a k means approach to clustering disease progressions
play

A k-means approach to clustering disease progressions Duc Thanh Anh - PowerPoint PPT Presentation

Sep 24, 2022 •368 likes •543 views

A k-means approach to clustering disease progressions Duc Thanh Anh Luong Varun Chandola Department of Computer Science & Engineering University at Buffalo IEEE ICHI 2017 August 26, 2017 Outline Motivation K-means approach An

  1. A k-means approach to clustering disease progressions Duc Thanh Anh Luong Varun Chandola Department of Computer Science & Engineering University at Buffalo IEEE ICHI 2017 August 26, 2017

  2. Outline • Motivation • K-means approach • An application for Chronic Kidney Disease • Generating patient-specific disease profiles

  3. Motivation • Find subgroup of patients have similar disease progression • Identify the underlying mechanism of the subgroup • Provide better treatment for each subgroup

  4. Motivation • Different patients have different disease progressions • Consider the case of Chronic Kidney Disease 90 ● Are there few general trends of ● ● ● ● disease progressions? ● patient ID ● 70 8562280 ● ● ● ● ● ● ● ● ● ● ● ● 8563881 ● ● ● ● ● ● ● ● ● ● ● ● ● 8567589 ● ● ● ● ● ● ● ● ● ● ● ● eGFR ● ● ● ● 8571050 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 8582794 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 50 ● ● ● ● ● 8587204 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 8587950 ● ● ● ● ● ● ● ● ● ● ● Can we group patients by their ● ● ● ● ● ● ● ● 8601598 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 8602147 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● progressions into few groups? ● 8602554 ● ● ● ● ● ● 30 ● ● ● ● ● ● ● ● ● ● ● ● ● ● 0 1000 2000 3000 days from first clinical record

  5. Motivation 75 eGFR 50 Trajectories of 500 25 patients 0 0 300 600 900 days from first clinical record Trajectories after being clustered

  6. Clustering problem and k-means algorithm • Cluster a set of data points into k clusters • Can be solved by K-means approach Bishop, Christopher M. Pattern recognition and machine learning . Springer, 2006.

  7. K-means approach ● 55 ● ● ● 50 eGFR ● ● Patient disease Data object ● ● progression 45 ● ● ● 40 50 0 500 1000 1500 days from first clinical record 45 ● 40 eGFR Distance ● ● ● 70 35 metric ● ● ● ● ● ● ● ● ● ● ● 60 ● ● ● 30 ● ● eGFR ● ● ● centroid 50 ● red 0 300 600 900 ● days from first clinical record ● ● ● 40 ● ● ● ● ● ● ● ● ● ● ● ● ● ● 30 ● Centroid Regression line ● ● 0 400 800 1200 days from first clinical record

  8. K-means approach randomly assign patient into k Initial step clusters No Perform regression for each cluster Update to obtain “centroid” step Yes No patient move End to another group? Assignme Assign patient to the the cluster nt step that has closest centroid

  9. Dataset & Preprocessing DARTNet patients (n = 69,817) Excluded Invalid birth year and sex Excluded Number of serum creatinine value (n = 6,418) records < 1 (n = 181) Excluded Invalid data records (n = 9) “Preprocessed” DARTNet patients (n = 63,209) Having eGFR values less than 60 for more than three months (n = 29,585) Excluded Observation duration < 1 Excluded year (n = 5,285) Number of serum creatinine records < 10 (n = 17,158) Final CKD cohort (n = 7,142)

  10. Clustering result

  11. Demographic distribution in clusters 0.73% 4.56% 5.91% 80 (52) (326) (422) 9.49% (678) 14.86% (1061) 60 11.55% (825) age ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 13.72% ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 40 (980) ● ● ● ● ● ● ● ● ● ● ● ● ● Cluster 1 ● ● 9.61% ● ● ● ● ● ● Cluster 2 ● ● ● (686) Cluster 3 ● ● ● ● ● Cluster 4 ● ● Cluster 5 ● Cluster 6 Cluster 7 ● ● Cluster 8 20 14.59% Cluster 9 14.98% ● (1042) Cluster 10 (1070) ● ● ● 2.5 5.0 7.5 10.0 cluster

  12. Other clinical markers

  13. Generating patient-specific disease profiles ● 40 residuals ● 35 eGFR ● ● Gaussian processes ● ● ● ● ● 30 ● ● 0 300 600 900 days from first clinical record Rasmussen, Carl Edward, and Christopher KI Williams. Gaussian processes for machine learning . Vol. 1. Cambridge: MIT press, 2006.

  14. Generating patient-specific disease profiles Patient 391 Cluster 5 100 100 cluster's trajectory cluster's trajectory individual predicted trajectory individual predicted trajectory 80 upper and lower limit 80 upper and lower limit actual eGFR value actual eGFR value ● ● 60 eGFR 60 40 40 20 20 0 0 0 200 400 600 800 1000 0 200 400 600 800 1000 day day

  15. Conclusion & Future Work • Clustering disease progressions – k-means approach • Generating individual prediction – Gaussian processes • Extend the approach to cope with multiple clinical markers • Give quantitative evaluation of clusters • Tightness • Separation

  16. Thank you

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

1 K-means clustering The K-means clustering algorithm can be seen as applying the EM algorithm to

1 K-means clustering The K-means clustering algorithm can be seen as applying the EM algorithm to

Statistical Modeling and Analysis of Neural Data (NEU 560) Princeton University, Spring 2018 Jonathan Pillow Lecture 18 notes: K-means and Factor Analysis Tues, 4.17 1 K-means clustering The K-means clustering algorithm can be seen as

547 views • 3 slides
Clustering: Hierarchical Clustering and K- Means Clustering Machine

Clustering: Hierarchical Clustering and K- Means Clustering Machine

Clustering: Hierarchical Clustering and K- Means Clustering Machine Learning 10-601B Seyoung Kim Many of these slides are derived from William Cohen,

660 views • 46 slides
Multi-variable Optimization K-means clustering K-means clustering on points is finding K

Multi-variable Optimization K-means clustering K-means clustering on points is finding K

Multi-variable Optimization K-means clustering K-means clustering on points is finding K central locations that reduce the distance of each point to the nearest central location (summed over all points) K=3 K-means clustering For

383 views • 10 slides
Clustering and Dimensionality Reduction Preview Clustering K -means clustering

Clustering and Dimensionality Reduction Preview Clustering K -means clustering

Clustering and Dimensionality Reduction Preview Clustering K -means clustering Mixture models Hierarchical clustering Dimensionality reduction Principal component analysis Multidimensional scaling Isomap

1.61k views • 17 slides
k -means clustering Method to automatically separate data sets into distinct groups. Clustering

k -means clustering Method to automatically separate data sets into distinct groups. Clustering

k -means clustering Method to automatically separate data sets into distinct groups. Clustering example Clustering example k -means clustering algorithm 1. Start with k randomly chosen means 2. Color data points by the shortest distance to any

1.56k views • 19 slides
Data Clustering: Data Clustering: 50 Years Beyond K means 50 Years Beyond K means 50 Years

Data Clustering: Data Clustering: 50 Years Beyond K means 50 Years Beyond K means 50 Years

Data Clustering: Data Clustering: 50 Years Beyond K means 50 Years Beyond K means 50 Years Beyond K-means 50 Years Beyond K-means Anil K. Jain Department of Computer Science Michigan State University King-Sun Fu King-Sun Fu King-Sun

1.03k views • 39 slides
Basics of k-means clustering CLUS TERIN G METH ODS W ITH S CIP Y Shaumik Daityari Business

Basics of k-means clustering CLUS TERIN G METH ODS W ITH S CIP Y Shaumik Daityari Business

Basics of k-means clustering CLUS TERIN G METH ODS W ITH S CIP Y Shaumik Daityari Business Analyst Why k-means clustering? A critical drawback of hierarchical clustering: runtime K means runs signicantly faster on large datasets

691 views • 25 slides
Contents Clustering K-means Mixture of Gaussians Expectation Maximization

Contents Clustering K-means Mixture of Gaussians Expectation Maximization

Contents Clustering K-means Mixture of Gaussians Expectation Maximization Variational Methods 1 Introduction to Machine Learning CMU-10701 Clustering and EM Barnabs Pczos &amp; Aarti Singh Clustering 3 K- means

1.04k views • 67 slides
Introduction to Machine Learning, Clustering and EM Barnab s P czos Contents Clustering

Introduction to Machine Learning, Clustering and EM Barnab s P czos Contents Clustering

Introduction to Machine Learning, Clustering and EM Barnab s P czos Contents Clustering K-means Mixture of Gaussians Expectation Maximization Variational Methods 2 Clustering 3 K- means clustering What is clustering?

1.32k views • 66 slides
LECTURE 7 Clustering The k-means algorithm Hierarchical Clustering The DBSCAN algorithm

LECTURE 7 Clustering The k-means algorithm Hierarchical Clustering The DBSCAN algorithm

DATA MINING LECTURE 7 Clustering The k-means algorithm Hierarchical Clustering The DBSCAN algorithm Clustering Evaluation What is a Clustering? In general a grouping of objects such that the objects in a group (cluster) are similar (or

1.39k views • 110 slides
Clustering kMeans, Expectation Maximization, Self-Organizing Maps Outline K-means

Clustering kMeans, Expectation Maximization, Self-Organizing Maps Outline K-means

Clustering kMeans, Expectation Maximization, Self-Organizing Maps Outline K-means clustering Hierarchical clustering Incremental clustering Probability-based clustering Self-Organising Maps Classification vs. Clustering

1k views • 67 slides
Lecture 23: Spectral clustering Hierarchical clustering What is a good clustering?

Lecture 23: Spectral clustering Hierarchical clustering What is a good clustering?

Lecture 23: Spectral clustering Hierarchical clustering What is a good clustering? Aykut Erdem May 2016 Hacettepe University Last time K-Means An iterative clustering algorithm - Initialize: Pick K random points as cluster

932 views • 64 slides
K-MEANS++ OPTIMAL INITIALIZATION ALGORITHM An Improved K-means Clustering Method OVERVIEW

K-MEANS++ OPTIMAL INITIALIZATION ALGORITHM An Improved K-means Clustering Method OVERVIEW

K-MEANS++ OPTIMAL INITIALIZATION ALGORITHM An Improved K-means Clustering Method OVERVIEW K-means Clustering Algorithm K-means++ Initialization Algorithm Experiment Datasets Conclusion K-MEANS CLUSTERING ALGORITHM A

313 views • 14 slides
Graph Clustering Graph Clustering What is clustering? What is clustering? Finding patterns

Graph Clustering Graph Clustering What is clustering? What is clustering? Finding patterns

Graph Clustering Graph Clustering What is clustering? What is clustering? Finding patterns in data, or grouping similar groups of data-points together into clusters . Clustering algorithms for numeric data: Lloyds K-means, EM

688 views • 19 slides
K -means Clustering Ke Chen Reading: [7.3, EA], [9.1, CMB] COMP24111 Machine Learning Outline

K -means Clustering Ke Chen Reading: [7.3, EA], [9.1, CMB] COMP24111 Machine Learning Outline

K -means Clustering Ke Chen Reading: [7.3, EA], [9.1, CMB] COMP24111 Machine Learning Outline Introduction K -means Algorithm Example How K -means partitions? K -means Demo Relevant Issues Application:

416 views • 22 slides
EM, K-Means, GNG 4-6-16 Reading Quiz Which of the following can be considered an instance of the

EM, K-Means, GNG 4-6-16 Reading Quiz Which of the following can be considered an instance of the

EM, K-Means, GNG 4-6-16 Reading Quiz Which of the following can be considered an instance of the EM algorithm? a) Agglomerative clustering b) Divisive clustering c) K-means clustering d) Growing neural gas EM algorithm E step:

445 views • 17 slides
Coco Cloud Project Overview Aljosa Pasic Atos Spain Mission Seamless compliance and

Coco Cloud Project Overview Aljosa Pasic Atos Spain Mission Seamless compliance and

Coco Cloud Project Overview Aljosa Pasic Atos Spain Mission Seamless compliance and confidentiality for data shared in the cloud and mobile services, aligned to agreements considering legal , business , organizational regulations and user

356 views • 25 slides
Objective Cluster Identification A Finite State Machine Approach Want to find and identify

Objective Cluster Identification A Finite State Machine Approach Want to find and identify

Objective Cluster Identification A Finite State Machine Approach Want to find and identify homogeneous to Cluster Identification Using the patches in a 2D matrix, where: Hoshen-Kopelman Algorithm Cluster membership defined by adjacency

354 views • 21 slides
Aerotropolis Atlanta CIDs Presentation Freight Cluster Plan Pre-Bid Meeting Stan Reecy, Project

Aerotropolis Atlanta CIDs Presentation Freight Cluster Plan Pre-Bid Meeting Stan Reecy, Project

Aerotropolis Atlanta CIDs Presentation Freight Cluster Plan Pre-Bid Meeting Stan Reecy, Project Manager Presentation Contents Aerotropolis Overview Specific Challenges Trouble Spots Project Deliverables Plan Aspirations Aerotropolis

306 views • 9 slides
iGENI Presentation for ORCA Cluster Meeting, GEC 10 Presented By iGENI Consortium:

iGENI Presentation for ORCA Cluster Meeting, GEC 10 Presented By iGENI Consortium:

iGENI Presentation for ORCA Cluster Meeting, GEC 10 Presented By iGENI Consortium: International Center for Advanced Internet Research, Northwestern University et al GENI Engineering Conference 10 (GEC 10) March 15-17, 2011 San Juan, Puerto

385 views • 4 slides
HPC Clusters: Best Practices and Performance Study Agenda HPC at HPE System

HPC Clusters: Best Practices and Performance Study Agenda HPC at HPE System

HPC Clusters: Best Practices and Performance Study Agenda HPC at HPE System Configuration and Tuning Best Practices for Building Applications Intel Xeon Processors Efficient Methods in Executing Applications Tools and

879 views • 38 slides
Partnering value proposition Through partnering with the ETC and its ecosystem, partners have

Partnering value proposition Through partnering with the ETC and its ecosystem, partners have

Partnering value proposition Through partnering with the ETC and its ecosystem, partners have the ability to meet their objectives through constructive collaboration, which effectively channels their efforts. This provides scale, scope,

206 views • 6 slides
Farmer Clusters Pete Thompson Game &amp; Wildlife Conservation Trust Biodiversity Adviser

Farmer Clusters Pete Thompson Game &amp; Wildlife Conservation Trust Biodiversity Adviser

Farmer Clusters Pete Thompson Game &amp; Wildlife Conservation Trust Biodiversity Adviser Landscape scale - the best way to manage soil, water &amp; wildlife? So, what are Farmer Clusters? Farmer clusters set out to improve

509 views • 18 slides
Processing Big Data with Pentaho Rakesh Saha Pentaho Senior Product Manager, Hitachi Vantara

Processing Big Data with Pentaho Rakesh Saha Pentaho Senior Product Manager, Hitachi Vantara

Processing Big Data with Pentaho Rakesh Saha Pentaho Senior Product Manager, Hitachi Vantara Agenda Pentahos Latest and Upcoming Features for Processing Big Data Batch or Real-time Process big data visually in future-proof way

258 views • 25 slides

More recommend