INTRODUCTION:

Artificial intelligence (AI) is defined as a ‘field of science and engineering concerned with the computational understanding of what is commonly called intelligent behaviour, and with the creation of artefacts that exhibit such behaviour’¹. AI is a learning of how human brain thinks and acts, learn, decide and work. It attempts to solve problems and it outputs the intelligent software systems².

AI is simulation of the human intelligence process by computers. The process includes acquiring information, developing rules for using the information, drawing approximate or definite conclusions and self-corrections³.

The British mathematician Alan Turing⁴ (1950) was one of the founders of modern computer science and AI. He defined intelligent behaviour in a computer as the ability to achieve human-level performance in cognitive tasks; this later became popular as the ‘Turing test’. Since the middle of the last century, researchers have explored the potential applications of intelligent techniques in every field¹.

The sprouting idea of adopting AI in the medicine and pharmacy has shifted from hype to hope³.The expression “Medical Technology” is widely used to address a range of tools that can enable health professionals to provide patients and society with a better quality of life by performing early diagnosis, reducing complications, optimizing treatment and/or providing less invasive options, and reducing the length of hospitalization. Intelligent medical technologies (i.e., AI- powered) have been met with enthusiasm by the general population partly because it enables a 4P model of medicine (Predictive, Preventive, Personalized, and Participatory)⁵.

In pharmacy, the task of finding successful new drugs³, clinical trials⁶, drug design⁶ and formulation of pharmaceutical preparation⁶ is daunting, costly and predominantly most difficult. The technologies that incorporate AI have become versatile tools that can be applied ubiquitously in various fields of pharmacy, such as identification and validation of drug targets, designing of new drugs, drug repurposing, improving the R&D efficiency, aggregating and analysing biomedicine information and refining the decision-making process to recruit patients for clinical trials. These potential uses of AI provide the opportunity to counter the inefficiencies and uncertainties that arise while minimizing bias and human intervention in the process³.

Stages of Ai^7-8:

Figure 1. Stages of AI

Types of Ai⁹:

Based on the functionality of Artificial intelligence system, it can be categories into four types:

Figure 2. Types of AI

Sub-Fields of Ai:

AI research has been divided into subfields such as:

· Expert systems

· Natural language processing

· Robotics

· Machine vision

· Speech recognition

· Evolutionary computation

· Machine learning

Figure 3. Sub-fields of AI^10-11

Expert systems (ES)¹²:

Expert systems are branch of AI and were developed by the AI community in the mid-1960s. Expert system is a computer program designed to model the problem solving ability of a human expert. The program models the following characteristics of the human expert: knowledge, Reasoning, Conclusion and Explanations.

The ES system models the knowledge of the human expert both in terms of content and structure. Reasoning is modeled by using procedures and control structures which process the knowledge in a manner similar to the expert. Conclusions given by the system must be consistent with the findings of the human expert. The ES system also provides explanations similar to the human expert. These systems are rule-based system used as a way to store and manipulate knowledge to interpret. A rule consists of two parts: Condition (Antecedent) part and conclusion (action, consequent) part. i.e. IF (condition) THEN (actions).

Natural language processing¹³:

Natural Language processing (NLP)¹⁴ is an branch of computer science, artificial intelligence and linguistics concerned with the interactions between computers and human (natural) language. The field of NLP is deep and diverse. NLP is a collection of techniques used to extract grammatical structure and meaning from input in order to perform a useful task as a result, natural language generation builds output based on the rules of the target language and the task at hand. NLP is useful in the tutoring systems, duplicate detection, computer supported instruction and database interface fields as it provides a pathway for increased interactivity and productivity.¹³

Robotics¹⁵:

Robotics is an interdisciplinary integrative field, at the confluence of several areas, ranging from mechanical and electrical engineering to control theory and computer science, with recent extensions towards material physics, bioengineering or cognitive sciences. The AI-Robotics intersection is very rich. It covers issues such as: deliberate action, planning, acting, monitoring and goal reasoning.

Computer vision¹⁶:

Computer vision is a field of artificial intelligence (AI) that enables computers and systems to derive meaningful information from digital images, videos and other visual inputs — and take actions or make recommendations based on that information. If AI enables computers to think, computer vision enables them to see, observe and understand. Computer vision trains machines to perform these functions, but it has to do it in much less time with cameras, data and algorithms rather than retinas, optic nerves and a visual cortex. Two essential technologies are used to accomplish this: a type of machine learning called deep learning and a convolutional neural network¹⁷ (CNN).

Machine learning uses algorithmic models that enable a computer to teach itself about the context of visual data. If enough data is fed through the model, the computer will “look” at the data and teach itself to tell one image from another. A CNN helps a machine learning or deep learning model “look” by breaking images down into pixels that are given tags or labels.

Speech recognition^18-19:

Speech recognition can be defined as the process of converting speech signal to a sequence of words by means algorithm implemented as a computer program¹⁸.It is a conversion of speech to text in a system. Speech recognition is the machine on the statement or command of human speech to identify and understand and react accordingly. It is based on the voice as the research object, it allows the machine to automatically identify and understand human spoken language through speech signal processing and pattern recognition. Speech recognition is a cross-disciplinary and involves a wide range. It has a very close relationship with acoustics, phenotics, linguistics, information theory, pattern recognition theory and neurobiology disciplines.

The speech recognition system is essentially a pattern recognition system, includes speech signal, feature selection, feature extraction, acoustic models, learning models, modeling, word sequence¹⁹.

Evolutionary computation⁶:

Evolutionary computation is a general concept for several computer techniques based on natural evolution that is mechanism of natural selection and the survival of the most suitable. Therefore evolutionary computation can be used to solve real problems. The most common use of EC in medicine is “Genetic Algorithm”.

Machine Learning:

AI contains subfield called Machine learning²⁰, which uses statistical methods with the ability to learn with or without being explicitly programmed. It is primarily concerned with the design and development of algorithms that allow the system to learn from historical data³.

ML is based on the idea that system can learn from data, identify the patterns and make decision with minimum human intervention²¹.The learning process is done iteratively from analyzed data and new input data. This iterative aspect allows computers to identify hidden insights and repeated patterns and use these findings to adapt when exposed to a new data. The different types of data used in this learning process can vary from observations and examples to instructions and direct experience. The gained knowledge will help in producing reliable and repeated results. Thus, we can describe machine learning as a method that learns from past experiences and uses gained knowledge to do better in the future²².

ML is categorized into supervised, unsupervised, semi-supervised and reinforcement learning³.

Figure 4. Categories of Machine Learning

A. Supervised learning:

Supervised Learning is a machine learning approach where the inputs (features) and outputs (labels or targets) are known. The computer learns the patterns and relationships between features and labels in the training data set and then uses those patterns to predict the labels for previously unseen input features²³. Regression algorithms (continuous output) and classification algorithms²⁴ (discrete output) are considered as the main categories of supervised learning. Regression algorithms attempt to uncover the best function that fits points in the training dataset. Regression algorithms include the following main types: linear regression, multiple linear regression and polynomial regression. Classification algorithms, on the other hand, aim to uncover the best fit class for the input data through assigning each input to its correct class. In this case, the output of the predictive function is in the discrete form and its value is one of the different classes available²².

Figure 5. Supervised learning models

B. Unsupervised learning: Unsupervised Learning is a machine learning approach where the computer learns the patterns and relationships between input variables (features) without knowing the output variables²³. Algorithms included in unsupervised learning can be divided into two main categories, which are: clustering, and dimensionality reduction²².

Figure 6. Unsupervised Learning models:

C. Semi supervised learning: This method falls between the supervised and unsupervised learning methods where we have a large amount of input data, some of which are labeled and the rest are not. Many real life learning problems fall under this area of machine learning. The reason for that is that semi-supervised requires less human intervention since it utilizes very small amount of labeled data and a large amount of unlabeled data²².

D. Reinforcement learning:

Reinforcement learning²⁶ is a machine learning approach where the computer learns to make decisions on its own by making lots of decisions, learning from mistakes, and maximizing the benefit or reward²³.

Deep Learning²⁷:

Deep learning is a subset of machine learning, which is essentially a neural network with three or more layers. These neural networks attempt to simulate the behavior of the human brain—albeit far from matching its ability—allowing it to “learn” from large amounts of data. While a neural network with a single layer can still make approximate predictions, additional hidden layers can help to optimize and refine for accuracy. Deep learning neural networks, or artificial neural networks, attempts to mimic the human brain through a combination of data inputs, weights, and bias. These elements work together to accurately recognize, classify, and describe objects within the data. Deep neural networks consist of multiple layers of interconnected nodes, each building upon the previous layer to refine and optimize the prediction or categorization.

Deep learning models are capable of different types of learning as well, which are usually categorized as supervised learning, unsupervised learning, and reinforcement learning. Supervised learning utilizes labeled datasets to categorize or make predictions; this requires some kind of human intervention to label input data correctly. Common architectures for supervised learning include Deep Neural Network (DNN), Convolutional Neural Network (CNN) and Recurrent neural network (RNN). In contrast, unsupervised learning doesn’t require labeled datasets, and instead, it detects patterns in the data, clustering them by any distinguishing characteristics. The most common methods for unsupervised learning include Deep Auto encoder and Deep Boltzmann machines. Reinforcement learning is a process in which a model learns to become more accurate for performing an action in an environment based on feedback in order to maximize the reward.

Table 1. AI application in medicine

Diagnosis

· A late diagnosis of a disease, which results in delayed treatment and recovery, is fairly prevalent. A large number of people could be saved if we could diagnose an illness before it manifested in a person's body. Machine Learning technique is presently being investigated and used to speed up the diagnosis of numerous diseases, including cancer^{28,
29}.

· AI used in the clinical diagnosis, image analysis in radiology and histopathology, data interpretation in intensive care setting and waveform analysis.

· AI used in diagnosing cytological and histological specimens.

· Breast, gastric, thyroid, oral epithelial cells, urothelial cells, pleural and peritoneal effusion cytology have all been subjected to analysis by neural networks with varying degree of success. In radiology, it is possible to use both human observations and direct digitised images as inputs to the networks. Artificial intelligence’ have been used to interpret plain radiographs, ultrasound, CT, MRI, and radioisotope scans.

· Artificial intelligence pattern recognition ability has been used to analyse various wave forms including the interpretation of ECGs to diagnose myocardial infarction, atrial fibrillation, and ventricular arrythmias. Analysis of electro-enchalograms (EEG) by neural networks has led to its application in the diagnosis of epilepsy and sleep disorders. They have also been trained to analyse electromyographic (EMG) and Doppler ultrasound wave forms as well as haemodynamic patterns in intensive care patients¹.

Prognosis³⁰

· Prognostication is extremely important in planning appropriate treatment strategies and follow-up. Accurate identification of high-risk patients may facilitate targeted aggressive adjuvant therapy which may help cure the disease and prolong survival.

· Artificial intelligence with their ability to exploit non-linear relations between variables are particularly suitable to analyse complex cancer data. It has been demonstrated that neural networks can predict survival in patients with breast and colorectal cancer. Artificial intelligence have also shown to perform better than consultant colorectal surgeons in predicting outcome in patients with colorectal cancer¹.

AI in pharmacy:

Table 2. AI application in pharmacy

Clinical trials³¹	· Clinical trials of drugs are long-lasting and costly, and machine learning has several useful potential applications in helping to organize clinical trials. The application of an advanced, predictive analysis in identifying candidates for clinical trials, finding the best size sample for increased efficiency, adjusting the differences in patient recruitment sites and using electronic medical records to reduce data errors can lead to more efficient and more cost-effective testing. Machine learning can also be used for remote monitoring and access to real-time data for increased security; for example, monitoring biological and other signals for any sign of injury or death of the participants⁶.
Drug design and development	· The way in which AI is used in the design of new drugs is based on monitoring the interaction of the 3D models of molecules and target sites (receptors, enzymes,...) which can represent possible therapy. This is achieved by the application of deep learning based on the existing behavioral history of the molecules⁶.
Epidemiology³²	· By using machine learning and AI, the history of the epidemic can be studied, the activity of social media analyzed, and it can be predicted where and when the epidemic can occur with considerable accuracy⁶.
Formulation of pharmaceutical preparation	· Artificial intelligence have been successfully applied in designing compositions of pharmaceutical preparations, optimizing production processes, providing and controlling quality, predicting the stability of pharmaceutical preparations, in vitro testing the rate of release of the active substance from the pharmaceutical form and in vitro / in vivo correlation⁶.

FDA APPROVED TOOLS³³:

Table 3. AI approved tools

Name of device or algorithm	Description	Mention of AI in announcement
Arterys Cardio DL	Software analyzing cardiovascular images from MR	Deep learning
EnsoSleep	Diagnosis of sleep disorder	Automated algorithm
Arterys Oncology DL	Medical diagnostic application	Deep learning
Idx	Detection of diabetic retinopathy	AI
Koios DS for Breast	Diagnostic software for lesion suspicious for cancer	Machine learning
ContaCT	Stroke detection on CT	AI
OsteoDetect	x-ray wrist fracture diagnosis	Deep learning
Guardian Connect System	Predicting blood glucose changes	AI
EchoMD Automated Ejection Fraction Software	Echocardiogram analysis	Machine learning
DreaMed	Managing type 1 diabetes.	AI

CONCLUSION:

Artificial intelligence (AI) isn't a new concept, but it's becoming increasingly frequent in healthcare as data sets and processing capacity expand. In healthcare, augmented intelligence combines the strengths of computers and physicians to improve patient outcomes by making clinical tasks faster and easier. Artificial intelligence is slowly but steadily becoming an important part of the pharmaceutical sector and the healthcare team. With so much research being done around the world to increase the efficiency of manufacturing and other health-care-related activities, experts are looking into the possibility of using AI in every activity. AI not only improves productivity, but it also reduces errors, which are far more common when a human is in charge of the task. However, if we consider human employment, we must conclude that replacing humans with machines would result in widespread unemployment, and that all functions that were formerly performed by humans will soon be performed by AI. As Stephen Hawking put it, "this could spell the end of humanity." As a result, AI should be used in health care, but it must be designed in such a way that it works in tandem with people.

REFERENCES:

1. Ramesh AN, Kambhampati C, Monson JR, Drew PJ. Artificial intelligence in medicine. Ann R Coll Surg Engl. 2004;86(5):334-338. doi:10.1308/147870804290

2. Sindhu J, Renee Namratha. Impact of Artificial Intelligence in chosen Indian Commercial Bank –A Cost Benefit Analysis. Asian Journal of Management. 2019; 10(4):377-384.

3. Mak KK, Pichika MR. Artificial intelligence in drug development: present status and future prospects. Drug discovery today. 2019 Mar 1;24(3):773-80.doi.org/10.1016/j.drudis.2018.11.014

4. Aayush K., Vishal D., Hammad N., Manu K. S. Application of Artificial Intelligence in Curbing Air Pollution: The Case of India. Asian Journal of Management. 2020;11(3):285-290.

5. Briganti G, Le Moine O. Artificial intelligence in medicine: today and tomorrow. Frontiers in medicine. 2020 Feb 5;7:27. doi: 10.3389/fmed.2020.00027

6. Kostić EJ, Pavlović DA, Živković MD. Applications of artificial intelligence in medicine and pharmacy: Ethical aspects. Acta Medica Medianae. 2019;58(3):128-37. doi:10.5633/amm.2019.0319

7. Wirth N. Hello marketing, what can artificial intelligence help you with? International Journal of Market Research. 2018;60(5):435-438. doi:10.1177/1470785318776841

8. Kaplan A, Haenlein M. Siri, Siri, in my hand: Who’s the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence. Business Horizons. 2019 Jan 1;62(1):15-25. doi.org/10.1016/j.bushor.2018.08.004

9. “Types of Artificial intelligence” December 2021, https://www.forbes.com/sites/cognitiveworld/2019/06/19/7-types-of-artificial-intelligence/?sh=1d2bff64233e

10. Villanueva BM, Salenga ML. Bitter melon crop yield prediction using machine learning algorithm. Int. J. Adv. Comput. Sci. Appl. 2018 Mar 1;9:1-6.

11. Matheny M, Israni ST, Ahmed M, Whicher D. Artificial intelligence in health care: the hope, the hype, the promise, the peril. Washington, DC: National Academy of Medicine. 2019 Dec.

12. Pandit M. Expert system–A review article. International Journal of Engineering Sciences & Research Technology. 2013;2(6):1583-5.

13. Reshamwala A, Mishra D, Pawar P. Review on natural language processing. IRACST Engineering Science and Technology: An International Journal (ESTIJ). 2013;3(1):113-6.

14. Rashmi R., Nirmal Raj VK. A Study on the Implementation and the Impact of Artificial Intelligence in Banking Processes. Asian Journal of Management. 2021; 12(1):47-54.

15. Ingrand F, Ghallab M. Robotics and artificial intelligence: A perspective on deliberation functions. AI communications. 2014 Jan 1;27(1):63-80. DOI: 10.3233/AIC-130578

16. “What is computer vision?” December 2021, https://www.ibm.com/in-en/topics/computer-vision

17. Meenakshi K, Maragatham G. Computational Intelligence in Diagnosis and Prognosis of Gestational Diabetes using Deep Learning. Research J. Pharm. and Tech 2019; 12(8):3891-3895.

18. Gaikwad SK, Gawali BW, Yannawar P. A review on speech recognition technique. International Journal of Computer Applications. 2010 Nov;10(3):16-24.

19. Sharma MK, Kumar O. Speech recognition: A review. International Journal of Advanced Networking and Applications (IJANA). 2014:62-71.

20. Meenakshi K, Safa M, Karthick T, Sivaranjani N. A Novel Study of Machine Learning Algorithms for Classifying Health Care Data. Research J. Pharm. and Tech. 2017; 10(5): 1429-1432.

21. Chahal A, Gulia P. Machine learning and deep learning. International Journal of Innovative Technology and Exploring Engineering. 2019;8(12):4910-4.

22. A. B. Nassif, I. Shahin, I. Attili, M. Azzeh and K. Shaalan, "Speech Recognition Using Deep Neural Networks: A Systematic Review," in IEEE Access, vol. 7, pp. 19143-19165, 2019, doi: 10.1109/ACCESS.2019.2896880.

23. Scott D Nelson, PharmD, MS, Colin G Walsh, MD, MA, Casey A Olsen, PharmD, Andrew J McLaughlin, PharmD, MBA, Joseph R LeGrand, PharmD, MS, Nick Schutz, PharmD, Thomas A Lasko, MD, PhD, Demystifying artificial intelligence in pharmacy, American Journal of Health-System Pharmacy, Volume 77, Issue 19, 1 October 2020, Pages 1556–1570, doi.org/10.1093/ajhp/zxaa218

24. Hossein Niavand, Farzaneh Haghighat Nia. Data Mining, Applications Tools in Insurance Strategies. Asian Journal of Management. 2020; 11(1):38-42.

25. Megha Mishra, Vishnu Kumar Mishra, H.R. Sharma. An Impact of Machine Learning with Lexcio-Syntatics Features of Question Classification. Research J. Engineering and Tech. 3(4): Oct-Dec. 2012 page 327-331.

26. Avinash Dhole, Mohan Awasthy, Sanjay Kumar. Synchronous Q Learning Based Technique for Performance Improvement in Multi core Processors. Int. J. Tech. 2018; 8(2): 86-99.

27. “What is Deep Learning?” january 2022, https://www.ibm.com/cloud/learn/deep-learning

28. Merlin N.J., Shaji Selvin, Sukesh T.N.. Nano Technology –Potential Applications in Medicine. Asian J. Res. Pharm. Sci. 1(2): April-June 2011; Page 31-35.

29. Mohammed Yousef Shaheen. Adoption of machine learning for medical diagnosis. ScienceOpen Preprints. DOI: 10.14293/S2199-1006.1.SOR-.PPHMKA6.v1

30. Mikusev GI, Baikeev RF, Magomedov RO, Mikusev IE, Mishakin TS. Prognosis of Long-term Postoperative Complications after Surgical Treatment of Dupuytren's Disease (Contracture). Research Journal of Pharmacy and Technology. 2019 Mar 1;12(3):1055-65.

31. Patil SM, Sapkale GN, Kumbhar PB, Maske AP. Quality Assurance in Clinical Trial. Research J. Pharmacology and Pharmacodynamics. 2010; 2(1):01-04.

32. Sarfaraz Ahmad, Ambreen Shoaib, Sajid Ali, Sarfaraz Alam, Nawazish Alam, Maksood Ali, Ali Mujtaba, Ayaz Ahmad, Salahuddin Ansari, Mohammad Daud Ali. Epidemiology, risk, myths, pharmacotherapeutic management and socio economic burden due to novel COVID-19: A recent update. Research J. Pharm. and Tech 2020; 13(9):4435-4442.

33. “The Medical futurist” january 2022, https://medicalfuturist.com/fda-approved-ai-based-algorithms/

Received on 16.02.2022 Modified on 30.03.2022

Asian J. Pharm. Tech. 2022; 12(4):401-406.

DOI: 10.52711/2231-5713.2022.00061