AI in Pharmacy: Advancing Drug Discovery, Clinical Practice, Patient Care and Emerging Automated Technologies

Artificial Intelligence (AI) is an evolving branch of computer science dedicated to developing systems capable of performing tasks that typically require human intelligence—such as learning, reasoning, problem-solving, and decision-making¹⁸. The foundation of AI lies in gathering and processing vast amounts of data, applying sophisticated algorithms, and developing models that can produce accurate, reliable, and meaningful insights¹⁵. By combining advanced computational power with modern statistical and analytical techniques, AI enables faster data interpretation and more precise decision-making across scientific disciplines¹⁷.  

In recent decades, the field of pharmacy has undergone a remarkable transformation driven by technological advancements¹⁸. Increasing prescription demands, pharmacist shortages, higher operational costs, and declining reimbursement rates have pushed the industry toward automation and digital solutions². These innovations have improved workflow efficiency, reduced human error, and enhanced accuracy and safety in various pharmaceutical settings³. Among these technologies, AI has emerged as one of the most promising tools for improving the quality and effectiveness of pharmacy practice¹⁷.

AI is rapidly becoming a key driver in pharmacy and healthcare innovation. It can analyze large volumes of medical and pharmaceutical data to predict potential drug interactions, monitor therapeutic responses, and optimize medication management¹. By using predictive models and pattern-recognition systems, AI assists pharmacists in making more informed clinical decisions, personalizing treatment plans, and identifying possible risks before they impact patient outcomes⁵. Therefore, pharmacists must continuously upgrade their knowledge and readiness to integrate AI tools into their professional practice, as these technologies are no longer futuristic—they are becoming a part of everyday healthcare^5-7.

Clinical pharmacy, a specialized branch of the profession, focuses on optimizing medication therapy and improving patient outcomes through evidence-based care²⁹. Clinical pharmacists use data-driven approaches to design care plans, resolve drug-related problems, and collaborate with other healthcare professionals¹⁹. With the inclusion of AI, this role becomes even more dynamic, as intelligent systems can now assist in monitoring patient responses, automating documentation, and supporting clinical decision-making^24-26.

Beyond pharmacy practice, AI is also revolutionizing the drug discovery and development process, traditionally known to be time-consuming, complex, and costly. In the past, identifying and developing new drugs relied heavily on trial-and-error methods involving extensive laboratory work and animal testing³⁰. Today, AI accelerates this process by analyzing chemical and biological datasets to identify potential drug targets, predict molecular interactions, and design new therapeutic compounds. Machine learning models can screen millions of molecules virtually, selecting the most promising candidates for further testing—significantly reducing research timelines and costs²⁸. AI also enhances drug development by predicting pharmacokinetic and toxicity properties (ADMET), optimizing chemical synthesis routes, and supporting precision medicine approaches that tailor treatments to individual patients²².

Overall, the integration of AI into pharmacy and drug development represents a new era of intelligent, data-driven healthcare. By combining human expertise with computational intelligence, the pharmaceutical field is becoming more innovative, efficient, and patient focused. AI not only strengthens scientific discovery but also ensures safer, faster, and more effective treatments—paving the way for a smarter future in medicine.

The History and Development of Artificial Intelligence in Drug Discovery

Artificial Intelligence (AI) has been influencing the field of pharmaceutical sciences for nearly three decades. Its true impact began to grow in the late 1990s, when major advancements in algorithms and computational technology took place^17-18. Over the years, AI research has seen alternating phases of rapid progress and slowdowns, largely driven by the shift from traditional neural networks to more advanced deep learning systems. The combination of powerful algorithms, vast biomedical data, and enhanced computational capacity has been the driving force behind the growing role of AI in modern drug discovery and development¹⁶.

fig1: History and development of AI in drug discovery

In the early stages of drug research, scientists relied heaily on laboratory experiments to assess a compound’s absorption, distribution, metabolism, excretion, and toxicity— commonly known as ADMET properties². These tests, often performed through in vitro and in vivo methods, generated useful data but were expensive, time-consuming, and raised ethical concerns due to the use of animal models¹⁶. As the number of new molecular entities increased, researchers began exploring computer-based methods—known as in silico models—to predict these properties more efficiently³⁰. The continuous accumulation of experimental data and progress in computer science have since made in silico modelling and online prediction platforms essential tools for early pharmacokinetic and toxicity evaluation¹⁷.

In recent years, AI has evolved from being a theoretical concept to a practical technology in the pharmaceutical sector. Although no drug completely designed by AI has yet received full approval from global regulatory bodies, several AI-driven compounds have already reached and accelerated the early phases of clinical trials³. A particularly important development came when deep learning systems began to reverse-engineer chemical synthesis routes, an innovation compared by many experts to the revolutionary moment when AlphaGo mastered the complex^18-17.  Several successful examples highlight the growing influence of AI in real-world pharmaceutical research. AI platforms have been used to repurpose existing drugs for rare diseases in record time, identify new therapeutic targets, and even design novel molecules from scratch. For instance, some AI systems have managed to identify and design potential treatments for rare genetic conditions within a fraction of the time required by traditional methods. Another landmark achievement came when AI was used to design kinase inhibitors in only a few weeks—a process that once took months of laboratory testing. Although these compounds still require refinement to improve selectivity and safety, such achievements illustrate the speed and innovation AI brings to medicinal chemistry¹⁴.

DeepMind’s AlphaFold project also marked a revolutionary moment by accurately predicting the three-dimensional structures of proteins—an achievement scientists had struggled with for decades. More recently, AI-designed drug candidates have successfully advanced into mid-stage clinical trials, showing how computational intelligence can directly contribute to real-world pharmaceutical development¹⁸.

Despite these significant advances, AI in drug discovery still faces notable challenges. Issues such as data bias, limited interpretability of models, and poor generalization across diverse datasets can affect accuracy and reliability. Models that appear highly successful during training may fail to perform effectively in real-world scenarios if they lack transparency or are based on incomplete data^16-17. Therefore, while AI continues to transform the landscape of pharmaceutical research, its long-term success depends on continuous validation, ethical use of data, and strong collaboration between computational and experimental sciences¹⁸. The history and development of AI in drug discovery reflect a journey from early algorithmic experiments to an era of intelligent, data-driven research. The integration of AI with human expertise holds the promise to accelerate innovation, reduce costs, and ultimately improve the discovery of safer and more effective medicines for the future.

ARTIFICIAL INTELLIGENCE IN CLINICAL PHARMACY

Clinical pharmacy is a specialized field focused on optimizing medication use to promote health, wellness, and disease prevention through a pharmacist-driven approach to patient care¹⁹. Clinical pharmacists play a key role in ensuring safe, effective, and cost-efficient drug therapy to improve patient outcomes. They achieve this by designing pharmaceutical care plans, resolving drug therapy problems, participating in interdisciplinary rounds and consultations, educating patients during and after hospitalization, and maintaining thorough documentation and data management.

Artificial Intelligence (AI) and its subset, Machine Learning (ML), are increasingly being integrated into various core aspects of clinical pharmacy practice:

Medication management and Safety

• Drug-Drug Interaction (DDI) and Adverse Drug Reaction (ADR) Detection:

Records (EHRs), laboratory results, and medication profiles, to rapidly detect potential DDIs and predict the likelihood of ADRs before they occur¹⁹.

• Dosage Optimization:

AI systems can recommend personalized dosage adjustments by considering a patient’s genetic profile, organ function (such as kidney or liver health), and real-time clinical data, making them particularly valuable in fields like paediatrics, geriatrics, and polypharmacy²⁰.

• Medication Adherence Monitoring:

AI-driven tools can monitor and predict patient non-adherence, enabling pharmacists to take proactive measures by providing targeted support and timely reminders²².

Clinical Decision Support Systems (CDSS)

• Real-time Recommendations:

AI-enhanced Clinical Decision Support Systems (CDSS) provide pharmacists with instant, evidence-based guidance for drug selection and treatment planning by rapidly integrating current clinical guidelines with patient-specific data²².

• Risk Stratification:

Machine Learning (ML) models assist in identifying patients who are at high risk of specific clinical deteriorations, such as sepsis or falls, thereby enabling timely and preventive interventions²³.

Streamlining Operations and Research

• Automation:

AI-powered systems facilitate prescription verification, automated dispensing, and inventory management, minimizing human errors and allowing pharmacists to dedicate more time to direct patient care²⁶.

• Drug Discovery and Repurposing:

Although not a direct component of clinical pharmacy practice, AI significantly influences the availability of medications by expediting the discovery of new drug  candidates and uncovering novel therapeutic uses for existing drugs through the analysis of extensive chemical and biological databases²⁷.

• Clinical Trial Optimization:

AI aids in identifying appropriate candidates for clinical trials, thereby enhancing the efficiency and effectiveness of the research process²⁸.

Focus on Medication Safety and Adverse Drug Events (ADE) Prediction

The most immediate and critical application is in improving drug safety:

• Predicting ADEs And Drug interactions:

Governments and researchers are increasingly investing in AI to predict adverse drug events (ADEs) and potential drug interactions. A recent UK MHRA initiative, for example, leverages AI and real-world NHS data initially focusing on cardiovascular drugs to develop a tool that helps clinicians assess interactions and personalize prescriptions more safely and effectively¹⁹.

• Structured ADE data:

Structured ADE data being advanced through studies like the one by CedarsSinai, which developed AI models such as OnSIDES to analyze thousands of drug labels and clinical records, generating comprehensive, machine-readable databases of potential adverse drug events that enhance predictive models in patient care²².

• Machine Learning Algorithms:

Review studies indicate that machine learning algorithms such as Random Forest and Support Vector Machines (SVMs) are commonly used to predict drugrelated side effects, with the most effective models integrating chemical, biological, and phenotypic features from patient data²⁰.

Personalized Medicine and Pharmacogenomics PGx AI is proving indispensable for tailoring drug therapy to an individual's unique biological makeup:

• Championing PGx Implementation:  

Articles emphasize that AI can empower pharmacists to lead the implementation of pharmacogenomics (PGx) by rapidly analyzing patients’ genetic profiles, identifying actionable variants, and generating prescribing recommendations such as dosage adjustments or alternative drug options²⁶.

• Generative AI for Consultation: 

Specialized AI assistants, such as those developed using GPT-4, are currently being tested to efficiently and accurately respond to patient and provider inquiries about pharmacogenomic (PGx) testing results²⁷.

Enhancing Clinical Workflow and Decision Support

AI tools are being integrated into daily practice to augment the pharmacist's role:

• Clinical Decision Support System (CDSS): 

AI algorithms integrated into Clinical Decision Support Systems (CDSS) analyze large volumes of patient data, including electronic health records (EHRs) and lab results, to provide personalized, evidence-based recommendations that assist pharmacists in selecting optimal therapeutic regimens²³.

• Operational Efficiency:

In both hospital and retail settings, AI has been shown to greatly enhance pharmaceutical inventory management, improve the accuracy of drug dispensing, and streamline drug distribution processes, enabling clinical pharmacists to focus more on complex aspects of patient care²⁵.

• AI for Patients at Home: 

Researchers are creating AI-based systems designed to reduce medication errors in patients, especially those with cognitive disabilities, by assisting with and monitoring their at-home treatment plans²⁴.

ARTIFICIAL INTELLIGENCE IN PATIENT CARE

AI can significantly influence the role of pharmacists, shifting their focus from dispensing medications towards offering a wider array of patient-care services. By leveraging AI, pharmacists can help patients optimize the benefits of their medicines and improve their health, while also fostering greater collaboration across different healthcare entities serving the same patient. For patients, AI offers the potential for improved healthcare services and acts as a useful tool for finding cost-effective healthcare options, enhancing communication with professionals, optimizing wearable data, providing daily lifestyle guidance (including diet and exercise), and supporting treatment compliance and adherence^1-8.

AI's Role in Disease Diagnosis and Treatment

AI is rapidly integrating into the healthcare sector, significantly improving disease diagnosis, which is essential for creating thoughtful treatments and ensuring patient wellbeing. Within the biotech field, AI is recognized as playing a crucial supporting role in the efforts to combat and control viruses and other diseases, potentially accelerating the discovery of solutions faster than would otherwise be possible⁵.  It specifically refers to the exhibition of human-like intelligence or behaviours by a computer or machine. While robotics is the creation of machines for tasks (often repetitive), and automation is the process where a designed program allows machines like traditional robots to independently move or carry objects using sensors, AI focuses on developing machines with the ability to perform tasks typically associated with a human being⁶.

AI's Impact on Pharmacy: From Hype to Hope

AI technology is rapidly transforming pharmacy, moving drug discovery, drug delivery formulation, development, and other healthcare applications from mere hype to tangible hope. The use of AI models enables predictions of in vivo responses, pharmacokinetic parameters of therapeutics, and appropriate dosage. This capability, particularly in pharmacokinetic prediction via in silico models, significantly aids in improving a drug's effectiveness and affordability⁴.

Two Main Categories of AI Advancements

Advancements in AI technology, as applied to this field, broadly fall into two main categories:

• Traditional Computing Approaches: These include methods like expert systems, which are designed to mimic human experiences and derive conclusions based on fundamental rules and stored knowledge⁷.
• Artificial Neural Networks (ANNs): These systems represent the second, more advanced category, working to simulate the functions of the human brain for processing complex data and making predictions².

AI has numerous applications in hospital-based healthcare systems, particularly for individualizing patient care by helping to organize dosage forms, select appropriate administration routes, and determine the best treatment policies.

Maintaining of medical records:

AI systems significantly simplify the complex task of maintaining patient medical records by streamlining the collection, storage, normalization, and tracing of data. For instance, the Google DeepMind Health project has proven beneficial in achieving faster, better healthcare by quickly retrieving medical records, even assisting institutions like the Moorfields Eye Hospital NHS in improving eye treatments².

Treatment plan designing: 

AI technology is crucial for designing effective treatment plans, especially during patient critical conditions where selecting a suitable approach is challenging; the system considers prior patient data, clinical expertise, and reports. A prime example is IBM Watson for Oncology, a cognitive decision support system that rapidly analyses a patient's data against millions of pages of literature, hundreds of journals, textbooks, and thousands of historical cancer cases (gleaned from collaboration with Memorial Sloan Kettering Cancer Centre) to provide informed treatment options for oncology clinicians⁷.

Assisting in repetitive tasks: 

AI technology helps automate repetitive medical tasks, such as examining imaging from X-rays, radiology, ECHO, and ECGs, to effectively detect and identify diseases or disorders. For example, IBM's Medical Sieve is a "cognitive assistant" algorithm with strong analytical and reasoning capabilities, and specialized computer programs utilizing deep learning are now being developed—often through medical startups—to analyze nearly all types of imaging data for specific body parts and conditions, ultimately improving patient outcomes¹.

Health support and medication assistance:

AI technology is increasingly recognized as efficient in health support services and medication assistance. Examples include Molly, a virtual nurse start-up designed with a pleasant voice and face to guide patients through their treatment and support those with chronic conditions between doctor visits. Additionally, the Ai Cure app uses a smartphone webcam to monitor patients and help them manage their conditions, proving particularly useful for individuals with severe medication needs or those participating in clinical trials⁴.

Accuracy of medicine:

AI is having a significant impact on genomics and genetic development, enabling precise analysis of genetic data. Systems like Deep Genomics observe patterns in genetic information and medical records to identify disease-linked mutations, informing doctors about the cellular effects of DNA alteration. Furthermore, AI technology like Human Longevity helps pinpoint the exact location of cancer and vascular diseases in their early stages, while an algorithm designed by Craig Venter can even predict a patient's physical characteristics based on their DNA⁵.

Drug creation:

Developing pharmaceuticals is an extremely long and expensive process, often taking over a decade and costing billions. However, AI is accelerating this: the Atom wise technology, using supercomputers and molecular databases, can quickly identify potential therapies, demonstrated by its ability to complete a virtual drug search for the Ebola virus in just one day (a task that typically takes months or years) and identify two promising existing drugs. Similarly, a Boston-based biopharma company leverages big data and AI to analyze biological and environmental patient data to understand why some individuals survive diseases, aiding in drug discovery, design, and overall healthcare problem-solving⁵.

AI helps people in the health care system:

The "open AI ecosystem," recognized as a top 10 promising technology in 2016, is useful for collecting and comparing data from social awareness algorithms. In healthcare, these ecosystems can analyze the vast patient data (including lifelong medical history and treatment information) to provide personalized suggestions regarding the patient's lifestyle and habits¹.

Healthcare system analysis:

When healthcare data is fully digitized, retrieval becomes simple; for instance, the Netherlands maintains 97% of its treatment invoices, physician, and hospital names in digital format for easy access. Local company Zorgprisma Publiek utilizes IBM Watson cloud technology to analyze these invoices, allowing them to immediately recognize and address any mishaps, ultimately improving care and reducing unnecessary patient hospitalizations¹⁶.

The core objective of AI in healthcare is to analyze the link between prevention/treatment methods and patient outcomes, leading to applications in areas like diagnosis, personalized medicine, and drug development. Given the growing focus on quality care, pharmacies, as highly accessible and affordable stakeholders, can use the current technology boom (including health trackers and wearables) to evolve into health management centers—moving beyond just dispensing medication—by providing personalized advice, guidance, and an expanded suite of services (like immunizations and screenings) while using real-time data to monitor and follow up with at-risk patients⁷.

Healthcare continues to advance rapidly, with technological innovations transforming medicine much like other industries. In today’s era, where precision, efficiency, and personalized care are vital, robotics and automation are becoming essential in pharmacy and medicine. These innovations are reshaping healthcare by improving patient outcomes, optimizing workflows, and enhancing overall efficiency^1-4.

THE ROLE OF AI ROBOTICS IN PHARMACY

Artificial intelligence (AI) and robotic technologies are revolutionizing healthcare and medicine. AI supports healthcare professionals by delivering detailed, personalized insights about each patient, leading to more effective treatments and better outcomes. By automating routine tasks, AI and robotics enhance the efficiency of healthcare delivery, reduce costs, and improve patient care. Additionally, AI and supercomputing are expediting drug discovery and design, paving the way for new and more effective therapies. In pharmacy, robotic process automation (RPA) is transforming operations by managing repetitive, time-consuming tasks like data entry and administrative functions with speed and accuracy.  These technologies bring precision, accuracy, and automation to multiple areas of pharmaceutical manufacturing, patient care, and medical procedures. With an aging global population and a growing demand for healthcare services, professionals and institutions are facing greater challenges such as limited resources, escalating costs, and the rising need for personalized treatment^6-10.

Robotic Technologies in Pharmaceutical Manufacturing

Artificial intelligence (AI), a field of computer science, is often confused with robotics and automation. Innovative robotic technologies are increasingly implemented in the pharmaceutical manufacturing industry, which has witnessed significant advancements in recent years, with these technologies playing an integral role in enhancing efficiency, precision, and cost-effectiveness⁹.

Robot Pharmacy at UCSF Medical Centre:

The UCSF Medical Centre employs robotic technology in its pharmacy to track and prepare medications, aiming to enhance patient safety. This technology is credited with flawlessly preparing 350,000 doses of medication and has proven itself to be significantly superior to humans in both the size of the doses handled and the accuracy of medication delivery.

Fig:2 Applications of artificial intelligence and robotics in the pharmaceutical and medical fields

Robotic Technologies in Pharmaceutical Manufacturing

In recent years, the pharmaceutical manufacturing industry has seen major progress, with innovative robotic technologies becoming essential for improving accuracy, productivity, and overall cost efficiency.

Robotic Drug Delivery Systems:

New robotic drug delivery systems are being designed to enhance the safety, efficiency, and overall patient experience in medication administration. These advanced systems utilize technologies like artificial intelligence, machine learning, and automation to achieve optimal treatment results. Such innovations are set to transform how drugs are delivered, enabling more accurate and personalized therapies. This section explores examples including robot-assisted delivery methods, implantable drug delivery devices, and nanorobots used for targeted drug delivery⁸.

3D Printing for Personalized Medicine:

One of the most promising applications of 3D printing in drug production is the creation of personalized medications, allowing drugs to be tailored to specific patients by adjusting the dosage, release profile, and even the combination of active pharmaceutical ingredients (APIs) based on individual needs, ultimately leading to more effective treatments and reduced side effects¹³.

Personalised medicine

A key advancement of 3D printing in pharmaceutical manufacturing is the ability to produce personalized medications. This technology allows drugs to be customized for individual patients by modifying the dosage, release rate, and combination of active ingredients according to their specific requirements, resulting in more effective treatments with fewer side effects.

Fig3:3D printing in pharmacy and medicines