CNN Based Model for Malaria Diagnosis with Knowledge Distillation

K. M.Faizullah Fuhad; Jannat Ferdousey Tuba; Tanzilur Rahman; Nabeel Mohammed

doi:10.1145/3408127.3408159

CNN Based Model for Malaria Diagnosis with Knowledge Distillation

K. M.Faizullah Fuhad, Jannat Ferdousey Tuba, Tanzilur Rahman, Nabeel Mohammed

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

Malaria is a deadly disease caused by plasmodium parasites and carried by female anopheles mosquitoes. Bangladesh is one of the major malaria prone country with poor infrastructures, unable to diagnose Malaria rapid at a low cost. Here we propose a custom8 layers CNN architecture that is only 233.60KB in size meaning it can easily fit and work into a low cost smartphone. The model has been optimized through different preprocessing methods and model pruning techniques like knowledge distillation ensuring that the accuracy does not reduce due to smaller model size. The accuracy received from the optimized model is~ 96.51% when tested on NIH data-set containing images of infected and uninfected cells. Proposed model is a big step towards a complete mobile based rapid diagnostic platform that can be used in any resource restricted and hilly areas of Bangladesh.

Original language	English
Title of host publication	ICDSP 2020 - 2020 4th International Conference on Digital Signal Processing, Proceedings
Publisher	Association for Computing Machinery
Pages	131-135
Number of pages	5
ISBN (Electronic)	9781450376877
DOIs	https://doi.org/10.1145/3408127.3408159
State	Published - 19 Jun 2020
Externally published	Yes
Event	4th International Conference on Digital Signal Processing, ICDSP 2020 - Virtual, Online, China Duration: 19 Jun 2020 → 21 Jun 2020

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	4th International Conference on Digital Signal Processing, ICDSP 2020
Country/Territory	China
City	Virtual, Online
Period	19/06/20 → 21/06/20

Bibliographical note

Publisher Copyright:
© 2020 ACM.

Keywords

Anopheles
CNN
Knowledge Distillation
Malaria diagnostic system
Model pruning
Plasmodium
Resource restricted

ASJC Scopus subject areas

Software
Human-Computer Interaction
Computer Vision and Pattern Recognition
Computer Networks and Communications

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1145/3408127.3408159

Cite this

@inproceedings{5cfe1b380702440a8375af00e9fdbbe7,

title = "CNN Based Model for Malaria Diagnosis with Knowledge Distillation",

abstract = "Malaria is a deadly disease caused by plasmodium parasites and carried by female anopheles mosquitoes. Bangladesh is one of the major malaria prone country with poor infrastructures, unable to diagnose Malaria rapid at a low cost. Here we propose a custom8 layers CNN architecture that is only 233.60KB in size meaning it can easily fit and work into a low cost smartphone. The model has been optimized through different preprocessing methods and model pruning techniques like knowledge distillation ensuring that the accuracy does not reduce due to smaller model size. The accuracy received from the optimized model is\textasciitilde{} 96.51\% when tested on NIH data-set containing images of infected and uninfected cells. Proposed model is a big step towards a complete mobile based rapid diagnostic platform that can be used in any resource restricted and hilly areas of Bangladesh.",

keywords = "Anopheles, CNN, Knowledge Distillation, Malaria diagnostic system, Model pruning, Plasmodium, Resource restricted",

author = "Fuhad, \{K. M.Faizullah\} and Tuba, \{Jannat Ferdousey\} and Tanzilur Rahman and Nabeel Mohammed",

note = "Publisher Copyright: {\textcopyright} 2020 ACM.; 4th International Conference on Digital Signal Processing, ICDSP 2020 ; Conference date: 19-06-2020 Through 21-06-2020",

year = "2020",

month = jun,

day = "19",

doi = "10.1145/3408127.3408159",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "131--135",

booktitle = "ICDSP 2020 - 2020 4th International Conference on Digital Signal Processing, Proceedings",

}

Fuhad, KMF, Tuba, JF, Rahman, T & Mohammed, N 2020, CNN Based Model for Malaria Diagnosis with Knowledge Distillation. in ICDSP 2020 - 2020 4th International Conference on Digital Signal Processing, Proceedings. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 131-135, 4th International Conference on Digital Signal Processing, ICDSP 2020, Virtual, Online, China, 19/06/20. https://doi.org/10.1145/3408127.3408159

CNN Based Model for Malaria Diagnosis with Knowledge Distillation. / Fuhad, K. M.Faizullah; Tuba, Jannat Ferdousey; Rahman, Tanzilur et al.
ICDSP 2020 - 2020 4th International Conference on Digital Signal Processing, Proceedings. Association for Computing Machinery, 2020. p. 131-135 (ACM International Conference Proceeding Series).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - CNN Based Model for Malaria Diagnosis with Knowledge Distillation

AU - Fuhad, K. M.Faizullah

AU - Tuba, Jannat Ferdousey

AU - Rahman, Tanzilur

AU - Mohammed, Nabeel

PY - 2020/6/19

Y1 - 2020/6/19

N2 - Malaria is a deadly disease caused by plasmodium parasites and carried by female anopheles mosquitoes. Bangladesh is one of the major malaria prone country with poor infrastructures, unable to diagnose Malaria rapid at a low cost. Here we propose a custom8 layers CNN architecture that is only 233.60KB in size meaning it can easily fit and work into a low cost smartphone. The model has been optimized through different preprocessing methods and model pruning techniques like knowledge distillation ensuring that the accuracy does not reduce due to smaller model size. The accuracy received from the optimized model is~ 96.51% when tested on NIH data-set containing images of infected and uninfected cells. Proposed model is a big step towards a complete mobile based rapid diagnostic platform that can be used in any resource restricted and hilly areas of Bangladesh.

AB - Malaria is a deadly disease caused by plasmodium parasites and carried by female anopheles mosquitoes. Bangladesh is one of the major malaria prone country with poor infrastructures, unable to diagnose Malaria rapid at a low cost. Here we propose a custom8 layers CNN architecture that is only 233.60KB in size meaning it can easily fit and work into a low cost smartphone. The model has been optimized through different preprocessing methods and model pruning techniques like knowledge distillation ensuring that the accuracy does not reduce due to smaller model size. The accuracy received from the optimized model is~ 96.51% when tested on NIH data-set containing images of infected and uninfected cells. Proposed model is a big step towards a complete mobile based rapid diagnostic platform that can be used in any resource restricted and hilly areas of Bangladesh.

KW - Anopheles

KW - CNN

KW - Knowledge Distillation

KW - Malaria diagnostic system

KW - Model pruning

KW - Plasmodium

KW - Resource restricted

UR - https://www.scopus.com/pages/publications/85091569009

U2 - 10.1145/3408127.3408159

DO - 10.1145/3408127.3408159

M3 - Conference contribution

AN - SCOPUS:85091569009

T3 - ACM International Conference Proceeding Series

SP - 131

EP - 135

BT - ICDSP 2020 - 2020 4th International Conference on Digital Signal Processing, Proceedings

PB - Association for Computing Machinery

T2 - 4th International Conference on Digital Signal Processing, ICDSP 2020

Y2 - 19 June 2020 through 21 June 2020

ER -

CNN Based Model for Malaria Diagnosis with Knowledge Distillation

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Fingerprint

Cite this