國立雲林科技大學 National Yunlin University of Science and Technology Predicting adequacy of vancomycin regimens: A learning-based classification approach to improving.

Slides:



Advertisements
Similar presentations
國立雲林科技大學 National Yunlin University of Science and Technology Application of LVQ to novelty detection using outlier training data Hyoung-joo Lee, Sungzoon.
Advertisements

Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Presenter : Yu Cheng Chen Author: Hichem.
Autocorrelation and Linkage Cause Bias in Evaluation of Relational Learners David Jensen and Jennifer Neville.
Evaluation.
Evaluation.
Ensemble Learning: An Introduction
Bagging LING 572 Fei Xia 1/24/06. Ensemble methods So far, we have covered several learning methods: FSA, HMM, DT, DL, TBL. Question: how to improve results?
Experimental Evaluation
Rotation Forest: A New Classifier Ensemble Method 交通大學 電子所 蕭晴駿 Juan J. Rodríguez and Ludmila I. Kuncheva.
CSCI 347 / CS 4206: Data Mining Module 06: Evaluation Topic 01: Training, Testing, and Tuning Datasets.
2015 AprilUNIVERSITY OF HAIFA, DEPARTMENT OF STATISTICS, SEMINAR FOR M.A 1 Hastie, Tibshirani and Friedman.The Elements of Statistical Learning (2nd edition,
國立雲林科技大學 National Yunlin University of Science and Technology Building Reactive Characters for Dynamic Gaming Environments Peter Blackburn and Barry O’Sullivan,
Data Mining Practical Machine Learning Tools and Techniques Slides for Chapter 5 of Data Mining by I. H. Witten, E. Frank and M. A. Hall 報告人:黃子齊
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Unsupervised pattern recognition models for mixed feature-type.
Machine Learning1 Machine Learning: Summary Greg Grudic CSCI-4830.
Biostatistics IV An introduction to bootstrap. 2 Getting something from nothing? In Rudolph Erich Raspe's tale, Baron Munchausen had, in one of his many.
Intelligent Database Systems Lab Advisor : Dr. Hsu Graduate : Chien-Shing Chen Author : Satoshi Oyama Takashi Kokubo Toru lshida 國立雲林科技大學 National Yunlin.
NEURAL NETWORKS FOR DATA MINING
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology A data mining approach to the prediction of corporate failure.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology A Taxonomy of Similarity Mechanisms for Case-Based Reasoning.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Looking inside self-organizing map ensembles with resampling.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology On Data Labeling for Clustering Categorical Data Hung-Leng.
Intelligent Database Systems Lab N.Y.U.S.T. I. M. OpinionMiner: A Novel Machine Learning System for Web Opinion Mining and Extraction Presenter : Jiang-Shan.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Wireless Sensor Network Wireless Sensor Network Based.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Extracting meaningful labels for WEBSOM text archives Advisor.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology A self-organizing neural network using ideas from the immune.
Ensemble Methods: Bagging and Boosting
Ensemble Learning Spring 2009 Ben-Gurion University of the Negev.
Intelligent Database Systems Lab Advisor : Dr. Hsu Graduate : Chien-Ming Hsiao Author : Bing Liu Yiyuan Xia Philp S. Yu 國立雲林科技大學 National Yunlin University.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 GMDH-based feature ranking and selection for improved.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Mining Logs Files for Data-Driven System Management Advisor.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Evolving Reactive NPCs for the Real-Time Simulation Game.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Utilizing Marginal Net Utility for Recommendation in E-commerce.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Efficient Optimal Linear Boosting of a Pair of Classifiers.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Yu Cheng Chen Author: Chung-hung.
Data Mining Practical Machine Learning Tools and Techniques By I. H. Witten, E. Frank and M. A. Hall Chapter 5: Credibility: Evaluating What’s Been Learned.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Sheng-Hsuan Wang Authors :
Intelligent Database Systems Lab Advisor : Dr.Hsu Graduate : Keng-Wei Chang Author : Lian Yan and David J. Miller 國立雲林科技大學 National Yunlin University of.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Model-based evaluation of clustering validation measures.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Rival-Model Penalized Self-Organizing Map Yiu-ming Cheung.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Information Loss of the Mahalanobis Distance in High Dimensions-
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Multiclass boosting with repartitioning Graduate : Chen,
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology O( ㏒ 2 M) Self-Organizing Map Algorithm Without Learning.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Enhanced neural gas network for prototype-based clustering.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 A self-organizing map for adaptive processing of structured.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Adaptive FIR Neural Model for Centroid Learning in Self-Organizing.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Cost- sensitive boosting for classification of imbalanced.
Validation methods.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Yu Cheng Chen Author: Wei Xu,
Intelligent Database Systems Lab N.Y.U.S.T. I. M. Predicting corporate bankruptcy using a self-organizing map: An empirical study to improve the forecasting.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Text Classification Improved through Multigram Models.
Intelligent Database Systems Lab Advisor : Dr. Hsu Graduate : Yu Cheng Chen Author : Yongqiang Cao Jianhong Wu 國立雲林科技大學 National Yunlin University of Science.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Dual clustering : integrating data clustering over optimization.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Presenter : Chien-Shing Chen Author: Gustavo.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Information Extraction from Wikipedia: Moving Down the Long.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Yu Cheng Chen Author: Lynette.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Chun Kai Chen Author : Andrew.
Data Mining Practical Machine Learning Tools and Techniques By I. H. Witten, E. Frank and M. A. Hall Chapter 5: Credibility: Evaluating What’s Been Learned.
國立雲林科技大學 National Yunlin University of Science and Technology Intelligent Database Systems Lab 1 Self-organizing map for cluster analysis of a breast cancer.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology A Nonlinear Mapping for Data Structure Analysis John W.
A Presentation on Adaptive Neuro-Fuzzy Inference System using Particle Swarm Optimization and it’s Application By Sumanta Kundu (En.R.No.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 f-information measures in medical image registration Presenter.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Investigating the Effect of Sampling Methods for Imbalanced.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Yu Cheng Chen Author: Michael.
Data Science Credibility: Evaluating What’s Been Learned
Ensemble Classifiers.
Machine Learning: Ensemble Methods
Data Mining Practical Machine Learning Tools and Techniques
network of simple neuron-like computing elements
Structure of a typical back-propagated multilayered perceptron used in this study. Structure of a typical back-propagated multilayered perceptron used.
Presentation transcript:

國立雲林科技大學 National Yunlin University of Science and Technology Predicting adequacy of vancomycin regimens: A learning-based classification approach to improving clinical decision making Paul Jen-Hwa Hua, Chih-Ping Wei, Tsang-Hsiang Cheng, Jian-Xun Chen, Decision Support Systems, (article in press). Presenter : Wei-Shen Tai Advisor : Professor Chung-Chian Hsu 2006/5/10

N.Y.U.S.T. I. M. Outline Introduction Data and automated classification techniques Classification system and evaluation design Evaluation result and discussion Conclusion Comments

N.Y.U.S.T. I. M. Motivation Clinicians' drug regimen decision making Of particularly alarming salience are problems surrounding sub- or overtherapeutic doses of high-alert medications. Managing the clinical use of vancomycin It is challenging because of its narrow therapeutic index (decision problem) and significant, lasting, adverse effects on patients (derived problem).

N.Y.U.S.T. I. M. Objective Decision support system to predict the adequacy of a vancomycin regimen is desirable. enhance the efficacy of initial regimen estimations by the nomogram and complement the pharmacokinetic analysis.

N.Y.U.S.T. I. M. Classification method Supervised learning techniques in artificial intelligence. Decision-tree induction C4.5 and a back propagation neural network Extend with Bagging.

N.Y.U.S.T. I. M. Bagging Bootstrap sampling To generate multiple training data sets from the original overall training data set create a distinct data set that consists of the same number of training instances as appear in the original data set. Construct base classifiers Based on machine learning techniques (e.g., C4.5 or the backpropagation neural network). This process terminates after reaching a specified number of iterations. Majority-voting scheme Integrates all constructed base classifiers when a new (unseen) instance have been classified.

N.Y.U.S.T. I. M. Evaluation and results Overall accuracy Significantly higher than that of the benchmark one-compartment pharmacokinetic model. Bagging can significantly improve the performance of each system. Insensitive fairly insensitive to the size of its training data set. C4.5 vs. NN Computation and performance

N.Y.U.S.T. I. M. Vancomycin and its clinical use Existed problem Vancomycin consistently has been identified as one of the top three adverse effect-producing pharmaceutical drugs in Taiwan between 1998 and General solution Clinicians often must supplement the nomogram with their experiences and patient condition assessments to adjust the regimen recommendations properly.

N.Y.U.S.T. I. M. Related prior research Pharmacokinetic model represents a mathematical scheme is crucial for estimating the elimination (discharge) rate of an administered drug. In general, an administered drug initially is distributed into a central compartment before diffusing into the peripheral compartment. Its prediction accuracy of peak and trough concentrations is limited.

N.Y.U.S.T. I. M. Bootstrap – A re-sampling method Fundamental idea Compute measures of our inference uncertainty from that estimated sampling distribution of f. Re-sampling using some form of re-sampling with replacement from the actual data, x, to generate B bootstrap samples, x*. Often, the data (sample) consist of n independent units and it then suffices to take a simple random sample of size n. Goal From the set of results of sample size B we measure our inference uncertainties from sample to (conceptual) population (see figure). Caution The bootstrap can work well for large sample sizes (n), but may not be reliable for small n (say 5, 10 or even 20), regardless of how many bootstrap samples, B, are used.

N.Y.U.S.T. I. M. Case tutorial for Bootstrap Sample data Consider a sample of weights of 27 rats (n = 27); the data are The sample mean of these data = , standard deviation = with cv = For illustration, what if we wanted an estimate of the standard error of cv. Processes First, we draw a random subsample of size 27 with replacement. Thus, while a weight of 63 appears in the actual sample, perhaps it would not appear in the subsample; or is could appear more than once. Second, the whole process is repeated B times (where we will let B = 1,000 reps for this example). Thus, we generate 1000 resample data sets (b = 1, 2, 3,..., 1000) and from each of these we compute the cv and store these values. Third, we obtain the standard error of the cv by taking the standard deviation of the 1000 cv values (corresponding to the 1000 bootstrap samples). The process is simple. In this case, the standard error is

N.Y.U.S.T. I. M. Experiment design A DSS based on Weka an open-source machine learning software. Analysis items Rregimen adequacy, 2 output nodes: appropriate and inappropriate. Peak concentrations, 3 output nodes (i.e., low, on-target, and high) Trough concentrations, 2 output nodes (i.e., on- target and high).

N.Y.U.S.T. I. M. Conclusions A solution for the vancomycin usage A decision support systems based on promising learning-based classification techniques in AI. Performance improvement Superior to the benchmark one-compartment pharmacokinetic model in prediction of the adequacy of vancomycin regimens.

N.Y.U.S.T. I. M. Comments Bagging magic When the number of sample data set is poor, it can (maybe) improve the accuracy of classification. Insensitivity test problem Maybe 40%, 60% and 80% of entire cases own high consistency or bagging magic causes this result also. Parameter (optimization) finding cost The iteration time for bagging, optimal number of hidden nodes and adequate parameter tuning for NN. 80/20 training/testing strategy Whether it will be better than 10 fold cross validation in the result or not? It makes the distribution of training sample may be inconsistent with original data set.