Review On Artificial Intelligence Revolutionizing the Pharmaceutical Industry

Dr. Shivappa Nagoba, Maithili Kamble, Avinash Swami, Nivrutti Kotsulwar, Mayur Upade, Amrapali Rajput,

doi:10.5281/zenodo.15393564

Review Paper | Open Access
Volume 03 | Issue 05 | Article Id IJPS/250304700

Review On Artificial Intelligence Revolutionizing the Pharmaceutical Industry
Dr. Shivappa Nagoba* Maithili Kamble Avinash Swami Nivrutti Kotsulwar Mayur Upade Amrapali Rajput
Department of Pharmaceutics, Channabasweshwar Pharmacy College (Degree), Latur-413512, Maharashtra, India.

Abstract

Artificial Intelligence (AI) and Machine Learning (ML) are revolutionizing the pharmaceutical industry, driving advancements across drug discovery, formulation, manufacturing, and clinical trials. AI tools such as molecular visualization software, predictive algorithms like Random Forest, and Principal Component Analysis (PCA) are pivotal in assessing drug stability and designing stable drug-polymer systems, which are essential for developing solid dispersions. Artificial Neural Networks (ANNs) have demonstrated superior accuracy in predicting the crystalline and amorphous content of drugs compared to traditional methods, enhancing the formulation process. AI also plays a crucial role in clinical trials, improving patient recruitment, optimizing selection, and enabling real-time monitoring, all of which reduce failure rates and lower costs. In pharmaceutical manufacturing, AI-driven technologies like computer vision for defect detection and predictive modeling for quality control help streamline operations and ensure product consistency. Furthermore, predictive maintenance techniques, such as Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks, prevent unexpected downtime by forecasting equipment issues. AI also contributes to more efficient product management across companies of all sizes by optimizing research and development, marketing, and analytics. While challenges such as data integrity and regulatory compliance persist, AI and ML continue to hold great promise for improving pharmaceutical processes, making them more efficient, cost-effective, and tailored to individual patient needs.

Keywords

Artificial Intelligence, Machine Learning, Artificial Neural Networks, Predictive Maintenance, etc.

Introduction

The pharmaceutical industry is continuously evolving to address the challenges of drug discovery and formulation, including high costs, extended timelines, and regulatory complexities. Artificial intelligence (AI), encompassing machine learning (ML), deep learning (DL), and natural language processing (NLP), has gained significant attention for its ability to process large datasets, identify patterns, and make predictions with remarkable accuracy. AI's incorporation into formulation and development processes presents previously unheard-of chances to boost productivity, cut expenses, and customize treatments to meet the needs of specific patients. The term Artificial Intelligence was first appeared in 1956. Artificial Intelligence (AI) refers to the capability of computers to mimic human intelligence while minimizing human intervention. It represents an advanced approach to leveraging machines for executing tasks that require physical effort and addressing complex problems in a manner resembling human cognition. AI plays a crucial role in demonstrating intelligent behaviour, acquiring knowledge, providing guidance, and supporting users. It encompasses a broad spectrum of functions, including learning, perception, problem-solving, and devising innovative solutions tailored to specific systems.

Figure no: 1 - Overview of Artificial Intelligence Subfields⁽⁴⁾.

Artificial Intelligence (AI) is a branch of science focused on the development of intelligent systems, particularly computer programs, designed to emulate human cognitive processes and decision-making. It aims to deliver outcomes that closely resemble human attention and reasoning.⁽¹⁾ ⁽²⁾ AI has numerous applications in the pharmaceutical field, including developing treatments for complex and rare diseases like Alzheimer’s and Parkinson’s. It enhances drug adherence and dosage optimization, improves data analytics, accelerates the identification of suitable patients for clinical trials, introduces automated robotic pharmacies for prescription fulfilment and dispensing, and streamlines marketing, logistics, and supply chain management.⁽³⁾ Figure no: 1 is a mind map summarizing the key components and subfields of Artificial Intelligence (AI). It organizes the relationships and methodologies within AI, focusing on its primary areas of application and study.⁽⁴⁾The main branches of Artificial Intelligence are as, Machine Learning (ML), Planning, Deep Learning (DL), Robotics, Natural Language Processing, Expert Systems, Speech, Vision, etc. and the deep knowledge about these branches is given in figure no. 1

Classification of Artificial Intelligence:

According to their calibre
According to their presence (figure no.2)

Based on their calibre, AI system is classified as follows-

Weak intelligence or Artificial narrow intelligence (ANI):

ANI represents AI systems that are designed to perform specific tasks and are limited to their designated scope. These systems cannot operate beyond the programmed functions. ⁽⁵⁾⁽¹⁾ It has features like-

Specializes in a single area or domain.
Lacks adaptability and versatility across tasks.
Does not possess self-learning abilities beyond predefined parameters.

Figure No. 2: Classification of Artificial Intelligence

Examples:

Siri: Apple's virtual assistant, capable of performing tasks like setting reminders and sending messages.
Facial Recognition Systems: Examples include Apple's Face ID and security surveillance tools.
Social Media Tagging: Algorithms used to identify faces in photos.
Google Translate: AI-driven language translation for predefined text and speech tasks.

Artificial General Intelligence (AGI) or Strong AI:

AGI refers to systems with the ability to understand and perform intellectual tasks on par with humans. These systems can adapt to unfamiliar tasks and situations without additional programming.⁽⁵⁾⁽¹⁾ The features of AGI are as -

Multi-functional and adaptable across various domains.
Mimics human intelligence to solve complex problems.
Capable of abstract reasoning, learning, and decision-making.

Examples:

Although AGI is still theoretical, advancements in research suggest that models like OpenAI's GPT aim to bridge the gap towards cross-domain intelligence.

Artificial Superintelligence (ASI):

ASI refers to AI systems surpassing human intelligence in every aspect, including creativity, decision-making, and problem-solving. This represents the ultimate level of AI development.⁽⁵⁾⁽¹⁾ The features of ASI are as-

Possesses advanced creativity and reasoning.
Far exceeds human cognitive abilities.
Autonomous and capable of innovations that redefine science, technology, and art.

Examples:

Currently non-existent, but envisioned as systems capable of solving humanity's greatest challenges or creating entirely new fields of knowledge.

AI can also be classified based on its capabilities, as follows:

Type 1: Reactive Machines:

These AI systems are designed to respond to specific inputs in real time, without the ability to learn from past experiences or store data. The features of reactive machines are as-

Operates solely on current data.
Does not store memory or learn from previous outcomes.
Focuses on performing specific tasks.

Examples:

Deep Blue: IBM's chess-playing AI that defeated Garry Kasparov in the 1990s.
AlphaGo: AI developed by Google, capable of mastering the game of Go by analyzing the board’s current state.

Type 2: Limited Memory:

This category of AI can use historical data for short-term decision-making. It leverages past experiences to make predictions or execute tasks but does not permanently store these experiences. The features of limited memory AI are as-

Uses temporary memory for problem-solving.
Frequently applied in dynamic systems, such as autonomous vehicles.

Examples:

Self-Driving Cars: Systems like Tesla’s AI utilize temporary data for lane detection, obstacle avoidance, and decision-making.
Recommendation Engines: Platforms like Amazon and Netflix analyze historical preferences to predict user interests.

Type 3: Theory of Mind:

This hypothetical type of AI would be capable of understanding human emotions, beliefs, intentions, and thoughts, enabling it to interact more naturally with people. The features of theory of mind AI are as-

Possesses advanced social and emotional intelligence.
Can predict and respond to human behaviour dynamically.

This AI category has not yet been realized, but it represents a significant milestone for the future of human-AI interaction.

Type 4: Self-awareness:

Self-aware AI represents the most advanced form of AI, capable of consciousness and understanding its own existence. This type of AI would also recognize the emotions and thoughts of others. The features of this AI are as-

Fully aware of its own identity.
Capable of introspection and self-improvement.
Would interact with humans on an entirely different level of cognition.

This form of AI remains entirely theoretical and is often depicted in science fiction.⁽⁵⁾⁽¹⁾

Advantages of AI in the Pharmaceutical Industry:

Artificial Intelligence offers the pharmaceutical sector the potential to address challenges that were previously impossible to solve through conventional data analysis.
AI can perform specialized tasks with greater accuracy, which leads to cost reduction and increased productivity.
It provides valuable insights that can significantly enhance the results of clinical trials.
AI can deeply analyze market dynamics and customer behaviours, offering a better understanding of their interactions.
It helps align unmet customer needs with innovative value offerings, both tangible and intangible.
AI improves the performance of antivirus detection systems and aids in the development of new AI algorithms.
AI supports the selection of patients for clinical trials and helps identify issues related to the efficacy and safety of compounds much earlier in the process.
When programmed correctly, AI systems can achieve a low error rate, offering precision, accuracy, and speed that often surpass human capabilities.
Robotic surgery and advanced techniques like robotic radiosurgery can potentially achieve precision beyond human capabilities.
AI is revolutionizing drug discovery by leveraging deep learning and natural language processing to analyze extensive bioscience data.

Disadvantages of AI:

AI lacks human empathy and cannot think independently; it only operates based on predefined programming.
There is concern that AI could negatively influence younger generations.
The potential exists for AI to be misused, possibly leading to widespread destruction on a large scale.
If robots replace humans in various fields, it could result in significant unemployment.
The cost and time involved in building, maintaining, and upgrading AI systems can be substantial.
If AI systems are controlled by malicious actors, they could lead to dangerous outcomes.⁽⁶⁾

Applications of Artificial Intelligence in Pharmaceutical Industry:

AI plays important role in the pharmaceutical field. The use of AI in pharmacy and medicine is given in figure no. 3

Figure no. 3: Applications of AI in Pharmaceutical Industry

AI in Drug Discovery:

Developing new drugs is a complex, costly, and time-intensive process, typically taking 10–15 years. Despite substantial financial investments by the pharmaceutical industry, the pursuit of groundbreaking drugs remains challenging. Only one out of ten potential drug candidates successfully pass phase I clinical trials and gains regulatory approval. AI used in different steps of drug discovery as drug design, chemical synthesis, poly-pharmacology, drug screening, and repurposing. ⁽⁷⁾Figure no. 3 illustrates the diverse applications of Artificial Intelligence (AI) within the drug discovery process. The drug discovery pipeline is categorized into five primary domains where AI significantly contributes:⁽⁸⁾AI is transforming drug discovery and development by enhancing research capabilities and outperforming traditional methods in areas like drug-target modeling and repositioning. In Drug Repositioning, AI speeds up finding new uses for existing drugs, reducing costs and time. AI was pivotal during COVID-19 in assessing antiviral drugs. While in Genomics, AI and ML applications in genomics improve drug development, emphasizing the need for integrated data systems. In Advanced Modeling, Techniques like hypergraph neural networks and bond-to-bond propensity help improve predictions and identify allosteric modulators for safer drugs.

Challenges: The lack of quality data and evolving computational models are key barriers to AI’s full potential. ⁽⁹⁾

Figure no. 4: Applications of Artificial Intelligence (AI) in Drug Discovery.

Artificial intelligence (AI) has transformed the process of small-molecule drug discovery, facilitating tasks such as virtual screening, quantitative structure-activity relationship (QSAR) modeling, and drug design. These tasks are broadly categorized into two main objectives:

predicting molecular properties
generating new molecules.

Small molecules can be represented in multiple ways, including fixed fingerprints, molecular graphs, SMILES strings, and visual formats like images. AI techniques, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), graph neural networks (GNNs), variational autoencoders (VAEs), generative adversarial networks (GANs), normalizing flow models, and transformers, are applied to these representations. Despite these advancements, challenges remain, such as the scarcity of labeled data for molecular property prediction and the complexity involved in generating molecules with specific properties. To address these issues, various learning paradigms have been developed. These include self-supervised learning for utilizing unlabeled data, reinforcement learning to optimize chemical space exploration, and methods like few-shot learning, metric learning, meta-learning, and active learning to improve model performance and adaptability. These innovations are instrumental in enhancing AI's ability to navigate chemical space, thereby expediting drug discovery and the development of new treatments.⁽¹⁰⁾ Figure no:4 shows details about the applications of AI in small-molecule drug discovery.

Figure no. 5: The applications of AI in small-molecule drug discovery⁽¹⁰⁾

AI in Drug formulation and Development:

AI applications helps pharmacists to make accurate and evidence based decision making and in predicting adverse drug events, optimize dosage form, predict drug interactions, preventing medication error, improving adherence, adding telemedicine, and provide medication therapy management services^.⁽¹¹⁾ Before drug formulation, pre-formulation study is important and the pharmaceutical industries are continuously working for the betterment of the formulation in public interest. To tackle obstacles M. Momeni.et.al introduced ML and DL for automated and enhanced Preformulation drug design.⁽¹²⁾ Use of Artificial Intelligence (AI) in pharmaceutical development, including drug discovery, formulation, and personalized medicine. AI improves efficiency by analyzing biological data, optimizing drug candidates, predicting pharmacokinetics and toxicity, and minimizing animal testing. It also enables tailored treatment through patient data analysis, offering great potential to enhance drug development and patient outcomes.⁽¹³⁾

AI in Drug Pre-Formulation Studies:

AI is transforming the drug discovery process by accelerating the identification of potential drug candidates through data analysis and predictive modeling.

Molecular Modeling Software: Tools like Schrödinger Suite and AMBER allow researchers to simulate how molecules interact with biological targets, helping predict the binding efficiency of drug candidates. Avogadro assists in visualizing and optimizing molecular structures to design new compounds.

Machine Learning Algorithms: AI algorithms, such as Random Forests and Neural Networks, are used to analyze biological datasets and predict the pharmacological properties of new drugs. Platforms like TensorFlow and Keras are widely used to train models for predicting a molecule's activity, toxicity, and drug-likeness.^(12,14)

Formulation Development

AI facilitates the optimization of pharmaceutical formulations to ensure the safety, stability, and effectiveness of drug products.

Predictive Analytics in Formulation: AI models help optimize excipient selection, drug solubility, and dissolution rates to enhance formulation performance. Tools like Scikit-learn enable predictive modeling for formulation parameters, ensuring stability and bioavailability.

Simulation Tools: Software such as Hildebrand Solubility Parameters and Flory-Huggins Interaction Models supports the prediction of drug-excipient compatibility and stability.

Neural Networks for Process Optimization: Advanced AI tools like ANNs (Artificial Neural Networks) predict the effect of various formulation parameters on final product quality, helping researchers design better drug delivery systems Neural networks are increasingly used in pharmaceutical formulation to identify optimal combinations of ingredients. They simulate complex interactions, reducing the reliance on physical experiments.

These models are evaluated using performance metrics such as mean squared error and correlation coefficients, ensuring their reliability in practical applications.^(6,15)

Solid Dispersion Formulations:

Solid dispersions enhance the solubility of poorly water-soluble drugs. Machine learning algorithms help optimize these formulations by analyzing the interplay between drug and excipient properties. Predictive tools like regression models and neural networks assist in fine-tuning parameters such as drug release profiles and stability.⁽¹⁶⁾

Crystalline and Amorphous Drug Stability Analysis:

Advanced algorithms analyze and predict the stability of crystalline and amorphous forms of drugs. By assessing factors like amorphization and miscibility, machine learning tools improve formulation stability predictions. Visualization tools and statistical methods further refine these models, enabling precise evaluation of formulation outcomes.^(17–20)

Applications in Tablet and Capsule Design:

AI-driven tools optimize tablet and capsule formulations by analyzing the impact of different variables, such as excipient ratios, pressure, and temperature. Simulation software models manufacturing processes, ensuring consistent product quality and reducing variability in production. AI contributes to improving the physical properties and production efficiency of tablets and capsules.

Design and Simulation: Chemoinformatics platforms like RDK it predict molecular behavior in tablet formulations. Tools such as Minitab and JMP use statistical methods to optimize tablet hardness, disintegration, and dissolution rates.

Optimization Algorithms: AI models help optimize manufacturing processes to produce consistent and high-quality pharmaceutical products.^(21–24)

Microemulsion and Encapsulation Processes:

Machine learning aids in designing microemulsions for delivering hydrophobic drugs effectively. Predictive models analyze the stability and efficiency of these formulations. AI technologies also refine encapsulation processes by predicting particle size, distribution, and release profiles. AI enhances the formulation of microemulsions and encapsulated drug products, ensuring consistent performance.

AI-Based Optimization: Machine learning models optimize particle size and distribution in microemulsions for improved drug release profiles. Tools like Packmol assist in molecular assembly for encapsulation studies.

Predictive Phase Behavior: AI models use experimental data to predict phase stability in microemulsions, reducing trial-and-error experimentation.^(25–28)

Manufacturing:

Figure no.6: AI in Pharmaceutical Manufacturing

Artificial Intelligence (AI) is revolutionizing pharmaceutical manufacturing by enhancing quality control processes and supporting decision-making. By utilizing AI’s data analysis, pattern recognition, and automation capabilities, pharmaceutical companies can ensure the production of safe and effective products. Below are key applications of AI in improving both quality control and decision-making. AI-driven computer vision systems are widely applied for visual inspections in pharmaceutical production. These systems assess images of products, packaging, and labels to detect defects, irregularities, or contamination, ensuring that only products meeting high-quality standards continue through the production pipeline. The AI-powered defect detection process works as follows:

Data Collection: Images of products, packaging, and labels are captured through sensors and cameras placed at various stages of the production process.
Data Preprocessing: The collected images undergo processes like resizing, normalization, and noise reduction to ensure consistency and clarity before analysis.
Training Data Creation: A dataset is curated, where images are labeled to indicate the presence or absence of defects, providing a foundation for training AI models.
Model Training: Convolutional Neural Networks (CNNs) are typically used to train the system to recognize patterns indicative of defects.
Feature Extraction: After training, the AI model identifies relevant features in the images, such as texture, color, and shape, that distinguish defective from non-defective products.
Real-Time Defect Detection: During production, the AI model continuously processes product images, identifying defects based on previously learned features.
Continuous Learning: The model improves over time by learning from new images and defects, reducing errors and improving accuracy.
Manufacturing Line Integration: The AI system is integrated into the production line, offering real-time feedback to prevent the further processing of defective products.

AI enhances predictive quality control by analyzing past production data to forecast whether future batches will meet required quality standards. Tools like regression analysis and random forest models are employed to predict potential quality issues before they arise. This predictive approach reduces the reliance on extensive manual testing, speeding up the release of batches and ensuring better resource utilization. AI, combined with Internet of Things (IoT) sensors, plays a crucial role in monitoring critical production parameters in real time. The system tracks variables such as temperature, pressure, and humidity, and if any parameter deviates from its predefined acceptable range, AI alerts operators to take immediate corrective actions. This prevents defective products from advancing in the production line, maintaining consistent product quality. AI systems can also detect anomalies—unexpected deviations from normal production patterns—by leveraging machine learning models. Detecting these anomalies early allows manufacturers to intervene before any defects or production errors occur. The anomaly detection process typically follows these steps:

Data Collection: Relevant data such as sensor readings and equipment performance metrics are gathered throughout the production process.
Data Preprocessing: The collected data is cleaned and prepared for analysis to ensure accuracy and consistency.
Feature Identification: Features that represent normal system behavior are selected to help the AI model learn what constitutes typical production performance.
Model Selection: AI utilizes various algorithms, including Isolation Forest, Support Vector Machines (SVM), and Autoencoders, for detecting anomalies in the data.
Training the Model: If machine learning algorithms are used, historical data is employed to train the system to recognize what constitutes normal behavior.
Anomaly Detection: Once trained, the model analyzes real-time data to identify any significant deviations from learned behavior and flags them as anomalies.
Threshold Settings: For some models, a threshold is established to define what constitutes an anomaly, helping to identify critical deviations.
Continuous Monitoring: AI continues to monitor production data in real time, issuing alerts when it detects deviations.
Adaptive Learning: Over time, the model adapts to new patterns as the production process evolves.
Model Validation and Fine-Tuning: The AI model is periodically tested against new data to refine its accuracy and improve its detection capability.

AI Tools and Software in Pharmaceutical Manufacturing:

Computer Vision Systems: Used for visual inspection, these systems employ Convolutional Neural Networks (CNNs) for effective image analysis.
Machine Learning Algorithms: Tools such as Random Forests and Support Vector Machines (SVMs) are used for predicting quality outcomes and optimizing manufacturing processes.
IoT Sensors: These sensors work in conjunction with AI to monitor real-time production conditions.
Anomaly Detection Algorithms: AI uses algorithms like Isolation Forest, Autoencoders, and One-Class SVM to detect unusual patterns in production data.

By integrating these AI tools, pharmaceutical manufacturers can significantly enhance their quality control processes, reduce defects, and streamline production while ensuring that only high-quality products are delivered to consumers.^(24,29)

Software and Tools for AI in Pharmaceuticals:

Python Libraries (Scikit-learn, TensorFlow, Keras): Widely used for building and evaluating machine learning models in drug development.

Molecular Modeling Tools: Platforms like Schrödinger provide simulations for understanding drug interactions at a molecular level.

Data Analytics Tools: Programs such as MATLAB and R enable the statistical analysis and visualization of experimental results, ensuring accurate interpretations. The application of artificial intelligence (AI) in predictive maintenance offers transformative potential for the pharmaceutical industry by efficiently analyzing large datasets to uncover valuable insights. By leveraging data analytics and machine learning techniques, AI enables a preventative approach to machine servicing, reducing the likelihood of unexpected downtime and system failures. A key aspect of predictive maintenance involves the use of advanced neural networks, such as Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks, which are particularly effective at analyzing time-series data. Tools like TensorFlow, Keras, and PyTorch facilitate the development and implementation of these models. RNNs and LSTMs are capable of recognizing intricate patterns within sensor data, maintenance histories, and environmental factors. Through continuous monitoring, these models can forecast potential equipment issues or repair requirements, allowing for proactive interventions that minimize disruptions in pharmaceutical manufacturing operations.⁽³⁰⁾

QA & QC:

Artificial intelligence (AI) and machine learning (ML) are reshaping pharmaceutical quality control by improving accuracy, efficiency, and compliance with regulations. It involves roles in enhancing supply chain management, real-time monitoring, predictive maintenance, data integrity, and advanced analytics. AI enables real-time tracking, anomaly detection, and compliance with standards, while ML optimizes spectroscopic analysis and quality assurance processes. Although challenges like data quality and regulatory adherence persist, AI and ML hold vast potential to enhance patient safety, product quality, and operational performance in the pharmaceutical sector.⁽³¹⁾several AI tools and software employed across various domains, including Convolutional Neural Networks (CNNs), specifically the Xception architecture for embryo implantation prediction in assisted reproductive technology (ART). Tools like the EmbryoScope (Vitrolife, Sweden) and the Leica Imaging System are used for time-lapse and high-magnification imaging, respectively, to support embryo quality assessments. In radiotherapy quality assurance, advanced machine learning methods such as Support Vector Classifiers (SVCs), Random Forest Models, and Poisson Regression with Lasso Regularization are utilized to predict gamma index pass rates and multileaf collimator (MLC) errors. Furthermore, FDA-regulated AI algorithms and Deep Learning Models are applied in medical imaging and diagnostics, with specific implementations like AI-based auto-segmentation in radiation therapy and diagnostic tools for detecting intracranial hemorrhages. These tools collectively improve precision, streamline workflows, and enhance quality assurance processes across medical fields.^(32–34)

Product Management:

Artificial intelligence (AI) is reshaping the pharmaceutical sector by transforming essential and supporting business processes. AI's impact on small, medium, and large pharmaceutical companies through interviews. Small firms utilize AI to enhance research and development, data management, reporting, and human resources. Large companies focus on optimizing production, sales, marketing, and analytics through AI. Medium-sized firms adopt AI based on their specific business needs and areas of expertise. The study underscores how AI drives innovation and operational improvements across companies of varying sizes.⁽³⁵⁾

Clinical Trial Design and Monitoring:

Clinical trials are a significant part of the drug development process, consuming up to half of the total time and cost to bring a new drug to market, with failure rates contributing to large financial losses. Challenges such as poor patient selection, ineffective recruitment, and inadequate monitoring of patients during trials result in many trials failing. These issues contribute to the high cost and inefficiency of the process, with fewer drugs making it to the market despite increased investment. Artificial Intelligence (AI) offers solutions to these problems by improving the matching of patients to trials, enhancing recruitment strategies, and providing real-time monitoring of patients. AI technologies like machine learning, natural language processing, and human-machine interfaces can analyze diverse data from electronic health records, medical literature, and trial databases to optimize patient adherence and assess clinical endpoints more efficiently. By improving various stages of clinical trial design and execution, AI can help reduce trial failure rates, lower R&D costs, and increase the success of drug development.⁽³⁶⁾The AI platform was used to monitor real-time medication adherence by confirming drug ingestion, with its accuracy compared to mDOT (medication directly observed therapy) for assessing dosing behavior. Other AI tools included ingestible sensor technology, which accurately tracks drug ingestion, and breathalyzer monitoring, which was employed for assessing adherence through plasma drug concentrations. These technologies offer valuable insights into medication adherence but require further validation in real-world settings to confirm their reliability and usability outside controlled environments.⁽³⁷⁾

CONCLUSION:

It is concluded that, Artificial Intelligence (AI) and Machine Learning (ML) transforming the pharmaceutical industry by streamlining drug development, formulation, and manufacturing. These technologies improve drug discovery, optimize formulations, enhance clinical trial design, and enable real-time monitoring, leading to increased efficiency, accuracy, and cost savings. AI aids in optimizing drug delivery systems, predicting dissolution rates, and improving quality control, all of which contribute to more personalized medicine. Despite challenges like data quality and regulatory hurdles, AI offers great potential to enhance pharmaceutical processes and patient outcomes. Ongoing investment and collaboration will help fully realize AI's capabilities in advancing drug development.

ACKNOWLEDGMENT:

This review paper was independently regarded, researched, and authored by the writers, without any external support. The authors express gratitude to one another for their dedicated collaboration, thoroughly evaluating several drafts to enhance the manuscript.

REFERENCES

Das S, Dey R, Nayak AK. Artificial intelligence in pharmacy. Indian J Pharm Educ Res. 2021;55(2):304–18.
Singh D, Singh R, Gehlot A, Akram SV, Priyadarshi N, Twala B. An Imperative Role of Digitalization in Monitoring Cattle Health for Sustainability. Electron. 2022;11(17):1–14.
- PBW, - SPM, - VLG. Future Directions of AI In Pharma?: Innovation In Pharmaceutical Industry. Int J Multidiscip Res. 2023;5(3):1–11.
Mukhamediev RI, Popova Y, Kuchin Y, Zaitseva E, Kalimoldayev A, Symagulov A, et al. Review of Artificial Intelligence and Machine Learning Technologies: Classification, Restrictions, Opportunities and Challenges. Mathematics. 2022;10(15):1–25.
Jarab AS, Abu Heshmeh SR, Al Meslamani AZ. Artificial intelligence (AI) in pharmacy: an overview of innovations. J Med Econ. 2023;26(1):1261–5.
Tamanna Sharma, Abhinav Mankoo, Vivek Sood. Artificial intelligence in advanced pharmacy. Int J Sci Res Arch. 2021;2(1):047–54.
Ali KA, Mohin S, Mondal P, Goswami S, Ghosh S, Choudhuri S. Influence of artificial intelligence in modern pharmaceutical formulation and drug development. Futur J Pharm Sci [Internet]. 2024;10(1):1–15. Available from: https://doi.org/10.1186/s43094-024-00625-1
Çelik IN, Arslan FK, Tun R, Yildiz I. Artificial Intelligence on Drug Discovery and Development. Ankara Univ Eczac Fak Derg. 2022;46(2):400–27.
Bittner MI, Farajnia S. AI in drug discovery: Applications, opportunities, and challenges. Patterns [Internet]. 2022;3(6):100529. Available from: https://doi.org/10.1016/j.patter.2022.100529
Deng J, Yang Z, Ojima I, Samaras D. Artificial Intelligence in Drug Discovery?: Applications and Techniques arXiv?: 2106 . 05386v4 [ cs . LG ] 2 Nov 2021. :1–65.
Chalasani SH, Syed J, Ramesh M, Patil V, Pramod Kumar TM. Artificial intelligence in the field of pharmacy practice: A literature review. Explor Res Clin Soc Pharm [Internet]. 2023;12(October):100346. Available from: https://doi.org/10.1016/j.rcsop.2023.100346
Momeni M, Afkanpour M, Rakhshani S, Mehrabian A, Tabesh H. A prediction model based on artificial intelligence techniques for disintegration time and hardness of fast disintegrating tablets in pre-formulation tests. BMC Med Inform Decis Mak. 2024;24(1):1–12.
Vora LK, Gholap AD, Jetha K, Thakur RRS, Solanki HK, Chavda VP. Artificial Intelligence in Pharmaceutical Technology and Drug Delivery Design. Vol. 15, Pharmaceutics. 2023. 1–46 p.
Khalid GM, Usman AG. Application of data-intelligence algorithms for modeling the compaction performance of new pharmaceutical excipients. Futur J Pharm Sci. 2021;7(1).
Kapoor DU, Sharma JB, Gandhi SM, Prajapati BG, Thanawuth K, Limmatvapirat S, et al. AI-driven design and optimization of nanoparticle-based drug delivery systems. Sci Eng Heal Stud. 2024;18:1–10.
Kumar B. Solid Dispersion- A Review. 2017;5(2).
Mendyk A, Pac?awski A, Szafraniec-Szcz?sny J, Antosik A, Jamróz W, Paluch M, et al. Data-Driven Modeling of the Bicalutamide Dissolution from Powder Systems. AAPS PharmSciTech. 2020;21(3):1–9.
Kapourani A, Valkanioti V, Kontogiannopoulos KN, Barmpalexis P. Determination of the physical state of a drug in amorphous solid dispersions using artificial neural networks and ATR-FTIR spectroscopy. Int J Pharm X [Internet]. 2020;2:100064. Available from: https://doi.org/10.1016/j.ijpx.2020.100064
Jiang J, Lu A, Ma X, Ouyang D, Williams RO. The applications of machine learning to predict the forming of chemically stable amorphous solid dispersions prepared by hot-melt extrusion. Int J Pharm X [Internet]. 2023;5(January):100164. Available from: https://doi.org/10.1016/j.ijpx.2023.100164
Hussain DZ, Suresh DM, Hemalata B, Shireen S. Articulation Expansion including Artificial Insemination Estimation of Dextromethorphan Tablet by Solid Dispersion Recipe for Solubility Enhancement. Int J Pharm Sci Rev Res. 2022;77(10):53–9.
Kamilov X, Zuparova Z, Abraeva Z, Yusupova S. Development of composition and technology of antidiabetic tablets based on medicinal plants. 2024;01047.
Péterfi O, Nagy ZK, Sipos E, Galata DL. Artificial Intelligence-based Prediction of In Vitro Dissolution Profile of Immediate Release Tablets with Near-infrared and Raman Spectroscopy. Period Polytech Chem Eng. 2023;67(1):18–30.
Khalid MH, Tuszy?ski PK, Kazemi P, Szlek J, Jachowicz R, Mendyk A. Transparent computational intelligence models for pharmaceutical tableting process. Complex Adapt Syst Model. 2016;4(1).
Aksu B, Paradkar A, De Matas M, Özer Ö, Güneri T, York P. Quality by design approach: Application of artificial intelligence techniques of tablets manufactured by direct compression. AAPS PharmSciTech. 2012;13(4):1138–46.
Ranchi M. Application of artificial neural networks in optimizing the fatty alcohol concentration in the formulation of an O / W emulsion. 2011;61:249–56.
Rodríguez-Dorado R, Landín M, Altai A, Russo P, Aquino RP, Del Gaudio P. A novel method for the production of core-shell microparticles by inverse gelation optimized with artificial intelligent tools. Int J Pharm. 2018;538(1–2):97–104.
Aumklad P, Suriyaamporn P, Panomsuk S, Pamornpathomkul B, Opanasopit P. Artificial intelligence-aided rational design and prediction model for progesterone-loaded self-microemulsifying drug delivery system formulations. Sci Eng Heal Stud. 2024;18:1–7.
Agatonovic-kustrin S, Glass BD, Wisch MH, Alany RG. Prediction of a Stable Microemulsion Formulation for the Oral Delivery of a Combination of Antitubercular Drugs Using ANN Methodology Prediction of a Stable Microemulsion Formulation for the Oral Delivery of a Combination of Antitubercular Drugs Using ANN. 2003;(February 2014).
Chandra Saha Associate Professor G, Nasrin Eni L, Saha H, Professor A, Kumar Parida P. Artificial Intelligence in Pharmaceutical Manufacturing: Enhancing Quality Control and Decision Making. Riv Ital di Filos Anal Jr [Internet]. 2023;14(2):2037–4445. Available from: https://rifanalitica.it
Mehta AK, Lanjewar P, Murthy DS, Ghildiyal P, Faldu R, Natrayan L. AI & Lean Management Principles Based Pharmaceutical Manufacturing Processes. 2023 10th IEEE Uttar Pradesh Sect Int Conf Electr Electron Comput Eng UPCON 2023. 2023;10(March 2024):1599–604.
Mundhra S, Kumar Kadiri S, Tiwari P, Student U. Harnessing AI and Machine Learning in Pharmaceutical Quality Assurance. J Pharm Qual Assur Qual Control. 2024;6(January):19–29.
Mahmood U, Shukla-Dave A, Chan HP, Drukker K, Samala RK, Chen Q, et al. Artificial intelligence in medicine: mitigating risks and maximizing benefits via quality assurance, quality control, and acceptance testing. BJR|Artificial Intell. 2024;1(1):1–8.
Simon L, Robert C, Meyer P. Artificial intelligence for quality assurance in radiotherapy. Cancer/Radiotherapie. 2021;25(6–7):623–6.
Cherouveim P, Jiang VS, Kanakasabapathy MK, Thirumalaraju P, Souter I, Dimitriadis I, et al. Quality assurance (QA) for monitoring the performance of assisted reproductive technology (ART) staff using artificial intelligence (AI). J Assist Reprod Genet [Internet]. 2023;40(2):241–9. Available from: https://doi.org/10.1007/s10815-022-02649-z
Kulkov I. The role of artificial intelligence in business transformation: A case of pharmaceutical companies. Technol Soc [Internet]. 2021;66(May):101629. Available from: https://doi.org/10.1016/j.techsoc.2021.101629
Harrer S, Shah P, Antony B, Hu J. Artificial Intelligence for Clinical Trial Design. Trends Pharmacol Sci. 2019;40(8):577–91.
Bain EE, Shafner L, Walling DP, Othman AA, Chuang-Stein C, Hinkle J, et al. Use of a novel artificial intelligence platform on mobile devices to assess dosing compliance in a phase 2 clinical trial in subjects with schizophrenia. JMIR mHealth uHealth. 2017;5(2).