Disease Detection Analytics: A Simple Linear Convex Programming Algorithm for Breast Cancer and Diabetes Incidence Decisions

Somnath Mukhopadhyay, Subhashish Samaddar, Adriano O. Solis, Asim Roy

Research output: Contribution to journalArticle

Abstract

In the last couple of decades, data analytics-based pattern classification methods for disease detection have gained much traction in healthcare research and applications. The current study builds linear programming (LP) models for detecting disease incidence. We propose sequential steps of a convex programming algorithm to construct decision boundary functions to classify patterns in disease detection data. We compare the performance of our LP-based classifier with others (neural network, decision tree, k-nearest-neighbor, logistic regression, naïve-Bayes, and support-vector-machine) on four datasets: two different ones for breast cancer, and one each for diabetes and diabetic retinopathy. Statistical tests reveal that the LP classifier did significantly better than the other methods in five out of eight false-positive and false-negative test cases. There is not a statistically significant difference in performance in the remaining three tests between the LP classifier and the best alternative method. Most importantly, the LP classifier has significantly superior performance in both diabetes detection and diabetic retinopathy data. The success of the proposed LP classifier results from avoiding “modeling noise” and “memorization of training data.” We recommend that our proposed LP classifier be among the set of classifiers for use in disease detection analytics.

Original languageEnglish (US)
JournalDecision Sciences
DOIs
StateAccepted/In press - Jan 1 2018

Fingerprint

Convex optimization
Medical problems
Linear programming
Classifiers
Statistical tests
Decision trees
Programming
Breast cancer
Diabetes
Classifier
Pattern recognition
Support vector machines
Logistics
Neural networks

Keywords

  • Clinical Data
  • Convex Programming
  • Decision Tree
  • Disease Detection Analytics
  • Linear Programming
  • Neural Network
  • Pattern Classification

ASJC Scopus subject areas

  • Business, Management and Accounting(all)
  • Strategy and Management
  • Information Systems and Management
  • Management of Technology and Innovation

Cite this

Disease Detection Analytics : A Simple Linear Convex Programming Algorithm for Breast Cancer and Diabetes Incidence Decisions. / Mukhopadhyay, Somnath; Samaddar, Subhashish; Solis, Adriano O.; Roy, Asim.

In: Decision Sciences, 01.01.2018.

Research output: Contribution to journalArticle

@article{689672fe05a84856bc782f0e0943527c,
title = "Disease Detection Analytics: A Simple Linear Convex Programming Algorithm for Breast Cancer and Diabetes Incidence Decisions",
abstract = "In the last couple of decades, data analytics-based pattern classification methods for disease detection have gained much traction in healthcare research and applications. The current study builds linear programming (LP) models for detecting disease incidence. We propose sequential steps of a convex programming algorithm to construct decision boundary functions to classify patterns in disease detection data. We compare the performance of our LP-based classifier with others (neural network, decision tree, k-nearest-neighbor, logistic regression, na{\"i}ve-Bayes, and support-vector-machine) on four datasets: two different ones for breast cancer, and one each for diabetes and diabetic retinopathy. Statistical tests reveal that the LP classifier did significantly better than the other methods in five out of eight false-positive and false-negative test cases. There is not a statistically significant difference in performance in the remaining three tests between the LP classifier and the best alternative method. Most importantly, the LP classifier has significantly superior performance in both diabetes detection and diabetic retinopathy data. The success of the proposed LP classifier results from avoiding “modeling noise” and “memorization of training data.” We recommend that our proposed LP classifier be among the set of classifiers for use in disease detection analytics.",
keywords = "Clinical Data, Convex Programming, Decision Tree, Disease Detection Analytics, Linear Programming, Neural Network, Pattern Classification",
author = "Somnath Mukhopadhyay and Subhashish Samaddar and Solis, {Adriano O.} and Asim Roy",
year = "2018",
month = "1",
day = "1",
doi = "10.1111/deci.12348",
language = "English (US)",
journal = "Decision Sciences",
issn = "0011-7315",
publisher = "Wiley-Blackwell",

}

TY - JOUR

T1 - Disease Detection Analytics

T2 - A Simple Linear Convex Programming Algorithm for Breast Cancer and Diabetes Incidence Decisions

AU - Mukhopadhyay, Somnath

AU - Samaddar, Subhashish

AU - Solis, Adriano O.

AU - Roy, Asim

PY - 2018/1/1

Y1 - 2018/1/1

N2 - In the last couple of decades, data analytics-based pattern classification methods for disease detection have gained much traction in healthcare research and applications. The current study builds linear programming (LP) models for detecting disease incidence. We propose sequential steps of a convex programming algorithm to construct decision boundary functions to classify patterns in disease detection data. We compare the performance of our LP-based classifier with others (neural network, decision tree, k-nearest-neighbor, logistic regression, naïve-Bayes, and support-vector-machine) on four datasets: two different ones for breast cancer, and one each for diabetes and diabetic retinopathy. Statistical tests reveal that the LP classifier did significantly better than the other methods in five out of eight false-positive and false-negative test cases. There is not a statistically significant difference in performance in the remaining three tests between the LP classifier and the best alternative method. Most importantly, the LP classifier has significantly superior performance in both diabetes detection and diabetic retinopathy data. The success of the proposed LP classifier results from avoiding “modeling noise” and “memorization of training data.” We recommend that our proposed LP classifier be among the set of classifiers for use in disease detection analytics.

AB - In the last couple of decades, data analytics-based pattern classification methods for disease detection have gained much traction in healthcare research and applications. The current study builds linear programming (LP) models for detecting disease incidence. We propose sequential steps of a convex programming algorithm to construct decision boundary functions to classify patterns in disease detection data. We compare the performance of our LP-based classifier with others (neural network, decision tree, k-nearest-neighbor, logistic regression, naïve-Bayes, and support-vector-machine) on four datasets: two different ones for breast cancer, and one each for diabetes and diabetic retinopathy. Statistical tests reveal that the LP classifier did significantly better than the other methods in five out of eight false-positive and false-negative test cases. There is not a statistically significant difference in performance in the remaining three tests between the LP classifier and the best alternative method. Most importantly, the LP classifier has significantly superior performance in both diabetes detection and diabetic retinopathy data. The success of the proposed LP classifier results from avoiding “modeling noise” and “memorization of training data.” We recommend that our proposed LP classifier be among the set of classifiers for use in disease detection analytics.

KW - Clinical Data

KW - Convex Programming

KW - Decision Tree

KW - Disease Detection Analytics

KW - Linear Programming

KW - Neural Network

KW - Pattern Classification

UR - http://www.scopus.com/inward/record.url?scp=85058975406&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058975406&partnerID=8YFLogxK

U2 - 10.1111/deci.12348

DO - 10.1111/deci.12348

M3 - Article

AN - SCOPUS:85058975406

JO - Decision Sciences

JF - Decision Sciences

SN - 0011-7315

ER -