper understanding of disease biology at the genomic and transcriptomic levels.

The advancement of NGS has driven the discovery of novel disease targets, as well as therapeutic identification, prenatal testing, or even disease diagnosis. ¹⁴ Several studies reported that targeted NGS enabled the detection of disease‐causing mutations for the clinical diagnosis of Mendelian diseases, which include hereditary metabolic disease, respiratory disease, cardiovascular disease, cancer, hereditary kidney disease, and neurogenetic diseases. ¹⁵ , ¹⁶ , ¹⁷ Metabolic syndrome (e.g., high blood pressure, high triglyceride, and high glucose) and metabolic diseases (e.g., type 2 diabetes mellitus, atherosclerosis, non‐alcoholic fatty liver disease, and obesity) are alarming health burdens globally as their etiologies are not completely elucidated, making it a challenge for enabling effective treatment and prevention. ¹⁸ The utility of NGS involves the storage of vast amounts of complex and heterogeneous data that requires high speed for generating and moving data around. The breakthrough in 5G is transmission speed, which offers up to 10 Gbps data transfer rate ⁵ has dramatically supported the healthcare application. ¹⁹ The ability to utilize this massive dataset through 5G technology leads to the acceleration of precision medicine.

Untap repositories of data from multi‐omics with greater precision

With the advent of NGS, a broader set of biomolecules (DNA, epigenetic markers, RNA, proteins, and metabolites) in our cells can be measured simultaneously and comprehensively, known as genomics, epigenomics, transcriptomics, proteomics, and metabolomics, respectively. ²⁰ Another dynamic view is the integration of the aforementioned omics (called multi‐omics) that can enable us to have a better understanding of the disease and its molecular mechanisms. ²¹ It is now slowly penetrating into clinical practices and healthcare systems to find the unique properties and dynamics of each tested patient to improve their quality of life by delivering customized targeted therapies or prevention measures, making it feasible to predict, diagnose, and treat diseases more precisely and personally than before. ²² , ²³ Multi‐omics studies are promising approaches to characterize metabolic diseases and unlock them for personalized medicine given their close interaction with genetic and environmental factors.

Multi‐omics data also offer opportunities to redefine the precision in the field of oncology, by uncovering potential novel biomarkers to classify cancer types and understanding the variability in treatment response to improve cancer treatment. ²⁴ Apart from these, the integrative multi‐omics data also provide important insights into the understanding of underlying mechanisms in human brain diseases, including neurodegenerative diseases (e.g., Alzheimer's and Parkinson diseases) and mental disorders (e.g., schizophrenia and bipolar disorder). ²⁵ Dissecting each molecular level from DNA, transcript, protein, and metabolic processes as well as integrating with the available clinical information can provide a holistic view of underlying disease mechanisms and guide the therapeutic decision for the diseases mentioned above more effectively (Figure 1).

FIGURE 1.

The transition from conventional molecular biology to multi‐omics integration. Incorporation of various omics technologies and data integration strategies to provide a holistic view of underlying disease mechanisms, involving bioinformatics (Created with BioRender.com).

Cancers of the breast, lungs, colon and rectum, prostate, and skin, as well as the stomach are associated with nearly 10 million global deaths in 2020. ²⁶ Cancer is characterized by abnormal cell growth that can invade and spread to other body parts during the metastasis phase. ²⁷ Each cancer type and cancer stage have inherent variability in the cellular state, driving precision medicine as a potential approach to enable targeted therapies for personalized care. ²⁸ Integrating multi‐omics and non‐omics data (e.g., demographic information, clinical data, and treatment) provides a deeper and holistic understanding of the development and progression of complex diseases, including cancer, to enable better prediction, diagnosis, and treatment for better patient care. ²⁰ The synergy between multi‐omics and non‐omics data and the advent of 5G network technology has been deployed to improve the healthcare system. Athieniti and Spyrou ²⁹ have comprehensively reviewed the incorporation of various omics technologies through computational tools, such as Multi‐Omics Factor Analysis (integrating datasets of genomics, epigenomics, transcriptomics, and proteomics) and more.

Nevertheless, issues are still associated with implementing multi‐omics approaches, such as the diversity and massive omics data from horizontal data ensemble that consists of similar types of data collected from different laboratories to vertical data ensemble, which uses various types of data collected for a group of individuals with matching information. ³⁰ In addition, omics data requires complex and multi‐step analysis pipelines using broad computational resources and varied data sources which are obtained from genomic, transcriptomic, proteomic, and metabolomic experiments. ³¹ Thus, the execution of such many analyses can be performed using 5G network technology. It helps to facilitate quick data transfer with greater internet connectivity and bandwidth in addition to super coverage as compared to 4G LTE ⁵ and supports advances in data‐rich fields.

The 5G network rollout is looking forward to faster downloads and seamless streaming of different types of data from medical devices, wearable sensors, and others to predict health risks and prevent diseases at the population level. ³² The utilization of 5G network technology can greatly enhance the parallelization of data analysis computation by dividing large datasets into smaller subtasks that can be executed concurrently on multiple machines leading to faster data processing speed, real‐time analytics, and overall system performance. ³³

Artificial intelligence‐driven precision medicine in public health

Infectious disease prediction can be a complex process involving various data sources and analytical methods to forecast the spread of infectious diseases. This includes analyzing data on infection incidence, risk factors, and population movements, and using models to simulate the spread of the infection. ³⁴ Generally, infectious prediction aims to identify potential outbreaks early, so that public health interventions can be implemented to reduce the spread of the infection and prevent large‐scale outbreaks. ³⁵ Moreover, predicting infectious diseases via integrating patient‐specific data offers advantages in transforming infectious disease medicine into the era of precision medicine. ³⁶

One of the key data sources used in infectious disease prediction is surveillance data, which includes information on the number of cases of infection and demographic information about the affected individuals from hospital records with their informed consent. ³⁷ With the usage of 5G network technology, it is now possible to collect and transmit the data in real‐time to identify patterns and trends in infection incidence and high‐risk populations. The 5G‐enabled mobile devices can collect and transmit data from wearables and other connected health devices, providing a more comprehensive and timely view of public management, such as detecting, quarantining, and tracking infectors. ³⁸ It will use AI algorithms to integrate and analyze the vast amounts of data collected, including patient health records, genetic information, lifestyle data, and real‐time monitoring data. For example, if a higher number of cases are being reported in a specific geographic area or among a certain age group, public health officials may investigate and quickly respond to outbreaks and take preventative measures to contain the spread of infectious diseases in that area or population. ³⁹

Herein, the capabilities of 5G technologies can be effectively utilized with universal high‐rate connectivity to collect and transmit data in real‐time, identifying areas where there may be an increased risk of infection transmissions, such as airports, hospitals, or train stations. ⁴⁰ The 5G networks are equipped to gather and combine substantial quantities of data from various sources, such as sensors, devices, infrastructure, and human interactions. For example, the investigators can access the data via phone's Global Positioning System or by linking to an application with a tracking function. Hence, 5G network solutions play a pivotal role in managing a city's various functions by aggregating multiple data layers, ranging from human movement, energy consumption, and traffic patterns. ⁴ This massive amount of data collected from various sources improves surveillance, and helps to identify patterns and signs of potential infectious disease outbreaks.

Hence, infectious disease prediction is a multidisciplinary field that involves the use of various data sources and analytical methods to forecast the spread of infectious diseases. The interdisciplinary field encompasses epidemiology, medical virology and microbiology, public health, data science, geographic information systems, climatology and environmental science, social science, and health informatics. The goal is to identify potential outbreaks early; so that public health interventions can be implemented to reduce the spread of the infection and prevent large‐scale outbreaks.

AI encompasses the utilization of computers and advanced technology to replicate intelligent behavior and critical thinking similar to human beings. Of late, the power of AI in the modern era shines particularly after the recent pandemic, such as the severe acute respiratory syndrome‐coronavirus 2 (SARS‐CoV‐2) outbreak that inadvertently disrupted human mankind's life. AI has incredible potential to greatly enhance the ability to predict and respond to infectious disease outbreaks. ⁴¹ Thus, combining the 5G network with cloud healthcare platforms and AI aids in precision medicine for tailoring infectious disease prediction by offering high speed, extreme bandwidth capacity, and low latency. ⁴ , ⁴² , ⁴³

Cloud computing refers to the delivery of computing resources, including storage, processing power, software applications, and other information technology (IT) resources, over the internet. Instead of relying on local servers or personal devices, cloud computing allows users to access and utilize these resources remotely through a network of servers located in data centers. In cloud computing, the term “cloud” symbolizes the internet, representing a network of servers and services that provide on‐demand access to computing resources. These resources are typically provided by cloud service providers, who maintain and manage the infrastructure and make it available to users as a service. ⁴⁴

Owing to the wide applications of 5G network technology, AI‐powered surveillance systems can be used to monitor data from various sources, such as EHRs, social media, and news articles, to identify early signs of an outbreak. For example, the clinical outcomes gathered from the EHRs are utilized for various analyses to investigate risk factors associated with diseases and treatment responses. Authorities have implemented data protection regulations and strict privacy protocols which necessitate AI algorithms accessing EHRs. The commonly known data protection regulations are General Data Protection Regulation (GDPR) and Health Insurance Portability and Accountability Act (HIPAA). ⁴⁵ , ⁴⁶ These systems can process and analyze large amounts of data in real‐time using several AI techniques, such as natural language processing, anomaly detection, and sentiment analysis, as shown in Table 1.

TABLE 1.

The various AI prediction methods and the key points which identifies the importance of each prediction method for infection disease prediction.

AI prediction methods	Source data
Data mining	To analyze large amounts of data, such as EHRs, social media, and news articles To identify patterns and trends that could indicate an impending outbreak
Modeling and simulation	To predict how an infectious disease is likely to spread, based on factors such as population density, travel patterns, and weather conditions
Machine learning	To identify patterns and features that are associated with infectious disease outbreaks To make predictions about future outbreaks
Computer vision	To automatically analyze medical images such as X‐rays, CT scans, and microscopy images To identify patients with infections, this can help with early detection and rapid response
Artificial neural network	To process the given inputs in the form of specific algorithms which include multiple hidden layers of nodes in order to produce an output prediction

Abbreviations: AI, artificial intelligence; EHR, electronic health record.

Natural language processing (NLP) can extract relevant information from texts, classify it, and infer the presence or absence of a disease in a certain area. ⁴⁷ The NLP often associated with unstructured text data, can also be applied to structured data. Structured data refers to data that are organized and stored in a specific format, typically in databases or spreadsheets, where each data element is assigned with specific field or column. Structured data follows a predefined schema and is highly organized, making it easily searchable and analyzable. On the other hand, unstructured data refers to data that lacks a specific structure or format. It includes text documents, social media posts, emails, audio and video recordings, and other forms of data that do not fit neatly into a tabular structure. ⁴⁸

For example, an increase in the number of posts about a specific symptom or a sudden spike in searches for a particular infection could indicate the birth of an outbreak. ⁴⁷ Anomaly detection algorithms can identify patterns in the data that deviate from the norm, which could indicate an outbreak. These algorithms can be applied to structured data, such as EHRs, to identify unusual spikes in the number of patients with a specific infection or symptom. These data are generally gleaned from the clinical notes from medical practitioners. ⁴⁹ , ⁵⁰ This requires establishment of an interconnected network with a focus on the real‐time sharing of medical data which highly depends on the 5G network communication.

NLP techniques can be applied to both data types (structured and unstructured data) in EHRs (Table 2). For structured data, NLP can help extract insights, patterns, and relationships from the data, such as identifying co‐occurrences of certain medical conditions or analyzing trends in vital signs. For unstructured data, NLP enables the extraction of meaningful information from clinical notes, such as identifying adverse drug reactions, predicting patient outcomes, or summarizing patient history for decision making. The combination of structured and unstructured data in EHRs provides a comprehensive view of a patient's medical history and current health status, allowing healthcare professionals to make more informed and personalized treatment decisions. However, given the sensitive nature of these data, it is crucial to ensure privacy and adhere to ethical guidelines when using NLP techniques to analyze EHRs. ⁵¹ , ⁵² , ⁵³

TABLE 2.

EHRs: (a) structured data and (b) unstructured data.

EHRs
Structured data quantifiable and searchable data in a predefined format		Unstructured data other textual information in free‐text format
Demographic information	Patient's name, age, gender, address, contact details	Clinical notes	Physician's progress notes (examination, symptoms, risk factor, family history), consultation notes, discharge summaries
Vital signs	Blood pressure, heart rate, temperature, respiratory rate	Radiology reports	Descriptions and interpretations of imaging studies: X‐rays, ultrasound, CT scan, MRIs
Laboratory results	Blood test, molecular test, urinalysis	Pathology reports	Findings from tissue biopsies or other diagnostic tests (e.g., immunostaining)
Medication history	Prescribed medications, dosage, frequency, start/end dates	Doctor‐patient communications	Messages exchanged between healthcare providers and patients
Problem list	Patient's medical issues or diagnoses (ICD‐10 codes assigned to each condition (e.g., hypertension – I10, diabetes – E11)	Nursing notes	Information recorded by nurses about patient care and observations

Abbreviations: CT, computed tomography; EHRs, electronic health records; ICD‐10, International Classification of Disease‐10th revision; MRI, magnetic resonance imaging.

Sentiment analysis, also referred to as opinion mining, is a computational method used to analyze text and determine the sentiment or subjective viewpoint conveyed within it. It involves categorizing the sentiment as positive, negative, or neutral by examining the words, phrases, and contextual cues utilized in the text. Sentiment analysis finds broad application in areas such as social media monitoring, brand reputation management, market research, and customer feedback analysis. By automatically evaluating sentiment in large volumes of text data, businesses and organizations can acquire invaluable insights to make informed decisions and comprehend public sentiment. ⁵⁴ , ⁵⁵

AI‐powered predictive models can be used to analyze historical data and predict future outbreaks. ⁵⁶ These models consider a wide range of factors, such as demographics, travel patterns, weather conditions, and the genetic makeup of the pathogen, to make accurate predictions. ⁵⁷ They can be used to model the spread of infection, identify hotspots of transmission, and predict the potential number of infected individuals in a certain geographic area. ⁵⁸ Predictive models can be created using several AI techniques, such as ML, statistical modeling, and artificial neural networks (ANNs). ⁵⁹

ANN represents an algorithm encompassing multiple hidden layers of nodes that process the given inputs to produce an output prediction. ⁶⁰ The other neural network including graph neural network (GNN) is a type of neural network designed to process and analyze structured data that can be represented as graphs. Graphs consist of nodes (also called vertices) connected by edges, and they can represent various real‐world relationships and interactions. Another neural network is the convolutional neural network (CNN), a type of deep learning (DL) model that is widely used for analyzing visual data, such as images and videos. CNNs are particularly effective in computer vision tasks like image classification, object detection, and image segmentation. ⁶¹ , ⁶² DL is a type of ML that mimics the workings of the human brain to understand and then analyze the data. ⁶³ Herein, 5G network technology plays a vital role in healthcare informatics involving various analytical methods and models and enables the connection of high‐performance computing devices over large distances. ⁶⁴

In recent years, a new healthcare model that could be delivered by 5G network technology enables computers to handle vast amounts of data. ML algorithms can be used to analyze large amounts of data to identify patterns and trends that are associated with infectious disease outbreaks. These algorithms can be trained on historical data from previous outbreaks, such as symptom reports, laboratory test results, and other data. ⁶⁵ , ⁶⁶ By harnessing the capabilities of precision medicine and leveraging big data, healthcare providers can access valuable environment‐related information, enabling them to develop highly effective personalized treatment approaches in the near future. ⁶⁷

Computer vision technology can be used to automatically analyze medical images, such as X‐rays, computed tomography (CT) scans, and microscopy images, ⁶⁸ to identify infected patients and to help with early detection which is crucial to provide rapid response. This can be performed using the high‐rate connectivity of 5G network technology which allows real‐time transmission of large medical data, such as medical images and videos. In addition, computer vision can be used to scan images and identify patterns that are associated with a specific infection, and then flag them for further analysis by public health officials.

Nevertheless, all of these AI‐driven systems are most effective when integrated with traditional surveillance and prediction methods, such as expert analysis, laboratory testing, and epidemiological investigation. ³⁷ However, there are also some challenges to be addressed to fully realize the potential of AI in infectious disease prediction (Table 3) and the related ethical issues in AI are shown in Table 4. This is because, even with the advancement of AI, these systems still require human judgment and expertise to interpret the results, for making sense of the data, and eventually appropriate action can be taken.

TABLE 3.

The challenges in AI models and its key points.

Challenges	Key points
Data quality and availability	Reliable, high‐quality data is essential for training and evaluating AI models Access to such data can be limited in many areas
Model interpretability	Some AI models, particularly deep learning models, can be difficult to interpret, which can make it hard to understand how they are arriving at their predictions
Robustness	AI models can be affected by the availability and distribution of the data used to train them Can be affected by changes in the environment or population. Examples: image classification, speech recognition and natural language processing

Abbreviation: AI, artificial intelligence.

TABLE 4.

The ethical issues in AI and its key points.

Ethical issues in AI	Key points
Privacy	The right to determine for oneself on when, how and for what purpose on the usage of personal data related to AI
Biasness	A phenomenon that arises when an AI algorithm delivers systematically skewed results as a consequence of erroneous assumptions (i.e., during machine learning process)
Explainability	The ability to interpret input, output, and behavior of AI model and how it leads to prediction outcome
Transparency	The need to openly inform and communicate with stakeholders on the system development, implementation and application of AI
Justifiability	Assessment of whether the decisions of an AI system are reasonable or valid based on the norms and rules of society
Reliability, safety, and consistency	Trust in AI systems will depend on whether they can be operated reliably, safely, and consistently even under unexpected conditions, especially for applications in fields affecting both lives and livelihoods, such as transportation, health care, and financial services – where consequential decisions are involved
Accountability	Transparency is crucial because a lack of it tends to lead to suspicion and reluctance

Abbreviation: AI, artificial intelligence.

Delivering the precision oncology with 5G network technology

Sathyanarayanan and colleagues conducted a systematic evaluation of ready‐to‐use bioinformatic tools to integrate multi‐omics data to identify cancer driver genes (multistaged model – accessing gene, function, and pathway levels) and to classify tumor subtype (meta‐dimensional model – classification performance). ⁶⁹ Although these multi‐omics tools improved gene identification and sample classification more than single omics data, sufficient sample size and the usage of control datasets can improve the performance further. However, the outcome remains room for improvement to facilitate the implementation in clinical practice with a user‐friendly interface, including actionable information for better patient care.

Precision medicine requires more sophisticated efforts to model patient‐specific data (multi‐omics and non‐omics data) with publicly available annotation databases (genes, variants, diseases, drugs, and biomarkers). ⁷⁰ Digital innovation, such as the 5G network technology, has created various opportunities for a myriad of new technologies in the field of precision medicine. The current cutting‐edge technologies including AI and ML algorithms can leverage classical statistical methods to model patient‐specific data across diverse ancestries, diseases, and populations to understand disease mechanisms moving a step forward to personalized medicine in terms of diagnosis, treatment, and prognosis of certain diseases, including cancer. ⁶⁴

The emerging hallmarks of cancer include cell cycle, apoptosis, proliferation, differentiation, growth, and others. ⁷¹ The key processes regulating these cellular functions are the phosphorylation of tyrosine, serine, and threonine residues, which is the primary function of kinase proteins. ⁷² CancerOmicsNet, an AI‐based GNN, is a useful predictive tool combining biological networks, genomics, inhibitor profiling, and gene‐disease associations to predict drug response, specifically in drug‐targeting kinase inhibitors across various cancer cell lines, aiming to predict the protein‐drug binding affinity and the gene‐disease association (identification of all genes that are involved in a disease). ⁷³

Using the AI‐based GNN together with the expansion of 5G network technology has paved the way for precision oncology. Studies reported that both pan‐CDK inhibitor JNJ‐7706621 and Src inhibitor PP1 were predicted by CancerOmicsNet, as being the most potent anti‐proliferative molecules against the PanC 04.03 pancreatic cancer cell line in a dose‐dependent manner. ⁷⁴ Meanwhile, AZD6482 and XMD8‐92 molecules exhibited anticancer properties in the human prostate cancer cell line DU 145, whereas GW2580 and PI‐103 molecules are found to be effective against HCC70, human triple‐negative mammary carcinoma cell line. ⁷⁴ Identifying candidate molecules through this approach is not only promising, but also circumvents the need for massive investment in clinical trials. Consequently, pharmaceutical companies can significantly reduce their investment costs in drug development.

Lung cancer is a common malignant tumor with a 10%–20% survival rate due to late diagnosis and limited treatment methods. ⁷⁵ Although the accurate diagnosis of early‐stage lung cancer and lung cancer staging is complex and highly dependent on the radiologist's experiences in interpreting CT and positron emission computed tomography (PET) images, the AI system based on 3D DL technology can be applied to improve diagnosis efficiency, involving multiparameter cluster analysis. ⁷⁶ , ⁷⁷ , ⁷⁸ The 3D DL technology is a subset of ML applications in AI by mimicking how the human brain works. Meanwhile, early diagnosis with the 3D DenseSharp and 3D DenseNets networks and cancer staging using the CNN model are core diagnostic components of the AI diagnostic system that can train and predict lung cancer diagnosis and staging more effectively ⁷⁹ , ⁸⁰ , ⁸¹ along with the extremely low latency and high data transmission speed of the 5G network technology.

Apart from that, the AI model can be applied to choose the best treatment for lung cancer, including surgical resection, radiotherapy, and drug treatment. ⁸² Various AI and ML methods are used to explore better surgery and therapeutic drugs for patients, by integrating patients' medical records, clinical data, and public databases to offer personalized treatment. ⁸³ , ⁸⁴ , ⁸⁵ In terms of the prognosis of lung cancer, the prognostic judgment can improve the survival benefit by advancing the treatment model via the incorporation of AI, such as advancing the iTEN (impact of treatment evolution in non‐small cell lung cancer) model by incorporating ANNs. ⁸⁶

Breast cancer is the leading cause of death among women across the globe. ⁷⁵ The etiology of breast cancer is complex, involving factors like gender, heredity, environment, and occupation. Examination of biopsy slides coupled with AI technologies including ML and DL approaches are important to train prediction models and interpret generalizable models. The generalizable model refers to the ability of a model to interpret unseen new data and thereby, accurately interpret and make predictions on new biopsy slides from patients that it has never encountered.

Additionally, establishing the ANNs using these approaches improves the accuracy and efficiency of breast cancer screening and classification. ⁸⁷ , ⁸⁸ Raafat et al. ⁸⁹ showed that AI‐based mammography using Fujifilm digital mammography system provides the abnormality scores and generally performed better than conventional mammography in detecting certain breast cancers, such as ductal carcinoma in situ, invasive lobular carcinoma, tubular carcinoma, and micropapillary carcinoma. Apart from diagnosis, AI‐based imaging provides breast cancer risk assessment and facilitates the risk stratification implementation coupled with demographic and genetic data to personalize breast cancer screening regimens. ⁹⁰ Advances in AI technologies coupled with 5G network have made the breast cancer risk assessment and prediction of cancer feasible.

Moreover, AI can be applied to identify novel cancer target identification and drug discovery using cancer‐related multi‐omics technologies. ⁹¹ Network‐based (identifying targets in a protein–protein interaction network) and ML‐based (identifying targets in complex biological networks) biology analysis algorithms are two commonly used AI algorithms to integrate multi‐omics data. ⁹¹ Network‐based biology analysis is important to integrate complicated biological data and identify the entity's relationships; nodes (e.g., genes, proteins, diseases, and drugs) and edges (e.g., biochemical physical or functional interactions between nodes) in the biological networks (e.g., gene–gene networks, protein–protein networks, and drug‐target networks). On the other hand, ML‐based biology analysis selects significant topological features for cancer based on the decision tree algorithm whereas identifying cancer targets and discovering drugs are based on DL algorithms. In brief, personalized medicine and even digitalized healthcare will be feasible with the integration of AI, ML, and DL, as well as 5G, which may be of great therapeutic value in cancer.

Disease modeling for precision medicine

One of the main limitations in drug development is the lack of a laboratory model to mimic the complex cell biology in our bodies. Hence, the advent of technologies of growing “mini‐organs,” the so‐called organoids are beneficial for drug screening to predict patient response to drug treatment. Various studies [92, 93, 94, 95, 96, 97, 98, 99, 100] are conducted using different approaches to bridge gaps between disease modeling organoids and precision medicine, as shown in Table 5.

TABLE 5.

Patient‐derived organoid‐based model for the prediction of drug sensitivity towards precision medicine.

Diseases	Optimizing methods	Tested drugs	References
Advanced colorectal cancer	High‐throughput sequencing platform with microfluidic organoid drug testing system	Afatinib, Dacomitinib, Dasatinib, Gefitinib, Imatinib, Lapatinib, Nilotinib, Olaparib, Osimertinib, Pazopanib, Regorafenib, Vemurafenib	[92]
Advanced clear cell renal cell carcinoma	Air‐liquid interface system	Toripalimab	[93]
Prostate cancer	Short‐term drug testing in ex vivo tissue culture	Combination of gemcitabine and Chk1 inhibitor MU380	[94]
Adeno‐ and squamous cell carcinoma NSCLC and pancreatic ductal adenocarcinoma	OrBITS	Afatinib, Carboplatin, Cisplatin, Erlotinib, Gefitinib, Osimertinib	[95]
GBM	Immersion bioprinting of hyaluronan and collagen bioink‐supported	p53 activator compound and TMZ	[96]
Salivary gland cancer	Ex vivo tissue culture of drug sensitivity	Cisplatin, Erlotinib, Lapatinib, Sunitinib, Crenigacestat, Monensin	[97]
Pancreatic ductal adenocarcinoma	3D immunofluorescence imaging	FOLFIRONOX, Gemcitabine, Pembrolizumab, Paclitaxel, Gemcitabine/Abraxane (paclitaxel)	[98]
Primary and metastatic brain tumors	Ex vivo tissue culture of drug response	Dabrafenib, Temozolomide, Trametinib, Vemurafenib	[99]
GB	FLIM‐based metabolic imaging	Temozolomide	[100]

Abbreviations: FLIM, fluorescence lifetime imaging microscopy; GB, glioblastoma; GBM, glioblastoma; OrBITS, Organoid Brightfield Identification‐based Therapy Screening; TMZ, temozolomide.

Patient‐derived organoid model systems have been used to bridge gaps between existing pancreatic ductal adenocarcinoma model systems and personalized medicine. Looking to the next wave of precision medicine, an automated system from organoid fabrication, manipulation, and drug evaluation functions is greatly required. ¹⁰¹ A study from Renner and colleagues demonstrated a fully automated high‐throughput workflow for human midbrain organoids capable of producing organoids with highly homogeneous size, structure, global gene expression, and cellular composition. ¹⁰² Similarly, other findings reported 1 week of high‐throughput screening for patients with cancer using an automated organoid platform with organoids were found to recapitulate 97% gene mutations in tumors and drug response prediction. ¹⁰³ Fully automated reproducible organoid models for high‐throughput drug screening assays therefore requires the 5G‐assisted smart healthcare networks.

Looking ahead to 5G network technology deployment in the precision medicine landscape, it can unleash opportunities across a number of different applications as well as to meet the demands of increasing data‐intensive applications, as shown in Figure 2.

FIGURE 2.

Opportunities of 5G network technology in the precision medicine landscape. Implementation of 5G network technology revolutionizes smart healthcare system, enabling high speed and real‐time data transmission as well as enhancement of connectivity for various precision medicine applications (Created with BioRender.com).

CHALLENGES AND PROSPECTIVES

There are several challenges that 5G network technology faces in its implementation of precision medicine (Figure 3). One major challenge is ensuring the security and privacy of patient data, as large amounts of sensitive medical information will be transmitted over the network. There is also the challenge of integrating 5G technology with existing medical equipment and systems, which may require updates or upgrades. Additionally, there is a need for the development of specialized applications and software that can take advantage of the high speeds and low latencies of 5G networks to support precision medicine.

FIGURE 3.

Main challenges of 5G network technology faces in its implementation towards precision medicine. The diagram highlights the key obstacles in 5G network technology, including cybersecurity, costing, and acceptance of patients, which need to be addressed for successful precision medicine applications (Created with BioRender.com).

In terms of perspective, 5G network technology has the potential to revolutionize precision medicine by enabling faster and more accurate data collection and analysis. With its high speeds and low latencies, 5G networks can support real‐time data and allow remote patient monitoring. Remote monitoring enables the collection and analysis of health‐related data from individuals who are located at a distance from healthcare providers and thus, offer more timely and effective treatments.

Additionally, 5G networks can support using more advanced medical technologies, such as virtual reality (VR) that creates an immersive, computer‐generated environment that simulates a realistic experience. For example, 5G‐powered VR medical training allows medical professionals to engage in realistic, hands‐on training. It enables them to practice medical simulations, and emergency scenarios in a safe and controlled environment. ¹⁰⁴ On the other hand, surgeons utilize augmented reality (AR) glasses, which combines elements of the physical world with digital content by providing necessary information of patients and their conditions to aid in their surgeries and procedures. ⁶³ With 5G network, data can be transmitted with very low latency, which is critical for the user wearing the AR smart glasses.

The 5G network will also bring several new use cases for healthcare, like telemedicine, remote monitoring, and real‐time analytics of data generated by medical devices and wearables. ¹⁰⁵ , ¹⁰⁶ These will give the ability to medical professionals to access patients in remote locations, or perform remote surgeries or consults. Consequently, 5G networks could provide an efficient, fast, and high‐quality network that enables the breakthrough of precision medicine by connecting the different domains in the healthcare ecosystem, like remote monitoring and telemedicine, large‐scale genomic data analysis, and AI‐powered decision support systems.

Privacy protection issues and cybersecurity

The safety and privacy of the enormous amount of data collected is crucial in this 5G‐driven precision medicine application, which raises ethical, social, and legal issues. The data includes genomics information, medical history, behavior, and social data, which is used to gain precision information, covering peoples' daily lives. ¹⁹ Additionally, data authenticity and integrity are important to capture patients' experiences and clinical outcomes to accelerate the growth of precision medicine in the current healthcare system. The patient experience encompasses the interactions that individuals have with the healthcare system, such as good communication with healthcare providers, getting timely appointments and easy access to information. ⁵ , ¹⁰⁷ Thus, effective regulation, policies, and business models are essential for network security, cybersecurity, and device protection, as well as data management and sharing to build a safe and well‐controlled system to protect human rights, privacy, and confidentiality data. ¹⁹ , ⁴⁰ Besides, effective data management and sharing to build a safe and well‐controlled system which also comprises distinguishing data ownership from data access and control is also crucial in recognizing individuals' rights and interests related to their health and genomic data. ¹⁰⁸

Meanwhile, leveraging 5G technology enables the sharing of valuable data with global partners to work together to customize disease‐prevention strategies requiring stringent regulations and systems with global protocols and infrastructures. ⁶⁷ Nevertheless, the shared data are limited to authorized individuals or organizations that involve data‐sharing agreements covering various aspects, including purpose, permitted use and scope, security, and restrictions on data usage. Besides that, access and regulation of data can also differ depending on whether the data are used for research purposes or clinical care. Healthcare regulations and patient consent requirements typically govern access to clinical data but access to research data may require approval from research ethics committees or institutional review boards. ¹⁰⁹ Authorities have implemented data protection regulations and strict privacy protocols which necessitate AI algorithms accessing EHRs. The commonly known data protection regulations are GDPR and HIPAA. ⁴⁵ , ¹⁰⁹

Recognizing the importance of data sharing for advancement in precision medicine from different entities has led to the beginning of a “Medical Information Commons” (MIC), an ecosystem to guide decision making about data control and access. ¹⁰⁸ MIC is defined as “networked environments to shared resources in diverse health, medical, and genomic data on large populations.” ¹¹⁰ Herein, the Global Alliance for Genomics and Health (GA4GH), an international, nonprofit alliance was formed in 2013 to create a common framework of standards and harmonized approaches to accelerate research and improve human health via responsible and secure sharing of genomic and health‐related data. ¹¹¹

Cost‐effective network

One of the major challenges of implementing 5G technology toward precision medicine is the cost of deploying the necessary infrastructure. ¹¹² This includes the cost of building and maintaining 5G cell towers and base stations, as well as upgrading existing equipment to be compatible with 5G networks. In addition, the challenges about requirements and specifications arise due to the complexity of 5G network system from an End‐to‐End perspective (device, radio network, core network, and data network) and a multilayer perspective. ¹¹³ , ¹¹⁴ These costs can be substantial, especially in rural or under‐served areas where the infrastructure may not be as developed.

In addition to the infrastructure costs, there are costs associated with upgrading existing medical equipment and systems to be compatible with 5G networks. ¹¹² This includes upgrading diagnostic devices, EHR systems, and other medical equipment to be able to transmit and receive data over 5G networks. The cost of these upgrades can be significant and a potential barrier for small and underfunded hospitals and clinics. ³⁷

However, despite these challenges, there are also significant cost savings that can be realized by implementing 5G technology toward precision medicine. One of the key benefits of 5G networks is the ability to support telemedicine and remote monitoring. ¹¹⁵ This allows patients to receive medical care remotely, reducing the need for travel and lodging, which can be significant cost savings. Additionally, 5G networks can support the use of more advanced medical technologies, such as VR which creates an immersive, computer‐generated environment that simulates a realistic experience and AR, which can improve the diagnostic and treatment process for patients, and make it less expensive. For example, 5G‐powered VR medical training allows medical professionals to engage in realistic, hands‐on training. It enables them to practice medical simulations, and emergency scenarios in a safe and controlled environment.

Furthermore, 5G networks can support real‐time analytics which can speed up the process of diagnosis and treatment, which in turn can reduce the overall cost of health care. The 5G networks can also enable the connectivity of a wide range of devices and systems which can significantly improve the way healthcare services are delivered and increase the efficiency of healthcare systems. ¹¹⁶ This can lead to cost savings in areas such as supply chain management, inventory, and logistics by providing real‐time data and improved communication between systems.

Although the upfront costs of deploying 5G infrastructure and upgrading equipment can be substantial, the long‐term benefits of 5G technology in terms of cost savings can be significant. These benefits include reduced need for in‐person visits, improved efficiency, and cost savings from advanced medical technologies and supply chain management. Healthcare providers and policymakers need to consider the long‐term cost savings when evaluating the implementation of 5G technology toward precision medicine.

Societal issue and acceptance of smart health care

The convergence of 5G technology and precision medicine promises the revolution of smart health care. However, these technologies are still hindered by challenges associated with patients' poor cognition and acceptance of smart healthcare due to their traditional concepts, living styles, and education level. ⁴² In addition, lack of acknowledgment and management of cultural sensitivities and socioeconomic hurdles may result in non‐acceptance of precision medicine. ³⁵

Arguably, the healthcare providers themselves are the main barrier in encouraging the widespread uptake of precision genomics by the community. There are concerns from physicians about exposing their patients to potential risks associated with utilizing pharmacogenomics as a guiding tool for drug therapy which may impede the adoption of this new therapeutic strategy. ¹¹⁷ For example, precision medicine oncology technologies were not widely incorporated in the genomic testing in their daily workflows in United States. ¹¹⁸ A survey of 1281 US oncologists revealed that only 38.2% were confident in using NGS only as a means to guide patient care. ¹¹⁹

Meanwhile, a study indicated varying levels of awareness of precision medicine among the health professionals in Korea, including caution in educating patients and health professionals related to precision medicine. ¹²⁰ On the other hand, clinicians, researchers, and clinician scientists in Germany also raised their uncertainties of a series of ethical, process‐related, and economic challenges in using precision medicine for chronic inflammation treatment. ¹²¹ Various stakeholders, including patients, integrated care team members, and health system leadership within the Southcentral Foundation (SCF) in South Central Alaska, US communities have also raised concerns regarding overpromising in precision medicine. In addition, difficulties in conducting precision medicine trials due to the time‐consuming nature of matching patients with appropriate medications and limited drug availability are some of the current challenges in the path of precision medicine. ¹²²

According to a recent survey involving 150 oncologists, representing 10% of UK oncologists, it was reported that 92.7% of them expressed the need for additional genomics training to equip themselves for advances in clinical practice for genomics in oncology. ¹²³ Therefore, innovation in educating healthcare providers and patients to acquire knowledge and skills in leveraging this rapid advance in precision medicine technologies is vital. This was supported by the findings from Rahman et al.¹²⁴ who suggested that tailored educational and training interventions are necessary to enhance the confidence level of oncologists and oncology nurses in adopting genomic testing into clinical practice in the era of precision medicine. For instance, the Genomics Education Programme (GEP), a part of the UK's National Health Service, can be used to prepare the healthcare workforce to harness the power of genomic medicine. ¹²⁴ Continuing engagement with communities of diverse populations in the education healthcare system is needed to introduce precision medicine effectively to augment resilience to future health crises.

CONCLUSION

In essence, 5G network technology has the potential to revolutionize precision medicine by enabling faster and more reliable communication of medical data, as well as the ability to connect medical devices and sensors remotely. This can lead to more efficient and accurate diagnoses, and improved patient outcomes. With regard to the beneficial aspects of 5G network technology, there are still enormous challenges in realizing precision medicine as the subsequent alternative treatment at this moment in time. Nevertheless, various cutting‐edge technologies, such as 5G wireless transmission technology and AI, are being developed to meet these challenges, hoping to bring subtle transformations and reshape smart healthcare via precision medicine approaches.

Next, ensuring data security and privacy, dealing with the increased complexity of the network infrastructure, and studying the potential health effects of 5G are all important issues that need to be overcome for precision medicine to harness the benefits of this technology fully. At the same time, a growing concern revolving around the potential adverse effects of 5G connections on human health and the environment. As of now, there are limited studies of the possible adverse effects of 5G network technology on human health in the long run, highlighting the need for further investigations to gain a deeper understanding of any causal links associated with exposure to 5G environment.

Digital transformation through the high bandwidth of 5G network technology in changing the precision medicine field to bring the greatest benefits across all healthcare systems is facing inevitable challenges. Hence, the synergy effects between these 5G network technologies and precision medicine require the support and engagement from all the stakeholders in the healthcare system with the ultimate goal of providing more accurate diagnoses, targeted prevention strategies, suitable treatments, and accelerating the discovery of novel therapies.

FUNDING INFORMATION

The authors gratefully acknowledge support by Malaysian Ministry of Higher Education under Fundamental Research Grant Scheme: FRGS/1/2022/TK08/XMU/02/10 and Xiamen University Malaysia under Research Fund Grant No: XMUMRF/2021‐C7/IECE/0018 and XMUMRF/2023‐C12/IECE/004.

CONFLICT OF INTEREST STATEMENT

The authors declared no competing interests for this work.

Kang CC, Lee TY, Lim WF, Yeo WWY. Opportunities and challenges of 5G network technology toward precision medicine. Clin Transl Sci. 2023;16:2078‐2094. doi: 10.1111/cts.13640