Intriguing of pharmaceutical product development processes with the help of artificial intelligence and deep/machine learning or artificial neural network

doi:10.21203/rs.3.rs-2112595/v1

Download PDF

Research Article

Intriguing of pharmaceutical product development processes with the help of artificial intelligence and deep/machine learning or artificial neural network

https://doi.org/10.21203/rs.3.rs-2112595/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

The objectives of current review are (1) to provide a historical overview of artificial intelligence and deep/machine learning (AI & D/ML) or Artificial Neural Network (ANN) (2) to update the financial dealings of pharma companies related to the application of AI & D/ML or ANN in drug discovery and development processes and (3) to showcase the application of AI & D/ML or ANN concept for optimization of analytical method conditions and formula of the dosage form. The optimization of analytical method conditions and formula of dosage form started with the employment of linear model such as design of experiment followed by non-linear model like AI & D/ML or ANN. Such type of linear and non-linear models blending in optimization processes nevertheless helped to suitably identify the influence of critical process parameters or critical material attributes on critical quality attributes. However, much of integration and understandable interpretation between the available data arised from clinical trials and the prevalence/progression of pandemic/endemic infections could potentially be ambitioned through the application of AI & D/ML or ANN.

Artificial neural network

artificial intelligence

deep learning

machine learning

linear and non-linear models

Producing pharmaceutical products without untoward effects but with desired qualities is not only the very basic requirement set by regulatory authorities but also indirectly decides the success of the pharma industry. The pharmaceutical product development process is a combined and coordinated work from numerous divisions of the pharma industry which usually starts from API discovery, synthesis and utilization to end with formulation development, market positioning and successful use at the consumer end.

Conventional API discovery and development process as well as optimization of analytical method and formula of dosage form routinely use the quality by testing (QbT) approach or one factor at a time (OFAT) strategy. Since the time consuming and chemical wastage in OFAT strategy or QbT approach is inevitable, the looking for a better alternative approach or strategy which shows less time consumption in conjunction with minimal chemical wastage becomes an urgent requirement. In terms of the response surface methodology (RSM) approach, the design of experiment (DoE) is being utilized effectively to optimize APIs synthetic process, analytical method development and formula optimization for final pharmaceutical products (Rahman et al., 2021; Rahman et al., 2020). It should be added that the RSM-linked DoEis based on a linear model and therefore it won’t consider the non-linear modelization concept to see the influence of independent factors on the response variables. Thus, depending solely on the DoEfor such optimization processes may end with an erroneous conclusion and thus necessitates applying another approach (preferably of non-linear model based) which judiciously eliminates the conclusion errors noticed with the DoE-based optimization process. One such non-linear model-based approach recently introduced in pharmaceutical product development process is artificial intelligence and deep/machine learning (AI & D/ML) or artificial neural network (ANN). Since the information processing capacity of ANN is related to the functioning of the normal human brain, the estimation of the process parameters is being carried out on the number of trials performed by varying the composition of the excipients and processing conditions (Ghate et al., 2019). AI is a division of computer science, involved in problem resolution and in creating machines that can perform tasks which would otherwise require intelligence and human operators (Senthuraman, 2020). In simple words, AI & D/ML is a branch of computer science that deals with problem-solving through the aid of symbolic programming (Krishnaveni et al. 2019). The DL consists of a neural network of multiple layers that aim to emulate how the information processing is carried out by the human brain after understanding complicated patterns and feature interactions. In practical situations, deep learning helps to give structure to unstructured data and enables machines to learn to classify data without assistance (Abhinav & Subrahmanyam, 2019; Gilvary et al. 2019).The ML is a subset of AI utilizing algorithm models and uses statistical methods with the ability to learn with or without being explicitly programmed (Abhinav & Subrahmanyam, 2019).It should be added that the AI & D/ML or ANN is based on a non-linear concept to study the independent factor’s influence on response variables. Non-linearity refers to a massive parallel network distributed throughout that allows for approximation and real-time operation to exhibit unpredictability and random behaviour. Figure 1 shows the possible areas of AI & D/ML or ANN in the pharmacy field of the healthcare system.

[Insert Fig. 1 here]

There have been several excellent articles in both reviews in general (Cavasotto & Di Filippo, 2021; Chaki et al., 2020; Damiati et al., 2020; Paul et al. 2021, Zhou et al. 2020) and an entire issue of a journal devoted solely to the subject of AI & D/ML or ANN (Artificial Intelliegence). Furthermore, as one of the non-linear models for pharmaceutical product optimization, the AI & D/ML or ANN is either combined with a linear model like RSM (Ghate et al., 2019; El Menshawe et al., 2014; Naveen et al., 2020) or only ANN-supported formula optimization of dosage form (Moghaddam et al., 2010; Manda et al., 2019; Lefnaoui et al., 2020) and food products (Rakhavan et al., 2016; Dash & Das 2021; Samson et al., 2016). Similarly, the optimization of analytical method conditions for APIs are also articulated through the AI & D/ML or ANN (Arabzadeh et al., 2019) with support from RSM (Rahman et al., 2021). The objectives of this “meta-review” are (1) to provide a historical overview of AI & D/ML or ANN, (2) to update the financial dealings of pharma companies related to the application of AI & D/ML or ANN in drug discovery and development processes and (3) to showcase the application of in AI & D/ML or ANN concept for optimization of analytical method conditions and formula of the dosage form.

Historical overview on AI & D/ML or ANN

Table 1 delineatesthe historical overview on AI & D/ML or ANN in ascending order of milestone years. The ANN has a history that dates to the precomputer era although the original goal of AI & D/ML or ANN is to solve problems mostly related to biology, the extrapolation of its applicability is much more in recent years and even the intriguing nature of AI & D/ML or ANN especially in medical diagnosis principles as well as in forecasting/predicting the disease pattern in a particular region or whole world. Because firstly the AI & D/ML or ANN helps us understand the impact of increasing/decreasing disease progression vertically or horizontally on computational time. Secondly, the AI & D/ML or ANN helps us understand the situations or cases where the model fits best. Thirdly, it also explains why the certain model works better in certain environment or situation. Samson et al., (2016) described the ANN as an information-processing paradigm that is related to biological nervous systems i.e., the human brain. In the early 1940s (McCulloch & Pitts, 1943), researchers developed the “threshold logic” model, which encompassed a two-pronged approach to computational models of ANN. This model is directed at identifying the biological neural networks separate from ascertaining the correlation of the neural networks to artificial intelligence. An unsupervised learning model utilizing neural plasticity and long-term potentiation has been introduced in the form of Hebbian learning (Morris, 1999). This type of learning was utilized in calculators and other computational instruments (Rochester et al., 1956; Farely & Clark, 1954). Development of the perception in the model added another dimension to AI & D/ML or ANN by incorporating a two-layer computer network in pattern recognition through an algorithm (Rosenblatt, 1958). Particularly, the ANNs comprise a set of nodes, each of which receives a separate input, which is finally converted to output. After the introduction of a clear definition of backpropagation (Werbos, 1975), the supervised learning method in which the ANN receives training in conjunction with optimization becomes easier to determine the loss of function during the anticipated output for each input value is known. In this way, the ANNs are linked to single or multiple algorithms to solve the problems (Paul et al., 2021).Although the AI & D/ML were coined around in the 1950s, it now becomes a slogan in pharma industries especially after finding out their benefits in handling increased volumes of raw data following the introduction of advanced algorithms (Sharma, 2019).

[Insert Table 1 here]

Pharma companies’ financial dealings related to the application of AI & D/ML or ANN in drug discovery and development processes

Table 2 displays the financial dealings of Pharma companies concerning the application part of AI & D/ML or ANN, particularly in drug discovery and development processes. The entry of AI & D/ML or ANN helps to shorten not only the new drug development period but also it significantly minimizes the utilization of manpower and considerable reduction in expenditure related to the API development. For example, the German-based biotechnology company, Evotec, has partnered with a UK-based company, Exscientia, for the small molecule drug discovery process. Within a short period of 8 months, the discovered small drug molecule entered Phase 1 clinical trials which might usually have taken 4-5 years to deliver the drug candidate from the traditional drug discovery process (without utilizing AI & D/ML or ANN).

[Insert Table 2 here]

AI & D/ML or ANN in optimization of analytical method conditions and formula of dosage form

Before entering the discussion related to optimization of analytical method conditions and formula of dosage form,it needs to be emphasized that the ANN simply mimics the principles of information processing handled by the human brain wherein the influence of critical material attributes variation on critical analytical attributes (CAAs) can be predicted by segregating different sets of data (generated from numerous trails) into training, testing and validating (Samson et al., 2016). For this purpose, the ANN must be coupled with an algorithm to attain the “best fit” optimum values for a method (Ghaheri et al., 2015). The ANN-linked algorithm produces a highly reliable and better predictor of the optimum values for a method than the RSM-based linear model (Sha & Edwards, 2007). However, the ANN relies on the number of experiments/trials conducted and consequently, it is highly likely that too high/a smaller number of trials would result in error and fault in the predictions (Ghate et al., 2019).Therefore, the ANN takes the trials of the linear model (RSM following face-centred central composite design (CCD) for generating the non-linear model [ANN-linked Levenberg-Marquardt (LM) algorithm] to predict the optimum regions for the studied CAAs (Rahman et al., 2021).Furthermore, the ANN-linked LM is a potent chemometrics method because of its high performance and good prediction for non-linear systems (Ghaedi, 2015). The typical network architecture of AI & D/ML or ANN is organized in three-different layers, viz., one input, one output and one or more hidden layers. Figure 2 portrays the schematic architecture of AI & D/ML or ANN having three input, ten hidden and three output layers (3:10:3). The architectural structure of AI & D/ML or ANN is the most common multi-layered perceptron (MLP) type which is built on four different elements, input, hidden and output layers along with connections or weights. Interestingly, the MLP type AI & D/ML or ANN works in two phases, training and testing. The training phase is based on the iterative demonstration of the available data pattern to teach the AI & D/ML or ANN for accomplishing the designated assignment. Figure 3 shows the various steps involved in developing the neural network. Other frequently used neuronal network combinations are the kohonen network, convolutional neural network (CNN) and recurrent neural network (RNN).

[Insert Figs. 2 & 3 here]

Figure 4 depicts the possible way to integrate AI & D/ML or ANN in the optimization of analytical method conditions and formula of the dosage form. It can be seen from Figure 4 that the AI & D/ML or ANN needs to be integrated with the DoE approach for optimizing conditions for analytical method and formula for the dosage form. This type of integration between AI & D/ML or ANN and DoE allows the coining of new terminology called, “double-stage systematic optimization”. The double-stage systematic optimization was therefore started initially by using the conventional DoEapproach and then by the application of AI & D/ML or ANN. For instance, Rahman et al., (2021) have used the RSM generated from face-centered CCD of DoE while the ANN is linked with the LM algorithm of AI & D/ML or ANN. Table 3 displays selected non-comprehensive publications showing the involvement of AI & D/ML or ANN in the optimization of analytical method conditions and the formula of the dosage form.

[Insert Figure 4 & Table 3 here]

Because of AI & D/ML or ANN’s advantage in dealing with complex and unstructured data, it is well suited for addressing a wide range of applications in the pharmaceutical sciences and easing up the process (Simões et al., 2020). Table 4 displays the various algorithms usage coupled with AI & D/ML or ANN in different pharmaceutical product development processes.

[Insert Table 4 here]

Representing the drug release process by using computationally simple empirical models is a challenging task since there are complicated interactions between formulation and processing variables. The effort in the pre-prescription step would be considerably reduced if AI & D/ML or ANN model could forecast drug release, and the accuracy of the predictions has been proven. Nagy et al., (2019) used the near-infrared (NIR) and Raman spectra to compare four three-layer ANN models to the standard partial least square (PLS) regression to predict the dissolution profile of extended-release anhydrous caffeine tablets. Brahima et al., (2017) used an MLP for the modelling of riboflavin release behaviour from poly(NIPA-co-AAc) hydrogels. The results showed that the function of ANN was validated, and when compared to the RSM using the mean square error (MSE), the ANN was more appropriate for predicting the release of riboflavin hydrogels and had great generality over the release behaviour of hydrogel. Additionally, Elman neural networks (ENNs) and other dynamic neural networks can also be used to forecast dissolution profiles. Petrovic et al., (2012) used both an ENN and an MLP to characterise the release curve of tablets and determined the wide applicability of ANN. Husseini et al., (2009) and Moussa et al., (2017) used ANN to optimise the ultrasonic release of APIs in preparations (such as liposomes and micelles) to keep therapeutic concentrations constant at specific sites. Han et al., (2018) predicted the disintegrating time of disintegrating oral tablets by neural network techniques. A few selected examples of analytical method condition optimization and formula optimization by integrating the AI & D/ML or ANN concept are narrated below.

SVM in formulation development

Wang et al. (2022) used particle swarm optimization along with the least square support vector machine (PSO-LSSVM) to simplify the optimization process. The results of the prediction model and Taguchi design were compared with PSO-LSSVM. Additionally, this model provided lower costs and a more efficient design of pharmaceutical formulation.

SVM in analytical method development

Keyvan et al., (2021)suggested UV spectrophotometric method development using feed-forward artificial neural network (FFNN) and least square support vector machine (LS-SVM) to simultaneously investigate Sofosbuvir and Daclatasvir in tablet production and biological fluid. Results indicated that the technique has a high potential for predicting component concentrations in dosage forms with a shorter analysis time.

GA in formulation development

Kumar and Kumar (2019) used integral hybrid GA with BPANN and RSM based on a central composite design considering water fraction, surfactant fraction, powder density and ultrasonication time as analysing parameters. Results indicated that the multi-objective hybrid GA model was successful in establishing robust results compared to the conventional method.

GA in analytical method development

In the study by Attia et al., (2021)GA-ANN was used to quantitatively analyse the UV absorption of velpatasvir and sofosbuvir which revealed some overlap, indicating difficulty in the simultaneous estimation of two drugs. GA-ANN proved to be effective in estimating drugs, with acceptable values of root mean square errors for calibration and prediction.

Autoencoder in analytical method development

Kensert et al. (2021)developed a deep one-dimensional convolutional autoencoder that simultaneously eliminates baseline noise and baseline drift to detect and quantify analytes in a mixture of chromatograms with high number and diversity surpassing the approaches like Savitzky-Golay smoothing, Gaussian smoothing and wavelet smoothing.

AI & D/ML or ANN in optimizing eutectic solvent system

The use of AI & ML or ANN to predict and select solvent systems is a very interesting integration of academia and industry. AI & D/ML and ANN can assist in optimizing eutectic systems by designing a solvent system based on appropriate properties. By involving AI & D/ML or ANN-based algorithms, the eutectic solvent system may be chosen by automatically separating the products from the reaction solution (self-precipitation) (Amar et al., 2019; Von Lilienfeld, 2018).

The DoE (linear model)-supported AI & D/ML or ANN concept is currently being established to optimize not only the analytical method conditions for single or dual drug quantification but also the formula development for dosage form. On the other hand, additional ways are also identified in recent years to omit the linear model support for AI & D/ML or ANN for intriguing of pharmaceutical product development processes. To go ahead directly with non-linear models, the employment of multiple-input-multiple-output and multiple-input-single output-based ANN architecture, self-organizing map, MLP network trained with backpropagation algorithm, three- or multi-layer feed-forward ANN, etc. are being proposed currently by Yun et al., (2020) and Ramakrishna et al., (2014)Such kind of direct involvement of AI & D/ML or ANN will speed up the production performance as well as reduce human errors leading to improve the quality of pharmaceutical products.

APIs- active pharmaceutical ingredients

AAE- adversarial autoencoder

AI & D/ML- artificial intelligence and deep/machine learning

AI-DSS- artificial intelligence-based decision support system

ANN- artificial neural network

AE- autoencoder

BPANN- back propagation artificial neural network

BLIS- bacteriocin-like inhibitory substances

CCD- central composite design

Chk 1- Checkpoint kinase 1

CGB- comprehensive grobner bases

CWT- continuous wavelet transform

CAAs- critical analytical attributes

CS- cuckoo search

DBN- deep belief network

DoE- design of experiment

dbCICA- docking-based comparative intermolecular contacts analysis

DruGAN- drug generative adversarial network

ENNs- elman neural network

FFNN- feedforward neural network

GA- genetic algorithm

IPO- initial public offering

kNN- k nearest neighbour

LS-SVM- least square support vector machine

LM- Levenberg-Marquardt

LC-MS/MS- liquid chromatography-mass spectrometry/mass spectrometry

LSTM- long short-term memory

MSE- mean square error

MLFN-12- multilayer feed forward neural network model

MLP- multilayer perceptron

MLR- multilinear regression

NB- naive bayes

NCE- new chemical entity

NIR- near infrared

OFAT- one-factor-at-a-time

ODT- oral disintegrating tablet

PLS- partial least square

PSO- particle swarm optimization

PBD- Plackett-Burman design

PAT- process analytical technology

QbT- quality by testing

RF- random forrest

RNN- recurrent neural network

RSD- relative standard deviation

RSM- response surface methodology

RBM- restricted Boltzmann machine

RP-HPLC- reverse phase-high performance liquid chromatography

SNEDDS- self-nanoemulsifying drug delivery system

SMILES- simplified molecular-input line-entry system

CSTR- single continuous-stirred tank reactor

SNARC- stochastic neural analogue reinforcement computer

SVM- support vector machine

Acknowledgements

The funding was provided by Department of Pharmaceuticals, Ministry of Chemicals and Fertilizers, Govt. of India to the authors Naitik Jariwala, Chandra Lekha Putta, Ketki Gatade, Manasi Umarji, Syed Nazrin Ruhina Rahman, Amoolya Sree, Atul Sayaji Kamble & Abhinab Goswami and by National Mission on Himalayan Studies (NMHS), Ministry of Environment, Forest & Climate Change (MoEF & CC) Nodal and Serving hub with G.B. Pant National Institute of Himalayan Environment & Sustainable Development, Govt. of India to Datta Maroti Pawde are greatly acknowledged.

CRediT (Contributor Roles Taxonomy) author statement

Naitik Jariwala: Conceptualization, Methodology, Software; Chandra Lekha Putta: Conceptualization, Methodology, Software; Ketki Gatade: Conceptualization, Methodology, Software; Manasi Umarji: Conceptualization, Methodology, Software; Syed Nazrin Ruhina Rahman: Data curation, Writing- Original draft preparation; Datta Maroti Pawde: Project administration; Amoolya Sree: Visualization, Investigation; Atul Sayaji Kamble: Visualization, Investigation; Abhinab Goswami: Software, Validation; Tamilvanan Shunmugaperumal: Writing- Reviewing and Editing, Supervision.

Conflict of Interest

The authors declare that they have no competing interest.

Funding Statement

There is no funding source.

Rahman SNR, Katari O, Pawde DM, Boddeda GSB, Goswami A, Mutheneni SR, Shunmugaperumal T (2021) Application of design of experiments^® approach-driven artificial intelligence and machine learning for systematic optimization of reverse phase high performance liquid chromatography method to analyze simultaneously two drugs (cyclosporin A and etodolac) in solution, human plasma, nanocapsules, and emulsions. AAPS PharmSciTech 22(4):1–22. https://doi.org/10.1208/s12249-021-02026-6
Rahman SNR, Pawde DM, Katari O, Hmingthansanga V, Shunmugaperumal T (2020) Systematic optimization, in vitro drug release, and preliminary nonclinical toxicity assessment of nonphospholipid-based topical ophthalmic emulsions containing 0.05 or 0.1% w/w cyclosporin A for dry-eye syndrome management. AAPS PharmSciTech 21(2):1–15. https://doi.org/10.1208/s12249-019-1587-5
Ghate VM, Kodoth AK, Raja S, Vishalakshi B, Lewis SA (2019) Development of MART for the rapid production of nanostructured lipid carriers loaded with all-trans retinoic acid for dermal delivery. AAPS PharmSciTech 20(4):1–16. https://doi.org/10.1208/s12249-019-1307-1
Sethuraman N (2020) Artificial intelligence: a new paradigm for pharmaceutical applications in formulations development. Indian J Pharm Educ Res 54(4):843–846. https://doi.org/10.5530/ijper.54.4.176
Krishnaveni C, Arvapalli S, Sharma JVC (2019) Artificial intelligence in pharma industry- A review. Int J Innovative Pharm Sci Res 7(10):37–50. https://doi.org/10.21276/IJIPSR.2019.07.10.506
Abhinav GVKS, Subrahmanyam SN (2019) Artificial intelligence in healthcare. J Drug Delivery Ther 9(5–s):164–166. https://doi.org/10.22270/jddt.v9i5-s.3634
Gilvary C, Madhukar N, Elkhader J, Elemento O (2019) The missing pieces of artificial intelligence in medicine. Trends Pharmacol Sci 40(8):555–564. https://doi.org/10.1016/j.tips.2019.06.001
Cavasotto CN, Di Filippo JI (2021) Artificial intelligence in the early stages of drug discovery. Arch Biochem Biophys 698:108730. https://doi.org/10.1016/j.abb.2020.108730
Chaki J, Ganesh ST, Cidham SK, Theertan SA (2020) Machine learning and artificial intelligence based diabetes mellitus detection and self-management: a systematic review. J King Saud University-Computer Inform Sci 34(6B):3204–3225. https://doi.org/10.1016/j.jksuci.2020.06.013
Damiati SA (2020) Digital pharmaceutical sciences. AAPS PharmSciTech 21(6):1–12. https://doi.org/10.1208/s12249-020-01747-4
Paul D, Sanap G, Shenoy S, Kalyane D, Kalia K, Tekade RK (2021) Artificial intelligence in drug discovery and development. Drug Discovery Today 26(1):80–93. https://dx.doi.org/10.1016%2Fj.drudis.2020.10.010
Zhou Y, Wang F, Tang J, Nussinov R, Cheng F (2020) Artificial intelligence in COVID-19 drug repurposing. Lancet Digit Health 2(12):e667–e676. https://doi.org/10.1016/s2589-7500(20)30192-8
Journal issue : Artificial Intelligence, An International Journal, ISSN: 0004-3702
El Menshawe SF, Aboud HM, Elkomy MH, Kharshoum RM, Abdeltwab AM (2020) A novel nanogel loaded with chitosan decorated bilosomes for transdermal delivery of terbutaline sulfate: Artificial neural network optimization, in vitro characterization and in vivo evaluation. Drug Delivery and Translational Research 10(2):471–485. https://doi.org/10.1007/s13346-019-00688-1
Naveen NR, Kurakula M, Gowthami B(2020) Process optimization by response surface methodology for preparation and evaluation of methotrexate loaded chitosan nanoparticles. Materials Today: Proceedings, 33, 2716–2724. https://doi.org/10.1016/j.matpr.2020.01.491
Moghaddam MG, Ahmad FBH, Basri M, Rahman MBA (2010) Artificial neural network modeling studies to predict the yield of enzymatic synthesis of betulinic acid ester. Electron J Biotechnol 13(3):3–4. http://dx.doi.org/10.2225/vol13-issue3-fulltext-9
Manda A, Walker RB, Khamanga SM (2019) An artificial neural network approach to predict the effects of formulation and process variables on prednisone release from a multipartite system. Pharmaceutics 11(3):1–18. https://doi.org/10.3390/pharmaceutics11030109
Lefnaoui S, Rebouh S, Bouhedda M, Yahoum MM (2020) Artificial neural network for modeling formulation and drug permeation of topical patches containing diclofenac sodium. Drug Delivery and Translational Research 10(1):168–184. https://doi.org/10.1007/s13346-019-00671-w
Rakhavan KR, Sudharsan K, Babuskin S, Sukumar M (2016) Design and characterization of spice fused tamarind starch edible packaging films. LWT-Food Sci Technol 68:642–652. https://doi.org/10.1016/j.lwt.2016.01.004
Dash KK, Das SK (2021) Modeling and optimization of microwave puffing of rice using artificial neural network and genetic algorithm. J Food Process Eng 44(1):1–16. https://doi.org/10.1111/jfpe.13577
Samson S, Basri M, Masoumi F, Malek HRA, Abedi Karjiban R (2016) An artificial neural network based analysis of factors controlling particle size in a virgin coconut oil-based nanoemulsion system containing copper peptide. PLoS ONE 11(7):e0157737. https://doi.org/10.1371/journal.pone.0157737
Arabzadeh V, Sohrabi MR, Goudarzi N, Davallo M (2019) Using artificial neural network and multivariate calibration methods for simultaneous spectrophotometric analysis of Emtricitabine and Tenofovir alafenamide fumarate in pharmaceutical formulation of HIV drug. Spectrochim Acta Part A Mol Biomol Spectrosc 215:266–275. https://doi.org/10.1016/j.saa.2019.02.077
Manikiran SS, Prasanthi NL (2019) Artificial Intelligence: Milestones and Role in Pharma and Healthcare Sector. Pharma times 51:9–56
McCulloch WS, Pitts W (1943) A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys 5(4):115–133. https://doi.org/10.1007/BF02478259
MORRIS RM(1999) Do hebb: The organization of behavior, Wiley: New york; 1949. Brain Research Bulletin, 50(5–6). https://doi.org/10.1016/s0361-9230(99)00182-3N
Rochester N, Holland J, Haibt L, Duda WJIT (1956) Tests on a cell assembly theory of the action of the brain, using a large digital computer. IRE Trans Inform Theory 2(3):80–93. https://doi.org/10.1109/TIT.1956.1056810
Farely BG, Clark WA (1954) Simulation of self-organizing systems by digital computers. IEEE Trans Prof Group Inform Theory PGIT –4:76–84. https://doi.org/10.1109/TIT.1954.1057468
Rosenblatt F (1958) The perceptron: A probabilistic model for information storage and organization in the brain. Psychol Rev 65(6):386–408. https://doi.org/10.1037/H0042519
Werbos PJ(1975) Beyond regression: new tools for prediction and analysis in the behavioral sciences. [Ph.D. thesis]. Harvard University;
Sharma R(2019), May 28 Artificial Intelligence-The future of pharma industry. Express Pharma. Retrieved August 11 2022, from https://www.expresspharma.in/amp/artificial-intelligence-the-future-of-pharma-industry/
Savage N Tapping into the drug discovery potential of AI. Biopharmadealmakers(2021), May 21 Nature. Retrieved August 1 2022, from https://www.nature.com/articles/d43747-021-00045-7
Ghaheri A, Shoar S, Naderan M, Hoseini SS (2015) The applications of genetic algorithms in medicine. Oman Med J 30(6):406–416. https://dx.doi.org/10.5001%2Fomj.2015.82
Sha W, Edwards KL (2007) The use of artificial neural networks in materials science based research. Mater Design 28(6):1747–1752. https://doi.org/10.1016/j.matdes.2007.02.009
Ghaedi A (2015) Simultaneous prediction of the thermodynamic properties of aqueous solution of ethylene glycol monoethyl ether using artificial neural network. J Mol Liq 207:327–333. https://doi.org/10.1016/j.molliq.2015.04.015
Abasi N, Sohrabi MR, Motiee F, Davallo M (2021) The application of artificial neural network and least square support vector machine methods based on spectrophotometry method for the rapid simultaneous estimation of triamcinolone, neomycin, and nystatin in skin ointment formulation. Optik 241:167210. https://doi.org/10.1016/j.ijleo.2021.167210
Sansare S, Duran T, Mohammadiarani H, Goyal M, Yenduri G, Costa A, Xu X, O'Connor T, Burgess D, Chaudhuri B (2021) Artificial neural networks in tandem with molecular descriptors as predictive tools for continuous liposome manufacturing. Int J Pharm 603:120713. https://doi.org/10.1016/j.ijpharm.2021.120713
Mswahili ME, Lee MJ, Martin GL, Kim J, Kim P, Choi GJ, Jeong YS (2021) Cocrystal prediction using machine learning models and descriptors. Appl Sci 11(3):1323. https://doi.org/10.3390/app11031323
Giese SH, Sinn LR, Wegner F, Rappsilber J (2021) Retention time prediction using neural networks increases identifications in crosslinking mass spectrometry. Nat Commun 12(1):1–11. https://doi.org/10.1038/s41467-021-23441-0
Obeid S, Madžarević M, Krkobabić M, Ibrić S (2021) Predicting drug release from diazepam FDM printed tablets using deep learning approach: Influence of process parameters and tablet surface/volume ratio. Int J Pharm 601:120507. https://doi.org/10.1016/j.ijpharm.2021.120507
Xu Y, Chen J, Yang D, Hu Y, Hu X, Jiang B, Ruan Z, Lou H (2021) Development of LC-MS/MS determination method and backpropagation artificial neural networks pharmacokinetic model of febuxostat in healthy subjects. J Clin Pharm Ther 46(2):333–342. https://doi.org/10.1111/jcpt.13285
Vu GTT, Phan NT, Nguyen HT, Nguyen HC, Tran YTH, Pham TB, Nguyen LT, Nguyen HD (2020) Application of the artificial neural network to optimize the formulation of self-nanoemulsifying drug delivery system containing rosuvastatin. J Appl Pharm Sci 10(9):1–11. https://doi.org/10.7324/japs.2020.10901
Abdullahi HU, Usman AG, Abba SI (2020) Modelling the absorbance of a bioactive compound in HPLC method using artificial neural network and multilinear regression methods. Dutse J Pure Appl Sci 6:362–371
Hasan MA, Hasan RO, Alrawi KF (2020) Simultaneous spectrophotometric estimation of atenolol and amlodipine besylate in pharmaceutical formulations. J Global Sci Res 5(12):1031–1045. https://doi.org/10.5281/zenodo.4587635
Simões MF, Silva G, Pinto AC, Fonseca M, Silva NE, Pinto RM, Simões S (2020) Artificial neural networks applied to quality-by-design: From formulation development to clinical outcome. Eur J Pharm Biopharm 152:282–295. http://dx.doi.org/10.1016/j.ejpb.2020.05.012
Patel L, Shukla T, Huang X, Ussery DW, Wang S (2020) Machine learning methods in drug discovery. Molecules 25(22):5277. https://doi.org/10.3390/molecules25225277
Lo YC, Rensi SE, Torng W, Altman RB (2018) Machine learning in chemoinformatics and drug discovery. Drug Discovery Today 23(8):1538–1546. https://doi.org/10.1016/j.drudis.2018.05.010
Stephenson N, Shane E, Chase J, Rowland J, Ries D, Justice N, Zhang J, Chan L, Cao R (2019) Survey of machine learning techniques in drug discovery. Curr Drug Metab 20(3):185–193. https://doi.org/10.2174/1389200219666180820112457
Ghasemi F, Mehridehnavi A, Pérez-Garrido A, Pérez-Sánchez H (2018) Neural network and deep-learning algorithms used in QSAR studies: merits and drawbacks. Drug Discovery Today 23(10):1784–1790. https://doi.org/10.1016/j.drudis.2018.06.016
Kadurin A, Nikolenko S, Khrabrov K, Aliper A, Zhavoronkov A (2017) druGAN: an advanced generative adversarial autoencoder model for de novo generation of new molecules with desired molecular properties in silico. Mol Pharm 14(9):3098–3104. https://doi.org/10.1021/acs.molpharmaceut.7b00346
Ferreira LL, Andricopulo AD (2019) ADMET modeling approaches in drug discovery. Drug Discovery Today 24(5):1157–1165. https://doi.org/10.1016/j.drudis.2019.03.015
Jaradat NJ, Khanfar MA, Habash M, Taha MO (2015) Combining docking-based comparative intermolecular contacts analysis and k-nearest neighbor correlation for the discovery of new check point kinase 1 inhibitors. J Comput Aided Mol Des 29(6):561–581. https://doi.org/10.1007/s10822-015-9848-1
Lee K, Lee M, Kim D (2017) Utilizing random Forest QSAR models with optimized parameters for target identification and its application to target-fishing server. BMC Bioinformatics 18(16):75–86. https://doi.org/10.1186/s12859-017-1960-x
Ghodmare P, Jibhakate M, Meshram N, Bele V, Ghire P (2018) A review paper on brief introduction of genetic algorithm. Int J Sci Technol Eng 4(8):42–44
Spiegel JO, Durrant JD (2020) AutoGrow4: an open-source genetic algorithm for de novo drug design and lead optimization. J Cheminform 12(1):1–16. https://doi.org/10.1186/s13321-020-00429-4
Hooshmand SA, Jamalkandi SA, Alavi SM, Masoudi-Nejad A (2021) Distinguishing drug/non-drug-like small molecules in drug discovery using deep belief network. Mol Diversity 25(2):827–838. https://doi.org/10.1007/s11030-020-10065-7
Zhang H, Yu P, Ren JX, Li XB, Wang HL, Ding L, Kong WB (2017) Development of novel prediction model for drug-induced mitochondrial toxicity by using naïve Bayes classifier method. Food Chem Toxicol 110:122–129. https://doi.org/10.1016/j.fct.2017.10.021
Ramadasan D, Chevaldonné M, Chateau T (2017) LMA: A generic and efficient implementation of the Levenberg-Marquardt Algorithm. Software: Pract Experience 47(11):1707–1727. https://doi.org/10.1002/spe.2497
Valizadeh M, Sohrabi MR, Motiee F (2021) Simple spectrophotometric method for simultaneous determination of salmeterol and fluticasone as anti-asthma drugs in inhalation spray based on artificial neural network and support vector regression. Optik 240:166879. https://doi.org/10.1016/j.ijleo.2021.166879
Wang H, Gu J, Wang S, Saporta G (2019) Spatial partial least squares autoregression: Algorithm and applications. Chemometr Intell Lab Syst 184(184):123–131. https://doi.org/10.1016/j.chemolab.2018.12.001
Tinkov O, Grigorev V, Grigoreva LD (2021) Virtual screening and molecular design of potential SARS-CoV-2 inhibitors. Mosc Univ Chem Bull 76(2):95–113. https://doi.org/10.3103/S0027131421020127
Sherwani F, Ibrahim BSKK, Asad MM (2021) Hybridized classification algorithms for data classification applications: A review. Egypt Inf J 22(2):185–192. https://doi.org/10.1016/j.eij.2020.07.004
Ajdarić J, Ibrić S, Pavlović A, Ignjatović L, Ivković B (2021) Prediction of drug stability using deep learning approach: Case study of esomeprazole 40 mg freeze-dried powder for solution. Pharmaceutics 13(6):829. https://doi.org/10.3390/pharmaceutics13060829
Mareli M, Twala B (2018) An adaptive Cuckoo search algorithm for optimisation. Appl Comput Inf 14(2):107–115. https://doi.org/10.1016/j.aci.2017.09.001
Hu T, Khishe M, Mohammadi M, Parvizi GR, Karim SHT, Rashid TA (2021) Realtime COVID-19 diagnosis from X-Ray images using deep CNN and extreme learning machines stabilized by chimp optimization algorithm. Biomed Signal Process Control 68:102764. https://doi.org/10.1016/j.bspc.2021.102764
Wong WC, Chee E, Li J, Wang X (2018) Recurrent neural network-based model predictive control for continuous pharmaceutical manufacturing. Mathematics 6(11):242. https://doi.org/10.3390/math6110242
Nagy B, Petra D, Galata DL, Démuth B, Borbás E, Marosi G, Nagy ZK, Farkas A (2019) Application of artificial neural networks for Process Analytical Technology-based dissolution testing. Int J Pharm 567:118464. http://dx.doi.org/10.1016/j.ijpharm.2019.118464
Brahima S, Boztepe C, Kunkul A, Yuceer M (2017) Modeling of drug release behavior of pH and temperature sensitive poly (NIPAAm-co-AAc) IPN hydrogels using response surface methodology and artificial neural networks. Mater Sci Engineering: C 75:425–432. http://dx.doi.org/10.1016/j.msec.2017.02.081
Petrović J, Ibrić S, Betz G, Đurić Z (2012) Optimization of matrix tablets controlled drug release using Elman dynamic neural networks and decision trees. Int J Pharm 428(2):57–67. http://dx.doi.org/10.1016/j.ijpharm.2012.02.031
Husseini GA, Mjalli FS, Pitt WG, Abdel-Jabbar NM (2009) Using artificial neural networks and model predictive control to optimize acoustically assisted doxorubicin release from polymeric micelles. Technol Cancer Res Treat 8(6):479–488. http://dx.doi.org/10.1177/153303460900800609
Moussa HG, Husseini GA, Abel-Jabbar N, Ahmad SE (2017) Use of model predictive control and artificial neural networks to optimize the ultrasonic release of a model drug from liposomes. IEEE Trans Nanobiosci 16(3):149–156. http://dx.doi.org/10.1109/tnb.2017.2661322
Han R, Yang Y, Li X, Ouyang D (2018) Predicting oral disintegrating tablet formulations by neural network techniques. Asian J Pharm Sci 13(4):336–342. http://dx.doi.org/10.1016/j.ajps.2018.01.003
Wang S, Yang J, Chen H, Chu K, Yu X, Wei Y, Zhang H, Rui M, Feng C (2022) A Strategy for the Effective Optimization of Pharmaceutical Formulations Based on Parameter-Optimized Support Vector Machine Model. AAPS PharmSciTech 23(1):1–12. http://dx.doi.org/10.1208/s12249-022-02210-2
Keyvan K, Sohrabi MR, Motiee F (2021) An intelligent method based on feed-forward artificial neural network and least square support vector machine for the simultaneous spectrophotometric estimation of anti hepatitis C virus drugs in pharmaceutical formulation and biological fluid. Spectrochim Acta Part A Mol Biomol Spectrosc 263:120190. http://dx.doi.org/10.1016/j.saa.2021.120190
Kumar H, Kumar V (2019) Ultrasound assisted synthesis of water-in-oil nanoemulsions: Parametric optimization using hybrid ANN-GA approach. Chem Eng Processing-Process Intensif 144:107649. http://dx.doi.org/10.1016/j.cep.2019.107649
Attia KA, El-Abasawi NM, El-Olemy A, Abdelazim AH, Goda AI, Shahin M, Zeid AM (2021) Simultaneous spectrophotometric quantitative analysis of velpatasvir and sofosbuvir in recently approved FDA pharmaceutical preparation using artificial neural networks and genetic algorithm artificial neural networks. Spectrochim Acta Part A Mol Biomol Spectrosc 251:119465. http://dx.doi.org/10.1016/j.saa.2021.119465
Kensert A, Collaerts G, Efthymiadis K, Van Broeck P, Desmet G, Cabooter D (2021) Deep convolutional autoencoder for the simultaneous removal of baseline noise and baseline drift in chromatograms. J Chromatogr A 1646:462093. http://dx.doi.org/10.1016/j.chroma.2021.462093
Amar Y, Schweidtmann AM, Deutsch P, Cao L, Lapkin A (2019) Machine learning and molecular descriptors enable rational solvent selection in asymmetric catalysis. Chem Sci 10(27):6697–6706. https://doi.org/10.1039/C9SC01844A
Von Lilienfeld OA (2018) Quantum machine learning in chemical compound space. Angew Chem Int Ed 57(16):4164–4169. https://doi.org/10.1002/anie.201709686
Yun S, Kang JM, Kim IM, Ha J (2020) Deep artificial noise: Deep learning-based precoding optimization for artificial noise scheme. IEEE Trans Veh Technol 69(3):465–3469. https://doi.org/10.1109/TVT.2020.2965959
RamaKrishna K, Ramam VA, Rao RS (2014) Mathematical Neural Network (MaNN) Models Part VI: Single-layer perceptron [SLP] and Multi-layer perceptron [MLP] Neural networks in ChEM-Lab. J Appl Chem 3(6):2209–2311

Table 1 Historical overview on artificial intelligence & deep/machine learning or artificial neural network in ascending order of milestone years [Partially taken from Manikiran & Prasanthi, 2019]

Milestone Years	Landmark of artificial intelligence
1943	Walter Pitts and Warren Mcculloch proved that the neurons can do logical operations like “and”, “or” or “not” when they are connected as a network.
1951	Marvin Minsky introduced the first neuronal network i.e., SNARC (Stochastic Neural Analogue Reinforcement Computer) which can solve the problem from the real world.
1956	The term artificial intelligence was coined at Dartmouth College Conference.
1958	Frank Rosenblatt created perceptrons which are neuronal networking that transmit information in one direction which is the basis of today’s AI progress.
1969	The symbolic representation of the problems is supported by Minsky in his book “perceptrons”.
1974-1980	“First AI winter” period.
1986	Georey Hinton developed back propagation algorithm design which is used in deep learning these days.
1987-1993	“AI winter” phase.
1997	IBM deep blue defeated the Garry Kasparov who is a Russian grandmaster in chess.
2013	British technology is used by the Google to perform efficient research on photos.
2016	Google DeepMind defeated the Go Champion Lee Sedol by the software alpha Go.
2017	Use of AI in analytical method optimization of pharma products.
2018	AI-based formula optimization for pharmaceutical products.
2019	Integration DoE with AI.
2020	AI concept in pharmaceutical unit operations including process analytical technology (PAT).
2021	AI utility in pharmaceutical labeling and package security system.

Table 2 Non-comprehensive financial dealings of pharma companies related to the application of artificial intelligence & deep/machine learning or artificial neural network in drug discovery process [Partially taken from Savage, (2021)]

Companies	Month & Year	Headline
ZebiAI Therapeutics, Relay Therapeutics (Cambridge, USA)	April 2021	Relay Therapeutics buys ZebiAI Therapeutics for $85 million and further $185 million in potential revolutionary payments.
Recursion Pharmaceuticals (Salt Lake City, UT, US)	April 2021	Recursion Pharmaceuticals finalizes $436 million initial public offering (IPO).
Exscientia (Oxford GB)	April 2021	Exscientia gets additional $225 million in a series D round driven by SoftBank Vision Fund 2.
Iktos (Paris, France), Pfizer (New York City, USA)	March 2021	A number of Pfizer’s small-molecule programs are supported by Iktos by its AI-driven de-novo design software.
Exscientia (Oxford GB)	March 2021	Exscientia finalizes $100 million Series C financing, with investors including Evotec, Bristol Myers Squibb and GT Healthcare.
Insitro (South San Francisco, CA, US)	March 2021	Canada Pension Plan Investment Board supported in financing Insitro to raise $400 million in Series C.
Valo Health (Boston, MA, USA)	March 2021	Valo Health, which is emerging its Opal computational drug discovery and development platform, increases $110 million to add to its $190 million upturned in January 2021 for its Series B funding round.
Exscientia (Oxford GB), University of Oxford (England)	February 2021	The University of Oxford and Exscientia collaborate for developing treatments for Alzheimer disease.
Roivant (Basel, New York), Silicon Therapeutics (Boston, USA)	February 2021	Silicon Therapeutics was sold to Roivant for $450 million, including its physics-based platform for insilico small-molecule drug design, to be combined with Roivant’s machine learning approaches.
Cellarity (Cambridge, MA)	February 2021	For the drug discovery approach related to modulating cellular behaviors, Cellarity raised $123 million in Series B funding.
AbCellera (Vancouver, Canada)	December 2020	AbCellera ends its IPO at $556 million.
Genesis Therapeutics (San Francisco, CA, US), Genentech (South San Francisco, CA, US)	October 2020	Alliance between Genesis Therapeutics and Genentech in multi-target drug development via Genesis’ graph machine learning to identify drug candidates for a wide range of disorders.
Insitro (South San Francisco, CA, US), Bristol Myers Squibb (New York, USA)	October 2020	Insitro Human platform, a machine learning technology is used to detect potential drug targets by rising prognostic models of amyotrophic lateral sclerosis and frontotemporal dementia. Then, Bristol Myers Squibb will select candidates to progress further.
Recursion Pharmaceuticals (Salt Lake City, UT, US), Bayer (Leverkusen, Germany)	September 2020	Deal between Bayer partners and Recursion Pharmaceuticals in drug development used Recursion’s AI-guided drug discovery platform for new small-molecule therapies for treating fibrotic diseases causing addition in Series D financing round.
Recursion Pharmaceuticals (Salt Lake City, UT, US)	September 2020	Recursion Pharmaceuticals, apply machine learning to cellular imaging data and raises $239 million in Series D financing round directed by Bayer’s investment department Leaps. Other investors include Baillie Gifford, Casdin Capital, Samsara BioCapital and Lux Capital.
XtalPi (Cambridge, Massachusetts, US)	September 2020	XtalPi is using quantum physics with AI to discover drug candidates and more than a dozen investment companies raised $318 million in Series C round for this start-up.
BioSymetrics (New York City, USA), Sema4 (Stamford, Connecticut, US), Johnson & Johnson (New Jersey, United States)	August 2020	J&J’s Janssen is collaborated with BioSymetrics and Sema4 aiming to apply machine learning to predict onset and severity of COVID-19 among different populations, involved developing new treatments and vaccines.
Atomwise (San Francisco, CA, US)	August 2020	Sanabil Invests around $123 million Series B funding for Atomwise in order to upkeep the advancement of its molecule identification software.
Relay Therapeutics (Cambridge, USA)	July 2020	Relay Therapeutics, focused on protein motion to design drug candidates, calls $400 million initial public offering (IPO).
Insitro (South San Francisco, CA, US)	May 2020	Insitro for its machine learning-based drug discovery approach, raised $143 million in Series B funding.
AbCellera (Vancouver, Canada)	May 2020	AbCellera for the expansion of its antibody drug discovery platform raised $105 million in Series B funding.
Schrödinger (New York City, USA)	February 2020	Bill Gates and David Shaw’s Drug discovery software company ends $232 million initial public offering (IPO).
Exscientia (Oxford GB), Bayer (Leverkusen, Germany)	January 2020	Bayer partnered with Exscientia to discover cardiovascular and oncology drug candidates by their artificial intelligence drug discovery platform i.e. “Centaur Chemist”.
ZebiAI Therapeutics, Google Accelerated Science	January 2020	ZebiAI Therapeutics partnered with Google Accelerated Science for discovering small-molecule drug candidates using machine learning and DNA-encoded library technologies.
Microsoft, Novartis (Basel Switzerland)	October 2019	Novartis announces collaboration with Microsoft to apply their AI algorithms to its large datasets in order to identify and develop therapeutics.
Exscientia (Oxford GB), Rallybio (New Haven, CT)	July 2019	Exscientia collaborates with Rallybio bringing its AI drug discovery platform Centaur Chemist to discover small molecule drugs for rare diseases.
Insitro (South San Francisco, CA, US), Gilead (Foster City, California, USA)	April 2019	Gilead is the first big pharma company to sign 3-year deal with Insitro for applying its Insitro Human platform to identify new drug targets for non-alcoholic steatohepatitis by producing experimental models of the disease. Insitro will be paid $15 million in a deal potentially worth $1 billion.
BenevolentAI (London United Kingdom), AstraZeneca (Cambridge, United Kingdom)	April 2019	AstraZeneca collaborated with BenevolentAI to detect new drug candidates for chronic kidney disease and idiopathic pulmonary fibrosis.

Table 3 Selected non-comprehensive publications showing the involvement of AI & D/ML or ANN in optimization of analytical method conditions and formula of dosage form

Publication title	Short summary of the publication
Application of design of experiments approach-driven artificial intelligence and machine learning for systematic optimization of reverse phase high performance liquid chromatography method to analyze simultaneously two drugs (cyclosporin A and etodolac) in solution, human plasma, nanocapsules and emulsions (Rahman et al., 2021).	- Optimization of the RP-HPLC method conditions using CCD-driven ANN linked with Levenberg-Marquardt (LM) models (Rahman et al., 2021).
The application of artificial neural network and least square support vector machine methods based on spectrophotometry method for the rapid simultaneous estimation of triamcinolone, neomycin and nystatin in skin ointment formulation (Abasi et al., 2021).	- ANN and LS-SVM were employed (Abasi et al., 2021).
Artificial neural networks in tandem with molecular descriptors as predictive tools for continuous liposome manufacturing (Sansare et al., 2021).	- The use of molecular descriptors as inputs in ANN has the potential to reduce predictive error and thus improve prediction accuracy (Sansare et al., 2021).
Cocrystal prediction using machine learning models and descriptors (Mswahili et al., 2021).	- SMILES descriptor values were developed and machine learning models were compared. ANN has the best accuracy, sensitivity and F1 score (Mswahili et al., 2021).
Retention time prediction using neural networks increases identifications in crosslinking mass spectrometry (Giese et al., 2021).	- The xiRT Siamese machine learning model predicts retention time with great accuracy (Giese et al., 2021).
Predicting drug release from diazepam fused deposition modeling printed tablets using deep learning approach: Influence of process parameters and tablet surface/volume ratio (Obeid et al., 2021).	- An established ANN model could predict drug release features (Obeid et al., 2021).
Development of LC-MS/MS determination method and backpropagation artificial neural networks pharmacokinetic model of febuxostat in healthy subjects (Xu et al. 2021).	- Back propagation artificial neural network (BPANN) model for forecasting febuxostat plasma concentrations was constructed (Xu et al. 2021).
Application of the artificial neural network to optimize the formulation of self- nanoemulsifying drug delivery system (SNEDDS) containing rosuvastatin (Vu et al., 2020).	- The authors were able to improve the formulation of rosuvastatin SNEDDS using an artificial neural network (Vu et al., 2020).
Modelling the absorbance of a bioactive compound in HPLC method using artificial neural network and multilinear regression (MLR) methods (Abdullahi et al., 2020).	- The ANN model performed marginally better than the MLR model (Abdullahi et al., 2020).
Simultaneous spectrophotometric estimation of atenolol and amlodipine besylate in pharmaceutical formulations (Hasan et al., 2020).	- ANN, PLS and principal component regression are three multivariate analytic approaches that have been created (Hasan et al., 2020).

Table 4 Various algorithms usage in different pharmaceutical product development processes

Algorithm name	Algorithm functioning	Application in pharma industry
Support Vector Machine (SVM)	According to Patel et al., (2020), the SVMs are managed machine learning algorithms which differentiate compounds into groups based on certain attribute selection by originating a hyperplane. SVM is a statistical method for mapping data into high-dimensional space to identify a lower dimensional hyperplane that maximizes the data separation using a nonlinear kernel (Lo et al., 2018).	Stephenson et al., (2019) utilized the SVM algorithm along with other algorithms to classify resistivity and susceptibility of tuberculosis drugs.
Autoencoder (AE)	AE is utilized layer by layer as an elementary unit for developing deep neural network (Ghasemi et al., 2018).	Kadurin et al., (2017) implemented DruGAN (Drug Generative Adversarial Network) which is set up on AAE (Adversarial Autoencoder) model to recognize novel molecular compounds having anticancer activity.
k Nearest Neighbor (kNN)	Ferreira & Andricopulo, (2019) outlined that kNN as a non-linear correlation optimization method which creates its own understanding from acquired knowledge of data.	Jaradat et al., (2015) applied combination of kNN with dbCICA (Docking-based comparative intermolecular contacts analysis) tool for selection and identification of anticancer drug asassociated with Chk 1 (Checkpoint Kinase 1) inhibitors.
Random Forest (RF)	According to Patel et al., (2020), RF is the most suitable for larger database having several attributes. RF is a collective learning method of various decision tree assembled on basis of different datasets and by using commonly elected program (Lo et al., 2018).	Lee et al., (2017) applied RF based virtual screening method to develop QSAR models with support of ChEMBL datasets.
Genetic algorithm (GA)	GA is an improvised algorithm developed to discover ideal result based on procedure with different steps (Ghodmare et al., 2018).	AutoGrow4: It is based on genetic algorithm and aids in prediction of drug-like compounds and optimizing pre-existing ligands. The process involves seed selection and pooling, then crossing over and filtering of compound and lastly it docks the compound and ranks them based on scoring (Spiegel & Durrant, 2020).
Deep belief network (DBN)	DBN is classified as generative unsupervised learning technique, developed by assembling restricted boltzmann machine (RBM) and is instructed in an acquisitive design (Hooshmand et al., 2021).	DBN algorithm is able to establish a classification system into drug like and non-drug like potential druggable candidates from ZINC datasets to accelerate drug discovery process (Hooshmand et al., 2021).
Naive Bayes (NB)	A probabilistic classification method that practices probability prior and Bayes’ theorem to predict membership by assuming feature independency (Lo et al., 2018).	Zhang et al., (2017) utilized NB classifier model for determining the mitochondrial toxicity caused by drugs.
Levenberg-Marquardt (LM) algorithm	LM algorithm is a very popular nonlinear optimization method (Ramadasan et al., 2017).	Valizadeh et al., (2021) performed simultaneous estimation of anti-asthmatic drugs in inhalation spray using LM.
Partial least square (PLS) algorithm	According to Wang et al., (2019), PLS is an efficient multivariate statistical data analysis method that has been proven to obtain good prediction accuracy which has gain numerous applications in bioinformatics, machine learning and chemometrics.	Tinkov et al., (2021) executed different types of molecular descriptors and machine learning methods by constructing QSAR models using PLS.
Backpropagation artificial neural network (BPANN)	Sherwani et al., (2021) applied the BPANN aimed to expedite the convergence of neural network to global optimum	Ajdarić et al., (2021) predicted stability of esomeprazole 40 mg freeze dried powder for solution using deep learning approach. They trained multilayer perceptron (MLP) network using back propagation network.
Cuckoo search algorithm	According to Mareli and Twala (2018), Cuckoo search is nature-inspired algorithm used extensively to solve optimization problems because it can maintain balance between local and global random walks using switching parameter.	Real-time detection of covid-19 using cuckoo search optimized extreme learning machine was done by Hu et al., (2021) using chimp optimization algorithm.
Recurrent Neural Network (RNN) algorithm	RNNs is a sub-class of ANNs, and they are structured to use hidden variables as a memory for capturing temporal dependencies between system and control variables (Wong et al., 2018).	Wong et al., (2018) demonstrated the use of RNNs for system identification of a complex reaction in a single continuous-stirred tank reactor for pharmaceutical API production.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Intriguing of pharmaceutical product development processes with the help of artificial intelligence and deep/machine learning or artificial neural network

Status:

Version 1

Abstract

Figures

Introduction

Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Status:

Version 1