Predicting Case Fatality of Dengue Epidemic: Statistical Machine Learning Towards a Virtual Doctor

Subhagata Chattopadhyay; Amit Chattopadhyay; Elias Aifantis

doi:10.12974/2311-8792.2021.07.2

Predicting Case Fatality of Dengue Epidemic: Statistical Machine Learning Towards a Virtual Doctor

Subhagata Chattopadhyay, Amit Chattopadhyay^*, Elias Aifantis

^*Corresponding author for this work

College of Engineering and Physical Sciences

Research output: Contribution to journal › Article › peer-review

Abstract

Dengue fever is a self-limiting communicable viral disease, transmitted through mosquito bites. Its Case Fatality Grade (CFG) varies across population due to variations in viral load, immunity of the patient, early diagnosis, and availability of high-end treatment facility. This study describes an initial effort to automate the process of Dengue CFG predictions. Two established Statistical Machine Learning (SML) algorithms, Multiple Linear Regressions (MLR) and Multinomial Logistic Regressions (MnLR), are combined to substitute the existing Deep Learning methods for clinical decision making. We consider a vector of eleven sign-symptoms (independent variables), each weighted between [0,1] on a 3-point scale - ‘Mild’ (CFG<=0.33), ‘Moderate’ (0.33<CFG< 0.66), and ‘Severe’ (CFG>0.66). Results show that both classifiers are effective in early screening with similar accuracy levels (68% for MLR versus 72% for MnLR) although precision levels are far superior with MnLR (88%) than MLR (61%). This study is a futuristic step towards Machine Learning (ML) aided clinical diagnostic paradigms, as an alternative to computationally intensive Artificial Intelligence.

Original language	English
Pages (from-to)	10-24
Number of pages	15
Journal	Journal of Nanotechnology in Diagnosis and Treatment
Volume	7
DOIs	https://doi.org/10.12974/2311-8792.2021.07.2
Publication status	Published - 15 Oct 2021

Bibliographical note

© 2021 Chattopadhyay et al.; Licensee Savvy Science Publisher.
This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License
(http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.

Funding: The authors gratefully acknowledge partial financial support from the H2020-MSCA-RISE2016 program, grant no. 734485, entitled ‘Fracture
Across Scales and Materials, Processes and Disciplines (FRAMED)’.

Keywords

Statistical machine learning
Dengue epidemic
Multiple linear regressions
Multinomial logistic regressions

Access to Document

10.12974/2311-8792.2021.07.2Licence: CC BY-NC 3.0

JNDT-V7A2-Chattopadhyay
© 2021 Chattopadhyay et al.; Licensee Savvy Science Publisher. This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.
Final published version, 1.32 MBLicence: CC BY-NC 3.0

Cite this

@article{baa993386b974703863820fe1f063f85,

title = "Predicting Case Fatality of Dengue Epidemic: Statistical Machine Learning Towards a Virtual Doctor",

abstract = "Dengue fever is a self-limiting communicable viral disease, transmitted through mosquito bites. Its Case Fatality Grade (CFG) varies across population due to variations in viral load, immunity of the patient, early diagnosis, and availability of high-end treatment facility. This study describes an initial effort to automate the process of Dengue CFG predictions. Two established Statistical Machine Learning (SML) algorithms, Multiple Linear Regressions (MLR) and Multinomial Logistic Regressions (MnLR), are combined to substitute the existing Deep Learning methods for clinical decision making. We consider a vector of eleven sign-symptoms (independent variables), each weighted between [0,1] on a 3-point scale - {\textquoteleft}Mild{\textquoteright} (CFG<=0.33), {\textquoteleft}Moderate{\textquoteright} (0.330.66). Results show that both classifiers are effective in early screening with similar accuracy levels (68% for MLR versus 72% for MnLR) although precision levels are far superior with MnLR (88%) than MLR (61%). This study is a futuristic step towards Machine Learning (ML) aided clinical diagnostic paradigms, as an alternative to computationally intensive Artificial Intelligence.",

keywords = "Statistical machine learning, Dengue epidemic, Multiple linear regressions, Multinomial logistic regressions",

author = "Subhagata Chattopadhyay and Amit Chattopadhyay and Elias Aifantis",

note = "{\textcopyright} 2021 Chattopadhyay et al.; Licensee Savvy Science Publisher. This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited. Funding: The authors gratefully acknowledge partial financial support from the H2020-MSCA-RISE2016 program, grant no. 734485, entitled {\textquoteleft}Fracture Across Scales and Materials, Processes and Disciplines (FRAMED){\textquoteright}.",

year = "2021",

month = oct,

day = "15",

doi = "10.12974/2311-8792.2021.07.2",

language = "English",

volume = "7",

pages = "10--24",

}

TY - JOUR

T1 - Predicting Case Fatality of Dengue Epidemic: Statistical Machine Learning Towards a Virtual Doctor

AU - Chattopadhyay, Subhagata

AU - Chattopadhyay, Amit

AU - Aifantis, Elias

N1 - © 2021 Chattopadhyay et al.; Licensee Savvy Science Publisher. This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited. Funding: The authors gratefully acknowledge partial financial support from the H2020-MSCA-RISE2016 program, grant no. 734485, entitled ‘Fracture Across Scales and Materials, Processes and Disciplines (FRAMED)’.

PY - 2021/10/15

Y1 - 2021/10/15

N2 - Dengue fever is a self-limiting communicable viral disease, transmitted through mosquito bites. Its Case Fatality Grade (CFG) varies across population due to variations in viral load, immunity of the patient, early diagnosis, and availability of high-end treatment facility. This study describes an initial effort to automate the process of Dengue CFG predictions. Two established Statistical Machine Learning (SML) algorithms, Multiple Linear Regressions (MLR) and Multinomial Logistic Regressions (MnLR), are combined to substitute the existing Deep Learning methods for clinical decision making. We consider a vector of eleven sign-symptoms (independent variables), each weighted between [0,1] on a 3-point scale - ‘Mild’ (CFG<=0.33), ‘Moderate’ (0.330.66). Results show that both classifiers are effective in early screening with similar accuracy levels (68% for MLR versus 72% for MnLR) although precision levels are far superior with MnLR (88%) than MLR (61%). This study is a futuristic step towards Machine Learning (ML) aided clinical diagnostic paradigms, as an alternative to computationally intensive Artificial Intelligence.

AB - Dengue fever is a self-limiting communicable viral disease, transmitted through mosquito bites. Its Case Fatality Grade (CFG) varies across population due to variations in viral load, immunity of the patient, early diagnosis, and availability of high-end treatment facility. This study describes an initial effort to automate the process of Dengue CFG predictions. Two established Statistical Machine Learning (SML) algorithms, Multiple Linear Regressions (MLR) and Multinomial Logistic Regressions (MnLR), are combined to substitute the existing Deep Learning methods for clinical decision making. We consider a vector of eleven sign-symptoms (independent variables), each weighted between [0,1] on a 3-point scale - ‘Mild’ (CFG<=0.33), ‘Moderate’ (0.330.66). Results show that both classifiers are effective in early screening with similar accuracy levels (68% for MLR versus 72% for MnLR) although precision levels are far superior with MnLR (88%) than MLR (61%). This study is a futuristic step towards Machine Learning (ML) aided clinical diagnostic paradigms, as an alternative to computationally intensive Artificial Intelligence.

KW - Statistical machine learning

KW - Dengue epidemic

KW - Multiple linear regressions

KW - Multinomial logistic regressions

UR - http://savvysciencepublisher.com/downloads/jndtv7a2/

U2 - 10.12974/2311-8792.2021.07.2

DO - 10.12974/2311-8792.2021.07.2

M3 - Article

SN - 2311-8792

VL - 7

SP - 10

EP - 24

JO - Journal of Nanotechnology in Diagnosis and Treatment

JF - Journal of Nanotechnology in Diagnosis and Treatment

ER -

Predicting Case Fatality of Dengue Epidemic: Statistical Machine Learning Towards a Virtual Doctor

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this