Impact of Artificial Intelligence on the Future of Clinical Pharmacy and Hospital Settings

Abstract

This review aimed to comprehensively evaluate the impact of artificial intelligence (AI) on the future of clinical pharmacy practice and hospital settings. To achieve this, a thorough literature search was conducted across major databases, including PubMed, Scopus, Web of Science, Embase, and Google Scholar, covering studies published up to July 2025. Relevant keywords – such as “artificial intelligence,” “machine learning,” “clinical pharmacy,” “hospital pharmacy,” “medication management,” and “decision support systems” – were used to identify studies assessing the applications, challenges, and outcomes of AI in these contexts. All eligible articles were narratively analyzed. The findings indicated that AI has a significant role in optimizing pharmaceutical inventory management, reducing medication errors, enhancing the accuracy and safety of drug dispensing, and streamlining drug distribution processes in hospital environments. Advancements in machine learning algorithms and clinical decision support systems not only assist pharmacists in selecting optimal therapeutic regimens and detecting drug–drug interactions but also facilitate the delivery of personalized patient care and prediction of medication needs. In addition, AI integration was shown to improve workflow efficiency, reduce administrative workload, and enhance both pharmacist training and job satisfaction. In conclusion, the integration of AI into clinical pharmacy and hospital settings emerges as a transformative factor, promising to elevate the quality of pharmaceutical care, improve patient outcomes, and shape a more dynamic and efficient future for the pharmacy profession.

INTRODUCTION

Artificial Intelligence (AI) is a vast field of science associated with intelligent machine learning (ML), mostly intelligent computerized programs, providing outcomes in a comparable manner to the human attention practice.[1] This process usually comprises collecting data, developing well-organized systems for the application of obtained data, and illustrating definite or fairly accurate results, self-corrections, and adjustments.[2,3]

In the recent years, AI technology has emerged as a highly crucial element in various research fields, finding valuable applications across industries. The field of pharmacy, in particular, has made significant strides in utilizing this enabling technology to enhance workflow efficiency, reduce operational expenses, and elevate safety, accuracy, and overall efficiency in all pharmacy settings. By implementing automated dispensing systems, pharmacists are able to allocate more time toward interacting with a larger number of patients, thereby improving their health outcomes.[4]

It is believed that the initial utilization of computers in pharmacies can be traced back to the 1980s. Since then, computers have found their way into various aspects of pharmacy operations, including data collection, retail pharmacy management, clinical research, drug storage, pharmacy education, clinical pharmacy, and many other areas. With the advent of AI, the potential for growth and advancement in the pharmacy sector is immeasurable. In the field of medicine, numerous expert systems have been developed to aid physicians in medical diagnosis.[5] In recent times, numerous initiatives have emerged that center around drug therapy. These initiatives aim to provide guidance on drug interactions, monitoring drug therapy, and selecting drug formularies. The field of pharmacy encompasses various facets that can be influenced by AI, and it is crucial for pharmacists to acknowledge these prospects as they may eventually materialize in the realm of pharmacy practice.[4]

The application of AI as an assist in the assortment and administration of personalized pharmacotherapy regimens is crucial, particularly when treating complicated and severe diseases in which most of patients require entirely new treatment protocols. Only through the application of computational devices reliant on AI can we accurately control the immense amount of biological, chemical, and medical data that need to be integrated to completely realize and fight complicated diseases. Similar to other mathematical techniques, the functions identified by AI modeling methods are most precise when applied to smaller output ranges rather than extrapolated across larger areas.[6]

The aim of AI in drug treatment is to lessen the data to an extent that makes the physician to interpret, giving them the access to information they have not formerly considered. AI-based black box decision support solutions contribute greater accuracy but with less transparency. The use of so-called black-box models has caused a lot of anxiety because their decision-making process is not always rational for people, but this stress should be alleviated if they prove successful and increase patient survival. Proving the predictive power of any medical black-box model requires extensive testing due to the impact of such software and the commonly used lack of interpretability. Models must be shown to unequivocally improve patient outcomes before they can be used in clinics. The introduction and wider use of deep learning (DL) in the selection and optimization of drug therapies make it even more difficult to achieve transparency in models. We must learn to trust the judgments and advice of our machines, just as children must learn to trust the judgments of their parents.[7]

In the past few years, the efficiency of AI technology in providing health support services and medication assistance has been widely acknowledged. Molly, a virtual nurse developed by a start-up, is equipped with a pleasant voice and a friendly face. Its primary objective is to assist patients in managing their treatment plans and provide support for their chronic conditions during visits to the doctor. AI Cure is a smartphone app that utilizes the device’s webcam to monitor patients and help them regulate their medical conditions. This application proves particularly beneficial for patients with complex medication requirements and those involved in clinical trials.[8]

Deep Genomics, an AI system, proves to be valuable in analyzing patterns within genetic information and medical records, enabling the identification of mutations and their connections to various diseases. By examining the alterations in DNA caused by genetic variations, this system provides doctors with insights into the cellular events taking place. The algorithm behind this AI system was developed by Craig Venter, renowned as the father of the human genome project. It offers information on patients’ physical traits by analyzing their DNA. In addition, the AI technology known as “Human Longevity” plays a crucial role in detecting the precise location of cancer and vascular diseases during their early stages.[9]

METHODS

In this review, we systematically explored the impact of AI on clinical pharmacy practice and hospital environments. Literature searches were conducted across multiple reputable databases, including PubMed, Scopus, Web of Science, Embase, and Google Scholar. The search strategy utilized a combination of relevant keywords and MeSH terms, such as “artificial intelligence,” “machine learning,” “clinical pharmacy,” “hospital pharmacy,” “medication management,” “decision support systems,” and “hospital settings.” Boolean operators (AND, OR) were applied to ensure comprehensive identification of pertinent studies. There were no time restrictions applied to the search in order to capture the full spectrum of available literature up to July 2025. Studies were included if they discussed the applications, challenges, and outcomes related to AI in clinical pharmacy or hospital settings. Duplicate records were removed, and titles, abstracts, and full texts were screened for eligibility. Data were then extracted and synthesized narratively.

RESULTS AND DISCUSSION

Artificial Intelligence in Hospital & Community Pharmacies

ML models enable the personalization of emails in hospital and community pharmacies at a level of speed and precision surpassing that of any human being. Chatbots have the potential to enhance the effectiveness of service provision by simulating interactions typically carried out by customers and sales or customer care representatives. They are equipped to autonomously address customer concerns and inquiries, redirecting complex issues to human personnel when necessary.[10] This concept can be implemented in the context of retail pharmacy, where chatbots can be tailored to replicate the interactions between pharmacists and patients. AI has proven to be a valuable tool in the field of inventory management. For instance, envision yourself as a retail pharmacist who can accurately anticipate the future needs of your patients, ensuring that you have the necessary medications in stock. By utilizing personalized software and AI-powered data analytics, you can even send timely reminders to patients regarding their drug requirements via email. This predictive capability of AI enables pharmacists to make informed decisions when procuring stock, ensuring that they have the right medications available when needed. It is worth noting that while there are existing inventory management software and applications used in retail pharmacy stock management, such as Mckessons, Liberty, Winpharm, PrimeRx, and WinRx, not all of them incorporate AI or ML. As an example, Blue Yonder, an AI company, has developed software specifically for the Otto group,[11] a German online and catalog retailer. The software has the capability to forecast Otto’s sales with a 90% accuracy rate over a 30-day period. This advancement has significantly reduced the delivery time for purchased items from a week or more to just 1–2 days by facilitating direct shipment from the supplier to the consumer, bypassing the need for warehouse storage. In an effort to enhance patient safety, the University of California San Francisco (UCSF) Medical Center has implemented robotic technology for the preparation and monitoring of medications. The institution reports that the technology has successfully dispensed 350,000 medication doses without any errors. The robotic system has demonstrated superior performance compared to human counterparts in terms of precision and efficiency, particularly in the preparation of oral and injectable medications, including hazardous chemotherapy drugs. This innovation has empowered UCSF pharmacists and nurses to focus on direct patient care and collaborate more closely with physicians. Within the pharmacy’s automated framework, computers receive medication orders electronically from UCSF’s medical professionals. Subsequently, the robotics pick, package, and dispense individual pill doses, which are then assembled onto a bar-coded plastic ring by the machines. This plastic ring contains all the medications that a patient needs to take within a 12-h timeframe. Furthermore, the automated system is equipped to prepare sterile chemotherapy preparations and accurately fill intravascular syringes with the appropriate medications.[12]

The main objective of AI applications in the field of healthcare is to examine the connections between prevention or treatment methods and the outcomes experienced by patients. AI programs have been created and utilized in various areas, including the diagnosis procedures and the development of treatment protocols,[13] drug development, personalized medicine,[14] and patient monitoring and care,[15] among others. Pharmacies have the opportunity to enhance value-based outcomes by embracing the ongoing technological advancements in healthcare. By transitioning from traditional medication fulfillment centers to comprehensive health management hubs, pharmacies can offer more personalized care to patients. Through the utilization of technology, such as health trackers and wearables, pharmacies can gather real-time data to closely monitor patients’ conditions and track their progress. In addition, technology enables pharmacies to expand their services beyond dispensing medications to include health advice, guidance, immunizations, screenings, medication therapy management, and disease state management. This shift towards leveraging technology in healthcare can ultimately improve the quality of care provided to patients.[16]

AI has proven to be a valuable tool in the analysis of data and the presentation of results, providing support for decision-making processes. This not only saves human effort, time, and money but also contributes to saving lives. The healthcare sector has greatly benefited from the advancements in AI, which can be attributed to various factors. First, the overall evolution of computers has led to faster data collection and more powerful data processing capabilities. In addition, there has been a significant increase in the availability of health-related data from personal and healthcare devices, as well as medical records. Furthermore, the development of pharmacogenomics and gene databases has played a crucial role in the advancement of AI in healthcare. Moreover, the expansion and industry adoption of electronic health records (EHRs), along with advancements in natural language processing and other computing technologies, have enabled machines to replicate certain human processes. These advancements have paved the way for AI to revolutionize the healthcare industry and improve patient outcomes.[17]

In the realm of medicine, technology giants such as Microsoft are making significant strides in the healthcare sector through the implementation of AI. Specifically, AI is proving to be invaluable in aiding physicians in the complex task of selecting the most suitable treatment options for cancer patients. By gathering and analyzing data from diverse databases pertaining to the specific condition, AI is empowering doctors to accurately identify and prescribe the optimal medications for individual patients.[18,19] Within the pharmacy settings, AI collaborates with researchers to enhance decision-making procedures related to current medications and broadening treatment options for various ailments. In addition, AI aids in accelerating the clinical trial phase by identifying suitable participants through the analysis of diverse data outlets.[20,21] Through the utilization of AI learning, it is possible to accurately forecast the timing and location of potential epidemic outbreaks by analyzing data from past outbreaks and various media outlets.

Within the hospital setting, AI is currently employed to mitigate medical errors and lower hospital readmission rates. Through the examination of patient information related to medical and medication errors, the underlying reasons for readmissions, as well as data from various internal and external sources, AI has the potential to detect and avert potential complications in high-risk patients, offer proactive care recommendations, and provide diagnostic assistance, alongside numerous other clinical uses.

Currently, in the field of pharmacy, there exists an initial version of AI known as the pharmacy management system. This system contains patient usage information, drug data, and has the capability to detect potential drug-related issues using clinical decision support screening. The future of pharmacy technology involves implementing a technology-driven information expert system that can promptly identify drug-related problems by analyzing patient data obtained from the pharmacy system and other external data sources. By incorporating workflow robotics, this advancement would alleviate pharmacists from the burden of solely identifying critical drug-related issues.[22,23]

AI has the potential to significantly impact and redirect the focus of pharmacists, shifting their attention from simply dispensing medications to offering a wider range of patient-care services. By harnessing the power of AI, pharmacists can assist individuals in maximizing the benefits of their medications and maintaining better overall health. Moreover, AI presents an opportunity for enhanced collaboration among various entities involved in the care of a single patient within the pharmacy setting. From the patient’s perspective, AI not only holds the promise of improved healthcare services delivered by healthcare professionals but also serves as a valuable tool for providing guidance on accessing the most cost-effective healthcare options and facilitating effective communication with healthcare providers. In addition, AI can optimize the utilization of data obtained from wearable devices, offer everyday lifestyle recommendations, integrate dietary and exercise plans, and support treatment compliance and adherence.

The integration of human expertise and resources with AI characterizes AI. As the exploration of AI progresses, with numerous intriguing applications underway, some may view it as a necessary component, even if they perceive it as a foe. Hence, it is highly advised that pharmacists acquire the necessary technical skills that facilitate AI integration. Comprehensive education and exposure to AI are imperative across all facets of pharmacy practice. Data science classes or pharmacy residencies focusing on AI subjects should be accessible to pharmacists interested in more hands-on participation in AI advancement, management, and utilization. Given the rapid advancements in these technologies, the pharmacy education system must remain adaptable to ensure that our profession is prepared to oversee these shifts in healthcare.

Pharmacy practice plays a vital role in the delivery of healthcare, ensuring the safe and effective utilization of medications in both clinical and community environments.[24] Clinical pharmacists (CP) in hospital settings work closely with healthcare providers to enhance medication utilization, oversee intricate treatment plans, and deliver individualized care. As personalized medicine gains traction, its significance in the healthcare system continues to grow. On the other hand, community pharmacists offer a wide range of healthcare services beyond simply dispensing medications, such as health education, preventive care, and guidance on medication safety and effectiveness.[25,26]

In the past few years, there has been a notable transition towards personalized medicine, which involves customizing treatments based on the unique characteristics of each patient. This shift requires adjustments in the tools and methodologies employed in the field of pharmacy practice. AI is a revolutionary technology that has the capacity to accelerate this transformation. AI, characterized by machines emulating intelligent human behavior, holds immense potential in healthcare, encompassing areas such as data analysis, disease prognosis, and tailored treatment suggestions.[27,28] This could have a significant impact on various aspects of patient care. For pharmacists, this innovative technology has the potential to streamline repetitive tasks, reduce errors, and deliver precise patient-centered care. Although the advantages of AI in healthcare are well-known, its integration into pharmacy practice is still in its early phases. While numerous exploratory and review articles have highlighted the potential of AI in pharmacy, there are fewer original studies demonstrating its effectiveness in enhancing pharmacy services. A systematic review by Ranchon et al. revealed that different AI techniques, particularly ML methods like natural language processing and deep learning, have been utilized in pharmacy practice. AI-powered tools analyze electronic prescriptions to identify potential errors such as incorrect dosages, drug interactions, or contraindications based on patient history. By examining context and semantics, AI can detect discrepancies that may be overlooked during manual medication order reviews. Deep learning algorithms are integrated into automated dispensing systems to ensure precise medication selection and dispensing. Furthermore, these algorithms predict medication restocking requirements by analyzing past usage and current inventory levels, thereby optimizing supply chains. AI chatbots and virtual assistants, equipped with natural language processing capabilities, provide patients with information on medications, advice on side effects, and answers to prescription-related inquiries. These AI resources can interpret patient queries, even those expressed in simple terms, and provide clear explanations that enhance patient comprehension and adherence.[29]

The transformative potential of AI’s advanced technologies is immense, as it is poised to revolutionize the practice of clinical pharmacy by seamlessly integrating its duties in a synergistic manner. This presents pharmacists with an unparalleled chance to enhance their collaboration with clinicians, particularly in hospital environments. Through AI-driven platforms, communication between pharmacists and clinicians is streamlined, providing real-time updates on patient medications, potential drug interactions, and dosage recommendations derived from integrated patient data. These seamless integrations establish pharmacists as indispensable members of healthcare teams, guaranteeing both patient safety and therapeutic effectiveness.

AI has the capability to process large amounts of patient data and clinical guidelines, which enables it to propose appropriate drug therapies, dosages, and combination medications. An example of this is IBM Watson for Oncology, a tool driven by AI that helps oncologists in identifying potential cancer drug treatments tailored to individual patient profiles. By analyzing a patient’s medical history along with extensive medical literature and clinical trial data, Watson recommends therapies that could be most effective for specific cancer types.[30]

AI has the capability to identify patterns in medication errors and adverse drug reactions, providing valuable insights into potential systemic issues. This, in turn, can inform strategies for improving the quality of healthcare. A notable example is Google’s AI, which has showcased its proficiency in predicting adverse events in hospitals and devising preventive measures. As a result, patient safety and the overall quality of care have been significantly enhanced.[31] For instance, in 2018, Google introduced a notable breakthrough in the realm of cardiovascular disease research. Utilizing ML methodologies, their team of researchers developed an AI model with the ability to assess an individual’s risk of heart disease by analyzing the retina.[32]

One instance to consider is the Chemical Substructure Representation, an AI tool capable of performing a comprehensive evaluation of concealed characteristics illustrating drugs or drug combinations through shared structures. Subsequently, it employs deep learning methods to establish a universal depiction of these pharmaceutical substances. This can be advantageous in predicting potential interactions between drugs and identifying adverse reactions, thereby augmenting the safety and effectiveness of pharmaceutical therapies.[33] AI’s predictive analytics capabilities possess the potential to optimize patient care through the anticipation of health trajectories, which are determined by current treatment plans and patient health histories. Notably, tools such as Google DeepMind have demonstrated their proficiency in accurately predicting hospital readmission rates for heart failure patients. This advancement equips pharmacists with a more precise tool for tailoring care strategies to individual patients, thereby enhancing the overall quality of healthcare delivery.[34]

AI plays a crucial role in patient education by converting complex medical advice into easily understandable and accessible formats. Interactive Chatbots like Buoy Health provide tailored guidance on medication usage, dietary adjustments, lifestyle changes, and treatment strategies, promoting enhanced comprehension and adherence. During consultations, AI’s capacity to quickly assess recent research, protocols, and patient data empowers pharmacists to promptly formulate evidence-backed treatment suggestions using resources such as PathAI, ultimately enhancing cooperative healthcare practices.[35] AI algorithms are highly beneficial in therapeutic drug monitoring (TDM), as they are capable of analyzing patient data, such as biomarkers, to evaluate the effectiveness and safety of medication therapies. For instance, platforms like Dosis personalized dosing utilize real-time patient information to recommend ideal medication dosages tailored to individual responses, ultimately leading to improved therapeutic results.[36] In general, AI has the potential to empower CP by allowing them to focus on providing exceptional, patient-centered care instead of being burdened by administrative tasks. This shift in responsibilities can result in enhanced patient outcomes and heightened job satisfaction and professional fulfillment among pharmacists.

AI has the potential to transform the landscape of community pharmacy practice by expanding its influence beyond the traditional roles of pharmacists. One key area where AI can make a significant impact is in supply chain management. Through the use of advanced algorithms, AI can process large volumes of data such as historical sales figures, seasonal variations, local health trends, marketing campaigns, and external variables like weather conditions or disease outbreaks. By analyzing this information, AI can accurately forecast the demand for different medications, enabling pharmacies to optimize their inventory levels. This proactive approach helps prevent stockouts of high-demand products and reduces the risk of overstocking items that may expire before they are sold. In addition, AI streamlines the reordering procedures in order to uphold the most efficient inventory levels. Through continuous monitoring of stock levels and the automatic generation of purchase orders when levels dip below specific thresholds, this technology not only saves staff time but also guarantees the constant availability of essential medications for purchase. Moreover, AI tools have the capability to assess suppliers based on various factors such as reliability, cost, delivery speed, and product quality. For example, this technology can analyze past data on delivery time, cost, and quality from each provider, subsequently recommending the most suitable supplier for each product. This aids pharmacies in maintaining a high-quality inventory while simultaneously minimizing expenditure.[37] AI has the potential to enhance Automated Dispensing Systems. It can significantly improve the accuracy and precision of dispensing, analyze past errors, and utilize ML algorithms for ongoing system optimization.[38] AI’s capacity to rapidly organize and categorize medications, predict maintenance requirements, and customize dispensing based on individual patient specifications will not only optimize operational efficiency but also promote personalized patient care. By integrating automated dispensing with other facets of pharmacy management, such as inventory control and EHRs, AI can further streamline medication processes from prescription creation to billing. AI’s capability to cross-reference dispensed medication with patients’ health records offers an additional safeguard, promptly notifying pharmacists of potential drug interactions or patient allergies, thereby significantly enhancing patient safety. In addition, AI aids in advancing public health monitoring. AI systems can analyze extensive health data sets to identify patterns in disease outbreaks, drug usage trends, and other public health concerns that may necessitate a response from pharmacists or public health organizations.

For example, through the analysis of data, AI has the potential to identify an increase in reported flu symptoms in specific areas.[39] Following this, pharmacists in those areas were then informed about possible outbreaks. This will encourage them to accumulate antiviral medications, flu shots, and over-the-counter remedies in preparation for an increase in demand. In addition, AI can assist pharmacists in addressing health equity issues by examining data such as zip codes, demographics, and medical records. Through analyzing zip codes alongside socioeconomic factors and health outcomes, AI can identify regions with significant health disparities, provide personalized patient suggestions, and offer insights for targeted policy interventions. This knowledge can lay the groundwork for more strategic and efficient healthcare programs. For instance, AI could identify areas with high rates of adult diabetes, leading to the development of educational initiatives, provision of complementary medications, or improved access to care for affected individuals. Lastly, AI tools can enhance the profitability of pharmacies. AI algorithms have the capability to forecast medication demand by analyzing data such as previous sales and local health trends. This ensures that inventory management is optimized, minimizing the occurrence of stockouts or overstocks that could have a negative impact on profits. In addition, the integration of AI leads to improvements in operational efficiency, which in turn contributes significantly to profitability. By automating tasks such as medication dispensing and inventory management, pharmacy staff are able to allocate their time and attention toward providing higher-value services such as patient counseling and care. This not only enhances the quality of service but also has the potential to create new revenue streams. AI tools like chatbots or automated reminders can enhance patient engagement. For instance, an AI chatbot can offer personalized medication advice, thereby promoting better adherence among patients. This has the potential to improve patient health outcomes and foster loyalty, as customers are more likely to return when they receive personalized and attentive service. From a cost-saving perspective, AI can help reduce dispensing errors that can result in costly reimbursements, reputational damage, patient harm, and equipment breakdowns. By preventing such errors, unexpected repair costs can be minimized. Furthermore, AI facilitates data-driven decision-making. For example, AI can analyze sales data to identify high-margin products, enabling pharmacies to focus their marketing efforts on these items to increase sales. AI can also identify trends that indicate new revenue opportunities, such as unmet needs for specific health products or services within local communities. Efficient staffing through the use of AI not only enhances patient satisfaction but also directly impacts the profitability of pharmacists. Optimizing staffing levels enables pharmacists to attend to a larger number of patients while maintaining the quality of service, resulting in increased revenue. In addition, satisfied customers often lead to repeat business and referrals through word-of-mouth, further enhancing profitability. By avoiding overstaffing and reducing overhead costs, combined with the increased revenue from efficient service, the overall financial health of the pharmacy is improved. The potential of AI to transform both hospital and community pharmacy settings is remarkable. AI has the capability to revolutionize pharmacy practice by streamlining administrative processes, making data-driven insights for better decision-making, and providing personalized patient care. This technology can significantly enhance operational efficiency by predicting medication demand, improving dispensing accuracy, and automating tasks like inventory management. Furthermore, AI’s involvement in patient engagement and education fosters loyalty and improves health outcomes. Through predictive analytics and data interpretation, AI can inform preventative care strategies, enhance quality control measures, and uncover revenue opportunities. Therefore, AI plays a crucial role in shaping the future of pharmacy, promising increased job satisfaction for pharmacists and more profitable and efficient pharmacy practices.

Decades of research in clinical pharmacology have focused on utilizing mathematical models to understand the relationship between drugs and changes in pharmacokinetics (PK) and pharmacodynamics over time. In the pursuit of improved therapies, there is a shift toward digitization. Digitized drug interaction databases contain both qualitative and quantitative data on various factors such as comedications, excipients, food products, organ function, and genetics that can impact drug exposure in the human body. In addition, digital biomarkers obtained through wearable and implantable devices offer real-time data, allowing clinicians to better comprehend how medications affect diseases and overall health. The integration of EHRs and genomic data has made digital twins and virtual populations feasible. AI models can now utilize physician notes and laboratory records for predictive modeling, thanks to advancements in natural language processing techniques. The Internet of Things enables seamless connectivity between medical devices and databases, facilitating the collection of vast amounts of electronic data through wearable devices. The combination of curated databases, real-world evidence, patient-centric sampling, and wearable technology lays the groundwork for AI-driven clinical pharmacology to revolutionize patient care. An AI/ML-powered dose-recommender system has the potential to revolutionize patient care. By integrating data from various domains such as safety and efficacy measures, EHRs, disease information, treatment history, and patient-reported outcomes, this system can offer personalized dosing options. This tailored approach not only improves treatment efficacy but also reduces the occurrence of adverse events. In the field of oncology, reinforcement learning-based algorithms have shown promise in predicting and modifying doses during treatment, enabling precision dosing for patients.[40] The Dose–Response Network employs DL techniques to predict individual patient outcomes across various points on dose–response curves. The reliability of AI in suggesting doses under hypothetical circumstances is uncertain. Nevertheless, generative adversarial networks have the potential to transform personalized dose–response curve forecasts by leveraging existing data to extrapolate to uncharted areas of the dose–response landscape.[41]

The influence of CP on prescribing guidelines is primarily determined by research focusing on drug-drug interactions (DDIs), drug–food interactions, bioavailability, and PK alterations in specific patient groups. Nevertheless, the available data in these areas are restricted when compared to the actual drug interactions observed in real-world clinical scenarios. AI combined with clinical pharmacists’ expertise has the capacity to extend beyond the patient cohorts studied in clinical trials. Novel algorithms utilizing knowledge graphs (KGs) have demonstrated the ability to forecast unforeseen adverse drug reactions,[42] DDIs, and drug–food interactions.[43] Bougiatiotis et al. indicated the effectiveness of biomedical literature knowledge graphs (KGs) and link prediction models in evaluating DDIs in both Alzheimer’s disease and lung cancer.[44] With the assistance of the KG framework, clinical pharmacologists can utilize existing physiologically based PK expertise and data sources to forecast potential and hazardous DDIs resulting from the concurrent administration of multiple drugs. This enables the incorporation of both recognized and unidentified (potential) DDIs in patient information leaflets.

In the realm of clinical practice, TDM provides dosage suggestions for medications with considerable variability in PK and a narrow therapeutic index. AI/ML has shown improved dosage recommendations for propofol and remifentanil, with reduced errors in predicting the bispectral index during anesthesia compared to conventional modeling techniques.[45] AI/ML approaches have advanced to the point where they can identify intricate and nonlinear connections, as well as comprehend the impact of both intrinsic and extrinsic factors on the variability of PK within various subpopulations. By integrating ML capabilities with population-based approaches, we anticipate a better understanding of PK variability and the ability to adjust dosage for specific patient subgroups. The US Food and Drug Administration recommends incorporating patients who have a high probability of exhibiting disease-related endpoints (prognostic indicators) and patients who are expected to respond positively to the treatment in order to enhance clinical trial strategies. (predictive indicators).[46]

The advancement of pharmacometric modeling automation, along with the development of predictive modeling based on neural networks/neural ordinary differential equations and innovative algorithm-based clinical trial designs, sets the groundwork for the future of model-based drug development (MBDD). It is hypothesized that the utilization of AI techniques to extract information from unstructured data (such as imaging/electronic records) could enhance current MBDD approaches by refining personalized forecasts and decision-making processes throughout clinical trials. Nevertheless, through the integration of AI and human intelligence, Michelson et al. conducted a swift meta-analysis to uncover findings suggesting ocular toxicity as a potential side effect of hydroxychloroquine in a significantly shorter timeframe (<30 min) compared to traditional meta-analyses.[47] Similarly, unsupervised ML assisted with the automated screening and study selection process for meta-analysis.[48]

Significant advancements have been achieved in the realm of data privacy, particularly with the enforcement of the influential “General Data Protection Regulation.” Healthcare organizations have shown increased awareness of data privacy issues and have established structural components such as infrastructure and governance boards to oversee data usage. Nevertheless, the implementation of these measures in a clinical environment necessitates robust information technology platforms, which are typically provided by large technology companies. In order to safeguard patient privacy and regulate the utilization of data by external parties, stringent data privacy regulations are imperative on a global scale. The integration of AI tools in clinical settings also raises ethical concerns related to data privacy and patient consent.

Currently, there is a lack of recommendations and guidelines for utilizing AI in clinical practice, highlighting the need for well-defined ethical standards that address patient consent and awareness of AI’s role in making life and health decisions.

The utilization of AI extends beyond the mere selection of the most optimal compound and prediction of its in vivo activity. It can also be harnessed to discover new applications for existing drugs with a well-established risk/benefit ratio. This process of drug repositioning is particularly beneficial for small patient populations where conducting clinical trials is impractical and the expenses associated with drug development are often deemed exorbitant. Various forms of drug repositioning exist, all of which necessitate extensive datasets and rigorous computational analyses. One approach involves comparing existing drugs with novel compounds to identify substances with similar properties that could effectively treat the same conditions as the original drug. Another approach focuses on comparing disease characteristics and pathways to identify drugs that may be effective for diseases with analogous features. Frequently, these two strategies are combined, such as by correlating gene expression profiles of a disease with those induced by various drugs to find potential matches. Al-Taie et al. employed this integrated approach to discover novel therapies for colorectal cancer (CRC). An AI model was trained to identify distinct subpopulations within the clinical data. Subsequently, the unique gene profiles of each subgroup were cross-referenced with the drug database to pinpoint potential candidates for repurposing. Out of the top 16 drugs identified through this method, 12 were cancer-related drugs, with eight being approved for CRC treatment, underscoring the promise of this innovative approach.[49]

Clinical trials and real-world evidence

Following the preclinical development of a novel medication, the commencement of the clinical trial phase ensues, leading to increased expenses. One of the primary obstacles encountered during clinical trials pertains to patient selection and enrollment. In cases where an insufficient number of patients can be recruited, the trial may be forced to halt. AI applications have the potential to enhance patient recruitment by analyzing extensive EHR data to identify eligible patients who are more likely to participate. This optimization ensures that less time and resources are wasted on patient recruitment. In addition, AI can aid in monitoring patients throughout the trial by utilizing EHR data mining and real-time data from devices like smartwatches and wearables worn by participants. Finally, AI methodologies can facilitate the collection of real-world evidence by extracting relevant data from EHRs and transforming this information into clinically significant outcomes.[50,51] In addition to its application in traditional clinical trials and data analysis, AI has the potential to forecast the drug response of an individual patient through the utilization of digital twins. These digital replicas accurately represent real-world patients and, once trained, can simulate the effects of various treatments on the patient’s disease progression.[52]

Drug treatment optimization

It is often beneficial and crucial to personalize the treatment for the numerous drugs available in the market. In the case of drugs with a narrow therapeutic range, TDM is employed to customize the dosage. The utilization of statistical prediction models is frequently employed to extrapolate TDM data and determine drug exposure as well as optimal treatment approaches.[53]

The utilization of AI in the selection and administration of personalized drug regimens is crucial, particularly when addressing intricate diseases that require unique treatment protocols for many patients. The remarkable achievements of modern medicine, such as vaccinations and antibiotics, have largely been effective across various patients without significant adjustments. However, the emergence of precision and personalized medicinal approaches aims to tackle diseases that lack universal solutions. To comprehensively comprehend and combat complex diseases like cancers, it is imperative to employ computational tools reliant on AI engines to integrate the extensive biological, chemical, and medical data available. It is important to note that there is no AI system that can serve as a universal solution, providing treatment recommendations for all diseases.[54]

To conclude, the undeniable impact of AI in clinical pharmacy and hospital settings is exhilarating. AI technology is revolutionizing the operations of healthcare professionals by predicting medication demand, enhancing dispensing accuracy, and automating tasks such as inventory management. Moreover, its role in patient engagement, education, and predictive analytics not only fosters loyalty and improves health outcomes but also informs preventative care strategies and uncovers revenue opportunities. As AI continues to shape the future of pharmacy, it not only promises increased job satisfaction among pharmacists but also offers the potential for more profitable and efficient practices in healthcare settings. The exciting potential of AI to transform the landscape of clinical pharmacy and hospital settings holds great promise for the future of healthcare. AI offers various benefits in clinical pharmacy, including:

1. Enhanced medication management: AI has the capability to assist pharmacists in precisely monitoring medication stock, forecasting demand, and minimizing errors during dispensing

2. Improved patient care: Utilizing AI tools enables the analysis of patient data for the customization of treatment plans, detection of potential drug interactions, and enhancement of medication adherence

3. Enhanced productivity: AI automation simplifies repetitive tasks such as data entry, enabling pharmacists to dedicate more time to intricate patient care activities

4. Financial benefits: Through the optimization of inventory management and reduction of medication errors, AI can assist pharmacies in cutting costs and functioning with greater efficiency

5. Sophisticated data analysis: AI algorithms have the capability to scrutinize extensive datasets to recognize patterns, forecast patient outcomes, and guide evidence-based decision-making in clinical practice.

In summary, the integration of AI in clinical pharmacy has the capacity to transform the approach pharmacists take in providing care, enhancing patient results, and improving the overall effectiveness of pharmacy operations.

AUTHORS’ CONTRIBUTIONS

Hadi Hamishehkar: Conceptualized the study, contributed to the design and methodology, supervised data interpretation, critically revised the manuscript, and approved the final version for submission.

Mehrdad Shahidi: Participated in manuscript drafting and editing, conducted literature review, coordinated the preparation of figures and tables, managed correspondence and communication with the journal, and approved the final manuscript for submission.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.