No Cover Image

Journal article 43 views 8 downloads

A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations

Maryam Bukhari Orcid Logo, Sadaf Yasmin Orcid Logo, Adnan Habib Orcid Logo, Cheng Cheng Orcid Logo, Farhan Ullah Orcid Logo, Jaeseok Yoo Orcid Logo, Daewon Lee Orcid Logo

Journal of Healthcare Engineering, Volume: 2023, Pages: 1 - 15

Swansea University Author: Cheng Cheng Orcid Logo

  • 65951.VoR.pdf

    PDF | Version of Record

    Copyright © 2023 Maryam Bukhari et al. This is an open access article distributed under the Creative Commons Attribution License.

    Download (4.37MB)

Check full text

DOI (Published version): 10.1155/2023/1847115

Abstract

Skin cancer remains one of the deadliest kinds of cancer, with a survival rate of about 18–20%. Early diagnosis and segmentation of the most lethal kind of cancer, melanoma, is a challenging and critical task. To diagnose medicinal conditions of melanoma lesions, different researchers proposed autom...

Full description

Published in: Journal of Healthcare Engineering
ISSN: 2040-2295 2040-2309
Published: Hindawi Limited 2023
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa65951
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2024-05-29T14:00:28Z
last_indexed 2024-05-29T14:00:28Z
id cronfa65951
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>65951</id><entry>2024-04-03</entry><title>A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations</title><swanseaauthors><author><sid>11ddf61c123b99e59b00fa1479367582</sid><ORCID>0000-0003-0371-9646</ORCID><firstname>Cheng</firstname><surname>Cheng</surname><name>Cheng Cheng</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2024-04-03</date><deptcode>MACS</deptcode><abstract>Skin cancer remains one of the deadliest kinds of cancer, with a survival rate of about 18–20%. Early diagnosis and segmentation of the most lethal kind of cancer, melanoma, is a challenging and critical task. To diagnose medicinal conditions of melanoma lesions, different researchers proposed automatic and traditional approaches to accurately segment the lesions. However, visual similarity among lesions and intraclass differences are very high, which leads to low-performance accuracy. Furthermore, traditional segmentation algorithms often require human inputs and cannot be utilized in automated systems. To address all of these issues, we provide an improved segmentation model based on depthwise separable convolutions that act on each spatial dimension of the image to segment the lesions. The fundamental idea behind these convolutions is to divide the feature learning steps into two simpler parts that are spatial learning of features and a step for channel combination. Besides this, we employ parallel multidilated filters to encode multiple parallel features and broaden the view of filters with dilations. Moreover, for performance evaluation, the proposed approach is evaluated on three different datasets including DermIS, DermQuest, and ISIC2016. The finding indicates that the suggested segmentation model has achieved the Dice score of 97% for DermIS and DermQuest and 94.7% for the ISBI2016 dataset, respectively.</abstract><type>Journal Article</type><journal>Journal of Healthcare Engineering</journal><volume>2023</volume><journalNumber/><paginationStart>1</paginationStart><paginationEnd>15</paginationEnd><publisher>Hindawi Limited</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>2040-2295</issnPrint><issnElectronic>2040-2309</issnElectronic><keywords/><publishedDay>6</publishedDay><publishedMonth>2</publishedMonth><publishedYear>2023</publishedYear><publishedDate>2023-02-06</publishedDate><doi>10.1155/2023/1847115</doi><url/><notes/><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm>Another institution paid the OA fee</apcterm><funders>his work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (No. 2021R1G1A1095460) and also by the Chung-Ang University, Research Scholarship Grants in 2021.</funders><projectreference/><lastEdited>2024-05-29T15:02:19.1365894</lastEdited><Created>2024-04-03T17:44:51.6481101</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>Maryam</firstname><surname>Bukhari</surname><orcid>0000-0003-2904-2223</orcid><order>1</order></author><author><firstname>Sadaf</firstname><surname>Yasmin</surname><orcid>0000-0003-3756-4865</orcid><order>2</order></author><author><firstname>Adnan</firstname><surname>Habib</surname><orcid>0000-0002-0778-259x</orcid><order>3</order></author><author><firstname>Cheng</firstname><surname>Cheng</surname><orcid>0000-0003-0371-9646</orcid><order>4</order></author><author><firstname>Farhan</firstname><surname>Ullah</surname><orcid>0000-0002-1030-1275</orcid><order>5</order></author><author><firstname>Jaeseok</firstname><surname>Yoo</surname><orcid>0000-0002-7204-1157</orcid><order>6</order></author><author><firstname>Daewon</firstname><surname>Lee</surname><orcid>0000-0002-3004-2901</orcid><order>7</order></author></authors><documents><document><filename>65951__30481__df2d5ea4d0574eb687b301c060cd198d.pdf</filename><originalFilename>65951.VoR.pdf</originalFilename><uploaded>2024-05-29T15:01:04.8984479</uploaded><type>Output</type><contentLength>4582131</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>Copyright © 2023 Maryam Bukhari et al. This is an open access article distributed under the Creative Commons Attribution License.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling v2 65951 2024-04-03 A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations 11ddf61c123b99e59b00fa1479367582 0000-0003-0371-9646 Cheng Cheng Cheng Cheng true false 2024-04-03 MACS Skin cancer remains one of the deadliest kinds of cancer, with a survival rate of about 18–20%. Early diagnosis and segmentation of the most lethal kind of cancer, melanoma, is a challenging and critical task. To diagnose medicinal conditions of melanoma lesions, different researchers proposed automatic and traditional approaches to accurately segment the lesions. However, visual similarity among lesions and intraclass differences are very high, which leads to low-performance accuracy. Furthermore, traditional segmentation algorithms often require human inputs and cannot be utilized in automated systems. To address all of these issues, we provide an improved segmentation model based on depthwise separable convolutions that act on each spatial dimension of the image to segment the lesions. The fundamental idea behind these convolutions is to divide the feature learning steps into two simpler parts that are spatial learning of features and a step for channel combination. Besides this, we employ parallel multidilated filters to encode multiple parallel features and broaden the view of filters with dilations. Moreover, for performance evaluation, the proposed approach is evaluated on three different datasets including DermIS, DermQuest, and ISIC2016. The finding indicates that the suggested segmentation model has achieved the Dice score of 97% for DermIS and DermQuest and 94.7% for the ISBI2016 dataset, respectively. Journal Article Journal of Healthcare Engineering 2023 1 15 Hindawi Limited 2040-2295 2040-2309 6 2 2023 2023-02-06 10.1155/2023/1847115 COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University Another institution paid the OA fee his work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea Government (MSIT) (No. 2021R1G1A1095460) and also by the Chung-Ang University, Research Scholarship Grants in 2021. 2024-05-29T15:02:19.1365894 2024-04-03T17:44:51.6481101 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science Maryam Bukhari 0000-0003-2904-2223 1 Sadaf Yasmin 0000-0003-3756-4865 2 Adnan Habib 0000-0002-0778-259x 3 Cheng Cheng 0000-0003-0371-9646 4 Farhan Ullah 0000-0002-1030-1275 5 Jaeseok Yoo 0000-0002-7204-1157 6 Daewon Lee 0000-0002-3004-2901 7 65951__30481__df2d5ea4d0574eb687b301c060cd198d.pdf 65951.VoR.pdf 2024-05-29T15:01:04.8984479 Output 4582131 application/pdf Version of Record true Copyright © 2023 Maryam Bukhari et al. This is an open access article distributed under the Creative Commons Attribution License. true eng https://creativecommons.org/licenses/by/4.0/
title A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
spellingShingle A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
Cheng Cheng
title_short A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
title_full A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
title_fullStr A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
title_full_unstemmed A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
title_sort A Novel Framework for Melanoma Lesion Segmentation Using Multiparallel Depthwise Separable and Dilated Convolutions with Swish Activations
author_id_str_mv 11ddf61c123b99e59b00fa1479367582
author_id_fullname_str_mv 11ddf61c123b99e59b00fa1479367582_***_Cheng Cheng
author Cheng Cheng
author2 Maryam Bukhari
Sadaf Yasmin
Adnan Habib
Cheng Cheng
Farhan Ullah
Jaeseok Yoo
Daewon Lee
format Journal article
container_title Journal of Healthcare Engineering
container_volume 2023
container_start_page 1
publishDate 2023
institution Swansea University
issn 2040-2295
2040-2309
doi_str_mv 10.1155/2023/1847115
publisher Hindawi Limited
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Skin cancer remains one of the deadliest kinds of cancer, with a survival rate of about 18–20%. Early diagnosis and segmentation of the most lethal kind of cancer, melanoma, is a challenging and critical task. To diagnose medicinal conditions of melanoma lesions, different researchers proposed automatic and traditional approaches to accurately segment the lesions. However, visual similarity among lesions and intraclass differences are very high, which leads to low-performance accuracy. Furthermore, traditional segmentation algorithms often require human inputs and cannot be utilized in automated systems. To address all of these issues, we provide an improved segmentation model based on depthwise separable convolutions that act on each spatial dimension of the image to segment the lesions. The fundamental idea behind these convolutions is to divide the feature learning steps into two simpler parts that are spatial learning of features and a step for channel combination. Besides this, we employ parallel multidilated filters to encode multiple parallel features and broaden the view of filters with dilations. Moreover, for performance evaluation, the proposed approach is evaluated on three different datasets including DermIS, DermQuest, and ISIC2016. The finding indicates that the suggested segmentation model has achieved the Dice score of 97% for DermIS and DermQuest and 94.7% for the ISBI2016 dataset, respectively.
published_date 2023-02-06T15:02:18Z
_version_ 1800395909404557312
score 11.012678

Similar Items