A Practical Framework Toward Prediction of Breaking Force and Disintegration of Tablet Formulations Using Machine Learning Tools  Ilgaz Akseli, Jingjin.

Slides:



Advertisements
Similar presentations
Appendix B: An Example of Back-propagation algorithm
Advertisements

Back-Propagation MLP Neural Network Optimizer ECE 539 Andrew Beckwith.
Previous Estimates of Mitochondrial DNA Mutation Level Variance Did Not Account for Sampling Error: Comparing the mtDNA Genetic Bottleneck in Mice and.
Functional Analysis of the Neurofibromatosis Type 2 Protein by Means of Disease- Causing Point Mutations Renee P. Stokowski, David R. Cox The American.
An Artificial Neural Network Approach to Surface Waviness Prediction in Surface Finishing Process by Chi Ngo ECE/ME 539 Class Project.
Gene Preference in Maple Syrup Urine Disease Mary M. Nellis, Dean J. Danner The American Journal of Human Genetics Volume 68, Issue 1, Pages (January.
Effects of Milk Powders in Milk Chocolate B. Liang, R.W. Hartel Journal of Dairy Science Volume 87, Issue 1, Pages (January 2004) DOI: /jds.S (04)
Date of download: 6/1/2016 Copyright © ASME. All rights reserved. From: Modeling of Direct Methanol Fuel Cell Using the Artificial Neural Network J. Fuel.
Clearance Prediction of HIV Protease Inhibitors in Man: Role of Hepatic Uptake Tom De Bruyn, Bruno Stieger, Patrick F. Augustijns, Pieter P. Annaert Journal.
Genetic Parameters and Trends in the Chilean Multibreed Dairy Cattle Population* M.A. Elzo, A. Jara, N. Barria Journal of Dairy Science Volume 87, Issue.
Patient request for pharmacist counseling and satisfaction: Automated prescription delivery system versus regular pick-up counter Jan D. Hirsch, PhD Journal.
Prediction of Pharmacokinetics Prior to In Vivo Studies. 1. Mechanism ‐ Based Prediction of Volume of Distribution Patrick Poulin, Frank ‐ Peter Theil.
Combining Models Foundations of Algorithms and Machine Learning (CS60020), IIT KGP, 2017: Indrajit Bhattacharya.
Big data classification using neural network
Hsiaoling Wang, Nagarani Pampati, William M
Pediatric Drugs-A Review of Commercially Available Oral Formulations
Figure 11.1 Linear system model for a signal s[n].
Design, Characterization, and Aerosol Dispersion Performance Modeling of Advanced Co-Spray Dried Antibiotics with Mannitol as Respirable Microparticles/Nanoparticles.
Ranga Rodrigo February 8, 2014
Chronological age prediction based on DNA methylation: Massive parallel sequencing and random forest regression  Jana Naue, Huub C.J. Hoefsloot, Olaf.
The Disposition of Oxymatrine in the Vascularly Perfused Rat Intestine-Liver Preparation and Its Metabolism in Rat Liver Microsomes  Li Hua Huang, Yun Ming.
Date of download: 12/22/2017 Copyright © ASME. All rights reserved.
Formulation of 3D Printed Tablet for Rapid Drug Release by Fused Deposition Modeling: Screening Polymers for Drug Release, Drug-Polymer Miscibility and.
Anna Middleton, J. Hewison, R.F. Mueller 
DNA Base-Calling from a Nanopore Using a Viterbi Algorithm
Self-Nanoemulsifying Lyophilized Tablets for Flash Oral Transmucosal Delivery of Vitamin K: Development and Clinical Evaluation  Khalid M. El-Say, Tarek.
Put a Face to a Name: A Randomized Controlled Trial Evaluating the Impact of Providing Clinician Photographs on Inpatients' Recall  Lora Appel, PhDc,
A Promising New Method to Estimate Drug-Polymer Solubility at Room Temperature  Matthias Manne Knopp, Natasha Gannon, Ilona Porsch, Malte Bille Rask, Niels.
Prof. Carolina Ruiz Department of Computer Science
A systematic review and meta-analysis of the accuracy of weight estimation systems used in paediatric emergency care in developing countries  Mike Wells,
Volume 10, Issue 6, Pages (June 2018)
Anna Middleton, J. Hewison, R.F. Mueller 
Neural Networks Advantages Criticism
Video Games for Neuro-Cognitive Optimization
Ian S. Howard, Daniel M. Wolpert, David W. Franklin  Current Biology 
network of simple neuron-like computing elements
Ian S. Howard, Daniel M. Wolpert, David W. Franklin  Current Biology 
Emre O. Neftci  iScience  Volume 5, Pages (July 2018) DOI: /j.isci
How to evaluate and predict the ecologic impact of antibiotics: the pharmaceutical industry view from research and development  R. Bax  Clinical Microbiology.
Generating Coherent Patterns of Activity from Chaotic Neural Networks
The Big Health Data–Intelligent Machine Paradox
Volume 87, Issue 1, Pages (July 2015)
Thomas Willems, Melissa Gymrek, G
Yukiyasu Kamitani, Frank Tong  Current Biology 
Confidence as Bayesian Probability: From Neural Origins to Behavior
Γ-TEMPy: Simultaneous Fitting of Components in 3D-EM Maps of Their Assembly Using a Genetic Algorithm  Arun Prasad Pandurangan, Daven Vasishtan, Frank.
Prediction of in-hospital mortality after ruptured abdominal aortic aneurysm repair using an artificial neural network  Eric S. Wise, MD, Kyle M. Hocking,
Volume 66, Issue 1, Pages (July 2004)
A Flexible Bayesian Framework for Modeling Haplotype Association with Disease, Allowing for Dominance Effects of the Underlying Causative Variants  Andrew.
Model generalization Brief summary of methods
Xing Hua, Haiming Xu, Yaning Yang, Jun Zhu, Pengyuan Liu, Yan Lu 
Sparseness and Expansion in Sensory Representations
TensorFlow: Biology’s Gateway to Deep Learning?
Volume 107, Issue 7, Pages (October 2014)
Johanna Jakobsdottir, Mary Sara McPeek 
Erratum The American Journal of Human Genetics
Neural Signatures of Economic Preferences for Risk and Ambiguity
Table of contents The Journal for Nurse Practitioners
Long-term Outcomes of Acute Kidney Injury: The Power and Pitfalls of Observational/Population-Based Studies  Charuhas V. Thakar, MD  American Journal.
Γ-TEMPy: Simultaneous Fitting of Components in 3D-EM Maps of Their Assembly Using a Genetic Algorithm  Arun Prasad Pandurangan, Daven Vasishtan, Frank.
J.G. Laffey, é. Tobin, J.F. Boylan, A.J. McShane 
Tao Wang, Robert C. Elston  The American Journal of Human Genetics 
Using Neural Networks as an Aid in the Determination of Disease Status: Comparison of Clinical Diagnosis to Neural-Network Predictions in a Pedigree with.
Xing Hua, Haiming Xu, Yaning Yang, Jun Zhu, Pengyuan Liu, Yan Lu 
Harold A. Nieuwboer, René Pool, Conor V. Dolan, Dorret I
Alice S. Whittemore, Jerry Halpern 
Anna Middleton, J. Hewison, R.F. Mueller 
Zuoheng Wang, Mary Sara McPeek  The American Journal of Human Genetics 
Primer on Neural networks
Prof. Carolina Ruiz Department of Computer Science
Presentation transcript:

A Practical Framework Toward Prediction of Breaking Force and Disintegration of Tablet Formulations Using Machine Learning Tools  Ilgaz Akseli, Jingjin Xie, Leon Schultz, Nadia Ladyzhynsky, Tommasina Bramante, Xiaorong He, Rich Deanne, Keith R. Horspool, Robert Schwabe  Journal of Pharmaceutical Sciences  Volume 106, Issue 1, Pages 234-247 (January 2017) DOI: 10.1016/j.xphs.2016.08.026 Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 1 Basic working principle of neural network. The input parameters in the input layer are represented by neurons. The values are forwarded to the neurons in the hidden layer for evaluation with weight and bias. The final decision was transmitted to the output layer. Note: there could be multiple inputs, hidden layer, and outputs in a typical neural network. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 2 Flow chart of typical genetic algorithm. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 3 Flow chart of typical support vector machine. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 4 The flow chart for a typical random forest algorithm. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 5 Schematics of the experimental setup for measuring ultrasonic properties of tablets (a). An example of a waveform captured using the developed setup (b). Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 6 The preparation of data for developing models. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 7 Predicted tablet breaking force of 10 randomly selected tablets from internal data samples. The blue line depicts the experimental results and red line with square markers depicts predicted results. Each data point represents the mean value of 10 experimental data samples. The error bars are the ±1 standard deviation. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 8 Predicted tablet breaking force of 10 randomly selected tablets from external data samples. The blue line depicts the experimental results and red line with square markers depicts predicted results. Each data point represents the mean value of 10 experimental data samples. The error bars are the ±1 standard deviation. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 9 Plotted relative importance of inputs in the prediction of tablet breaking force using internal data samples. The evaluation was achieved using internal data (cross-validation) (a) and external data as the test set (b). Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 10 Predicted disintegration time of 10 randomly selected tablets from internal data samples. The blue line depicts the experimental results and red line with square markers depicts predicted results. Each data point represents the mean value of 10 experimental data samples. The error bars are the ±1 standard deviation. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 11 Predicted disintegration time of 10 randomly selected tablets from external data samples. The blue line depicts the experimental results and red line with square markers depicts predicted results. Each data point represents the mean value of 10 experimental data samples. The error bars are the ±1 standard deviation. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 12 Plotted variable importance in the prediction of tablet disintegration time: relative importance using internal data for predictions (a) and relative importance using external data for predictions (b). Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 13 Modeling flow and implementation of the predicted models. Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions

Figure 14 Comparisons among different disintegration times from experiments and predictions. The comparison between experimental disintegration time and predicted disintegration time using the tablet breaking force obtained from experiments (a) and the comparison between experimental disintegration time and predicted disintegration time using the predicted tablet breaking force obtained with RF model (b). Journal of Pharmaceutical Sciences 2017 106, 234-247DOI: (10.1016/j.xphs.2016.08.026) Copyright © 2016 American Pharmacists Association® Terms and Conditions