Home | Articles
Published on:March 2021
Indian Journal of Pharmaceutical Education and Research, 2021; 55(1s):s149-s156
Original Article | doi:10.5530/ijper.55.1s.45

Investigation of Solubility of Mebendazole Drug using Linear Prediction and Multilayer Feed Forward Neural Network


Authors and affiliation (s):

Chathurappan Raja1, V Sampath Kumar2, Chinnakannu Jayakumar3,*

1Department of Chemical Engineering, National Institute of Technology, Raipur, Chhattisgarh, INDIA.

2Department of Production, RBP Technology (India) Pvt Limited, Chennai, Tamil Nadu, INDIA.

3Department of Applied Science and Technology, AC College of Technology, Anna University, Chennai, Tamil Nadu, INDIA.

Abstract:

Objectives: A systematic in-vitro study has been achieved to decide the stability of a selection of three presidential hydrotropes to decorate the obvious aqueous solubility of the sparingly water- soluble drug, mebendazole drug. This study is that the ANN model to be prediction the solubility of the mebendazole drug among the chemical substances. Methodology: These experimental data, together with a selection of recounted and estimate physico-chemical consequences of the hydrotropes are after that utilizing in-silico to set up a counterfeit neural system (ANN) to engage for desires mebendazole drug solubilization. These trial information, along with a determination of described and gauge physico-chemical outcomes of the hydrotropes are after that using in-silico to set up a counterfeit neural system (ANN) to engage for desires mebendazole tranquilize solubilization. The readied ANN transformed into once determined to exist particularly correct predictions of mebendazole drug solubilization in the presence of hydrotropes and was once for that purpose validated to provide a precious capacity through which hydrotrope sensibility could in like way be screened computationally. The readied ANN transformed into once resolved to exist especially address forecasts of mebendazole tranquilize solubilization within the sight of hydrotropes and was once for that reason approved to give a valuable limit through which hydrotrope sensibility could in like way be screened computationally. The artificial neural network for predicting the solubility properties of the hydrotropic-ester combination was once developed the utilization of MATLAB 2011. For growing the ANN, solubility records that changed into obtained from the experiments had been used. Results: The interest of the hydrotropic will become set as input to the neural network and thus, the precise solubility turn out to be set as target records. The set of rules used to prepare the community modified into the Levenberg Marquardt algorithm. Conclusion: Two hidden layers were expressed. Randomly chosen 80% of the data became used to train the network, It's far is inferred that in-silico screening of drug/hydrotrope structures the utilization of artificial neural networks presents specifically possible to decrease the want for large laboratory testing of those systems and will thus flexibly an economy in expressions of diminished costs and time in tranquilize framework improvement.

Key words: Mebendazole Drug, Solubility, Hydrotropes, Mathematical Model, Artificial Neural Networks.

 




 

Impact Factor

IJPER - An Official Publication of Association of Pharmaceutical Teachers of India is pleased to announce continued growth in the 2019 Release of Journal Citation Reports (source: 2018 Web of Science Data).

 

Impact Factor® as reported in the 2018 Journal Citation Reports® (Clarivate Analytics, 2019): 0.425

The Official Journal of Association of Pharmaceutical Teachers of India (APTI)
(Registered under Registration of Societies Act XXI of 1860 No. 122 of 1966-1967, Lucknow)

Indian Journal of Pharmaceutical Education and Research (IJPER) [ISSN-0019-5464] is the official journal of Association of Pharmaceutical Teachers of India (APTI) and is being published since 1967.

DOI HISTORY

IJPER uses reference linking service using Digital Object Identifiers (DOI) by Crossref. Articles from the year 2013 are being assigned DOIs for its permanent URLs