Population pharmacokinetic model selection assisted by machine learning

AbstractA fit-for-purpose structural and statistical model is the first major requirement in population pharmacometric model development. In this manuscript we discuss how this complex and computationally intensive task could benefit from supervised machine learning algorithms. We compared the classical pharmacometric approach with two machine learning methods, genetic algorithm and neural networks, in different scenarios based on simulated pharmacokinetic data. Genetic algorithm performance was assessed using a fitness function based on log-likelihood, whilst neural networks were trained using mean square error or binary cross-entropy loss. Machine learning provided a selection based only on statistical rules and achieved accurate selection. The minimization process of genetic algorithm was successful at allowing the algorithm to select plausible models. Neural network classification tasks achieved the most accurate results. Neural network regression tasks were less precise than neural network classification and genetic algorithm methods. The computational gain obtained by using machine learning was substantial, especially in the case of neural networks. We demonstrated that machine learning methods can greatly increase the efficiency of pharmacokinetic population model selection in case of large datasets or complex models requiring long run-times. Our results suggest that machine learning approaches can achieve a first fast selection of models which can be followed by more conventional pharmacometric approaches.

Download Full-text

Toward Audio Beehive Monitoring: Deep Learning vs. Standard Machine Learning in Classifying Beehive Audio Samples

Applied Sciences ◽

10.3390/app8091573 ◽

2018 ◽

Vol 8 (9) ◽

pp. 1573 ◽

Cited By ~ 10

Author(s):

Vladimir Kulyukin ◽

Sarbajit Mukherjee ◽

Prakhar Amlathe

Keyword(s):

Neural Network ◽

Machine Learning ◽

Neural Networks ◽

Deep Learning ◽

Convolutional Neural Network ◽

Convolutional Neural Networks ◽

Feature Engineering ◽

Learning Methods ◽

Machine Learning Methods

Electronic beehive monitoring extracts critical information on colony behavior and phenology without invasive beehive inspections and transportation costs. As an integral component of electronic beehive monitoring, audio beehive monitoring has the potential to automate the identification of various stressors for honeybee colonies from beehive audio samples. In this investigation, we designed several convolutional neural networks and compared their performance with four standard machine learning methods (logistic regression, k-nearest neighbors, support vector machines, and random forests) in classifying audio samples from microphones deployed above landing pads of Langstroth beehives. On a dataset of 10,260 audio samples where the training and testing samples were separated from the validation samples by beehive and location, a shallower raw audio convolutional neural network with a custom layer outperformed three deeper raw audio convolutional neural networks without custom layers and performed on par with the four machine learning methods trained to classify feature vectors extracted from raw audio samples. On a more challenging dataset of 12,914 audio samples where the training and testing samples were separated from the validation samples by beehive, location, time, and bee race, all raw audio convolutional neural networks performed better than the four machine learning methods and a convolutional neural network trained to classify spectrogram images of audio samples. A trained raw audio convolutional neural network was successfully tested in situ on a low voltage Raspberry Pi computer, which indicates that convolutional neural networks can be added to a repertoire of in situ audio classification algorithms for electronic beehive monitoring. The main trade-off between deep learning and standard machine learning is between feature engineering and training time: while the convolutional neural networks required no feature engineering and generalized better on the second, more challenging dataset, they took considerably more time to train than the machine learning methods. To ensure the replicability of our findings and to provide performance benchmarks for interested research and citizen science communities, we have made public our source code and our curated datasets.

Download Full-text

FRI0046 PHARMACOGENOMICS-DRIVEN INDIVIDUALIZED PREDICTION OF TREATMENT RESPONSE TO METHOTREXATE IN PATIENTS WITH RHEUMATOID ARTHRITIS: A MACHINE LEARNING APPROACH

Annals of the Rheumatic Diseases ◽

10.1136/annrheumdis-2020-eular.4993 ◽

2020 ◽

Vol 79 (Suppl 1) ◽

pp. 598.2-598

Author(s):

E. Myasoedova ◽

A. Athreya ◽

C. S. Crowson ◽

R. Weinshilboum ◽

L. Wang ◽

...

Keyword(s):

Rheumatoid Arthritis ◽

Machine Learning ◽

Supervised Machine Learning ◽

Research Support ◽

Eular Response ◽

Learning Methods ◽

Machine Learning Methods ◽

Early Ra ◽

Genome Wide

Background:Methotrexate (MTX) is the most common anchor drug for rheumatoid arthritis (RA), but the risk of missing the opportunity for early effective treatment with alternative medications is substantial given the delayed onset of MTX action and 30-40% inadequate response rate. There is a compelling need to accurately predicting MTX response prior to treatment initiation, which allows for effectively identifying patients at RA onset who are likely to respond to MTX.Objectives:To test the ability of machine learning approaches with clinical and genomic biomarkers to predict MTX response with replications in independent samples.Methods:Age, sex, clinical, serological and genome-wide association study (GWAS) data on patients with early RA of European ancestry from 647 patients (336 recruited in United Kingdom [UK]; 307 recruited across Europe; 70% female; 72% rheumatoid factor [RF] positive; mean age 54 years; mean baseline Disease Activity Score with 28-joint count [DAS28] 5.65) of the PhArmacogenetics of Methotrexate in RA (PAMERA) consortium was used in this study. The genomics data comprised 160 genome-wide significant single nucleotide polymorphisms (SNPs) with p<1×10-5 associated with risk of RA and MTX metabolism. DAS28 score was available at baseline and 3-month follow-up visit. Response to MTX monotherapy at the dose of ≥15 mg/week was defined as good or moderate by the EULAR response criteria at 3 months’ follow up visit. Supervised machine-learning methods were trained with 5-repeats and 10-fold cross-validation using data from PAMERA’s 336 UK patients. Class imbalance (higher % of MTX responders) in training was accounted by using simulated minority oversampling technique. Prediction performance was validated in PAMERA’s 307 European patients (not used in training).Results:Age, sex, RF positivity and baseline DAS28 data predicted MTX response with 58% accuracy of UK and European patients (p = 0.7). However, supervised machine-learning methods that combined demographics, RF positivity, baseline DAS28 and genomic SNPs predicted EULAR response at 3 months with area under the receiver operating curve (AUC) of 0.83 (p = 0.051) in UK patients, and achieved prediction accuracies (fraction of correctly predicted outcomes) of 76.2% (p = 0.054) in the European patients, with sensitivity of 72% and specificity of 77%. The addition of genomic data improved the predictive accuracies of MTX response by 19% and achieved cross-site replication. Baseline DAS28 scores and following SNPs rs12446816, rs13385025, rs113798271, and rs2372536 were among the top predictors of MTX response.Conclusion:Pharmacogenomic biomarkers combined with DAS28 scores predicted MTX response in patients with early RA more reliably than using demographics and DAS28 scores alone. Using pharmacogenomics biomarkers for identification of MTX responders at early stages of RA may help to guide effective RA treatment choices, including timely escalation of RA therapies. Further studies on personalized prediction of response to MTX and other anti-rheumatic treatments are warranted to optimize control of RA disease and improve outcomes in patients with RA.Disclosure of Interests:Elena Myasoedova: None declared, Arjun Athreya: None declared, Cynthia S. Crowson Grant/research support from: Pfizer research grant, Richard Weinshilboum Shareholder of: co-founder and stockholder in OneOme, Liewei Wang: None declared, Eric Matteson Grant/research support from: Pfizer, Consultant of: Boehringer Ingelheim, Gilead, TympoBio, Arena Pharmaceuticals, Speakers bureau: Simply Speaking

Download Full-text

Identifying Spatiotemporal Patterns in Land Use and Cover Samples from Satellite Image Time Series

Remote Sensing ◽

10.3390/rs13050974 ◽

2021 ◽

Vol 13 (5) ◽

pp. 974

Author(s):

Lorena Alves Santos ◽

Karine Ferreira ◽

Michelle Picoli ◽

Gilberto Camara ◽

Raul Zurita-Milla ◽

...

Keyword(s):

Machine Learning ◽

Land Use ◽

Time Series ◽

Satellite Image ◽

Spatiotemporal Patterns ◽

Supervised Machine Learning ◽

Learning Methods ◽

Self Organizing Maps ◽

Machine Learning Methods ◽

Land Use And Cover

The use of satellite image time series analysis and machine learning methods brings new opportunities and challenges for land use and cover changes (LUCC) mapping over large areas. One of these challenges is the need for samples that properly represent the high variability of land used and cover classes over large areas to train supervised machine learning methods and to produce accurate LUCC maps. This paper addresses this challenge and presents a method to identify spatiotemporal patterns in land use and cover samples to infer subclasses through the phenological and spectral information provided by satellite image time series. The proposed method uses self-organizing maps (SOMs) to reduce the data dimensionality creating primary clusters. From these primary clusters, it uses hierarchical clustering to create subclusters that recognize intra-class variability intrinsic to different regions and periods, mainly in large areas and multiple years. To show how the method works, we use MODIS image time series associated to samples of cropland and pasture classes over the Cerrado biome in Brazil. The results prove that the proposed method is suitable for identifying spatiotemporal patterns in land use and cover samples that can be used to infer subclasses, mainly for crop-types.

Download Full-text

Possibility of Autonomous Estimation of Shiba Goat’s Estrus and Non-Estrus Behavior by Machine Learning Methods

Animals ◽

10.3390/ani10050771 ◽

2020 ◽

Vol 10 (5) ◽

pp. 771

Author(s):

Toshiya Arakawa

Keyword(s):

Neural Network ◽

Machine Learning ◽

Random Forest ◽

Markov Models ◽

Tracking System ◽

Video Tracking ◽

Training Data ◽

Support Vector ◽

Learning Methods ◽

Machine Learning Methods

Mammalian behavior is typically monitored by observation. However, direct observation requires a substantial amount of effort and time, if the number of mammals to be observed is sufficiently large or if the observation is conducted for a prolonged period. In this study, machine learning methods as hidden Markov models (HMMs), random forests, support vector machines (SVMs), and neural networks, were applied to detect and estimate whether a goat is in estrus based on the goat’s behavior; thus, the adequacy of the method was verified. Goat’s tracking data was obtained using a video tracking system and used to estimate whether they, which are in “estrus” or “non-estrus”, were in either states: “approaching the male”, or “standing near the male”. Totally, the PC of random forest seems to be the highest. However, The percentage concordance (PC) value besides the goats whose data were used for training data sets is relatively low. It is suggested that random forest tend to over-fit to training data. Besides random forest, the PC of HMMs and SVMs is high. However, considering the calculation time and HMM’s advantage in that it is a time series model, HMM is better method. The PC of neural network is totally low, however, if the more goat’s data were acquired, neural network would be an adequate method for estimation.

Download Full-text

Landslide susceptibility mapping based on convolutional neural network and conventional machine learning methods

10.21203/rs.3.rs-190195/v1 ◽

2021 ◽

Author(s):

Rui Liu ◽

Xin Yang ◽

Chong Xu ◽

Luyao Li ◽

Xiangqiang Zeng

Keyword(s):

Neural Network ◽

Machine Learning ◽

Convolutional Neural Network ◽

Landslide Susceptibility ◽

Susceptibility Mapping ◽

Landslide Susceptibility Mapping ◽

Support Vector ◽

Learning Methods ◽

Machine Learning Methods ◽

Conventional Machine

Abstract Landslide susceptibility mapping (LSM) is a useful tool to estimate the probability of landslide occurrence, providing a scientific basis for natural hazards prevention, land use planning, and economic development in landslide-prone areas. To date, a large number of machine learning methods have been applied to LSM, and recently the advanced Convolutional Neural Network (CNN) has been gradually adopted to enhance the prediction accuracy of LSM. The objective of this study is to introduce a CNN based model in LSM and systematically compare its overall performance with the conventional machine learning models of random forest, logistic regression, and support vector machine. Herein, we selected the Jiuzhaigou region in Sichuan Province, China as the study area. A total number of 710 landslides and 12 predisposing factors were stacked to form spatial datasets for LSM. The ROC analysis and several statistical metrics, such as accuracy, root mean square error (RMSE), Kappa coefficient, sensitivity, and specificity were used to evaluate the performance of the models in the training and validation datasets. Finally, the trained models were calculated and the landslide susceptibility zones were mapped. Results suggest that both CNN and conventional machine-learning based models have a satisfactory performance (AUC: 85.72% − 90.17%). The CNN based model exhibits excellent good-of-fit and prediction capability, and achieves the highest performance (AUC: 90.17%) but also significantly reduces the salt-of-pepper effect, which indicates its great potential of application to LSM.

Download Full-text

Machine-learning in astronomy

Proceedings of the International Astronomical Union ◽

10.1017/s1743921314013672 ◽

2014 ◽

Vol 10 (S306) ◽

pp. 279-287 ◽

Cited By ~ 2

Author(s):

Michael Hobson ◽

Philip Graff ◽

Farhan Feroz ◽

Anthony Lasenby

Keyword(s):

Neural Network ◽

Machine Learning ◽

Neural Networks ◽

Gamma Ray ◽

Neural Network Training ◽

Training Algorithm ◽

Data Description ◽

Astronomical Data ◽

Machine Learning Methods ◽

Network Training

AbstractMachine-learning methods may be used to perform many tasks required in the analysis of astronomical data, including: data description and interpretation, pattern recognition, prediction, classification, compression, inference and many more. An intuitive and well-established approach to machine learning is the use of artificial neural networks (NNs), which consist of a group of interconnected nodes, each of which processes information that it receives and then passes this product on to other nodes via weighted connections. In particular, I discuss the first public release of the generic neural network training algorithm, calledSkyNet, and demonstrate its application to astronomical problems focusing on its use in the BAMBI package for accelerated Bayesian inference in cosmology, and the identification of gamma-ray bursters. TheSkyNetand BAMBI packages, which are fully parallelised using MPI, are available athttp://www.mrao.cam.ac.uk/software/.

Download Full-text

Identification of Village Building via Google Earth Images and Supervised Machine Learning Methods

Remote Sensing ◽

10.3390/rs8040271 ◽

2016 ◽

Vol 8 (4) ◽

pp. 271 ◽

Cited By ~ 29

Author(s):

Zhiling Guo ◽

Xiaowei Shao ◽

Yongwei Xu ◽

Hiroyuki Miyazaki ◽

Wataru Ohira ◽

...

Keyword(s):

Machine Learning ◽

Google Earth ◽

Supervised Machine Learning ◽

Learning Methods ◽

Machine Learning Methods

Download Full-text

Seeing It All: Evaluating Supervised Machine Learning Methods for the Classification of Diverse Otariid Behaviours

PLoS ONE ◽

10.1371/journal.pone.0166898 ◽

2016 ◽

Vol 11 (12) ◽

pp. e0166898 ◽

Cited By ~ 15

Author(s):

Monique A. Ladds ◽

Adam P. Thompson ◽

David J. Slip ◽

David P. Hocking ◽

Robert G. Harcourt

Keyword(s):

Machine Learning ◽

Supervised Machine Learning ◽

Learning Methods ◽

Machine Learning Methods

Download Full-text

Detecting Items with the Biggest Weight Based on Neural Network and Machine Learning Methods

Communications in Computer and Information Science - Data Stream Mining & Processing ◽

10.1007/978-3-030-61656-4_26 ◽

2020 ◽

pp. 383-396

Author(s):

Vitaliy Danylyk ◽

Victoria Vysotska ◽

Vasyl Lytvyn ◽

Svitlana Vyshemyrska ◽

Iryna Lurie ◽

...

Keyword(s):

Neural Network ◽

Machine Learning ◽

Learning Methods ◽

Machine Learning Methods

Download Full-text

Acoustic feature-based sentiment analysis of call center data

10.32469/10355/66751 ◽

2017 ◽

Author(s):

◽

Zeshan Peng

Keyword(s):

Machine Learning ◽

Neural Networks ◽

Emotion Recognition ◽

Sentiment Analysis ◽

Call Center ◽

Machine Learning Algorithms ◽

Language Recognition ◽

Acoustic Features ◽

Learning Methods ◽

Machine Learning Methods

With the advancement of machine learning methods, audio sentiment analysis has become an active research area in recent years. For example, business organizations are interested in persuasion tactics from vocal cues and acoustic measures in speech. A typical approach is to find a set of acoustic features from audio data that can indicate or predict a customer's attitude, opinion, or emotion state. For audio signals, acoustic features have been widely used in many machine learning applications, such as music classification, language recognition, emotion recognition, and so on. For emotion recognition, previous work shows that pitch and speech rate features are important features. This thesis work focuses on determining sentiment from call center audio records, each containing a conversation between a sales representative and a customer. The sentiment of an audio record is considered positive if the conversation ended with an appointment being made, and is negative otherwise. In this project, a data processing and machine learning pipeline for this problem has been developed. It consists of three major steps: 1) an audio record is split into segments by speaker turns; 2) acoustic features are extracted from each segment; and 3) classification models are trained on the acoustic features to predict sentiment. Different set of features have been used and different machine learning methods, including classical machine learning algorithms and deep neural networks, have been implemented in the pipeline. In our deep neural network method, the feature vectors of audio segments are stacked in temporal order into a feature matrix, which is fed into deep convolution neural networks as input. Experimental results based on real data shows that acoustic features, such as Mel frequency cepstral coefficients, timbre and Chroma features, are good indicators for sentiment. Temporal information in an audio record can be captured by deep convolutional neural networks for improved prediction accuracy.

Download Full-text