Cite Article

A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth

Choose citation format

BibTeX

@article{IJASEIT10227,
   author = {P.N. Kuan and S. Chua and E.B. Safawi and H.H. Wang},
   title = {A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth},
   journal = {International Journal on Advanced Science, Engineering and Information Technology},
   volume = {10},
   number = {1},
   year = {2020},
   pages = {145--150},
   keywords = {skin burn depth; burn images; classification; segmentation; image mining approach.},
   abstract = {

A correct first assessment of a skin burn depth is essential as it determines a correct first burn treatment provided to the patients. The objective of this paper is to conduct a comparative study of the different segmentation algorithms for the classification of different burn depths. Eight different hybrid segmentation algorithms were studied on a skin burn dataset comprising skin burn images categorized into three burn classes by medical experts; superficial partial thickness burn (SPTB), deep partial thickness burn (DPTB) and full thickness burn (FTB). Different sequences of the algorithm were experimented as each algorithm was able to segment differently, leading to different segmentation in the final output. The performance of the segmentation algorithms was evaluated by calculating the number of correctly segmented images for each burn depth. The empirical results showed that the segmentation algorithm that was able to segment most of the burn depths had achieved 40.24%, 60.42% and 6.25% of correctly segmented image for SPTB, DPTB and FTB respectively. Most of the segmentation algorithms could not segment well for FTB images because of the different nature of the burn wounds as some of the FTB images contained dark brown and black colors. It can be concluded that a good segmentation algorithm is required to ensure that the representative features of each burn depth can be extracted to contribute to higher accuracy of classification of skin burn depth.

},    issn = {2088-5334},    publisher = {INSIGHT - Indonesian Society for Knowledge and Human Development},    url = {http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=10227},    doi = {10.18517/ijaseit.10.1.10227} }

EndNote

%A Kuan, P.N.
%A Chua, S.
%A Safawi, E.B.
%A Wang, H.H.
%D 2020
%T A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth
%B 2020
%9 skin burn depth; burn images; classification; segmentation; image mining approach.
%! A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth
%K skin burn depth; burn images; classification; segmentation; image mining approach.
%X 

A correct first assessment of a skin burn depth is essential as it determines a correct first burn treatment provided to the patients. The objective of this paper is to conduct a comparative study of the different segmentation algorithms for the classification of different burn depths. Eight different hybrid segmentation algorithms were studied on a skin burn dataset comprising skin burn images categorized into three burn classes by medical experts; superficial partial thickness burn (SPTB), deep partial thickness burn (DPTB) and full thickness burn (FTB). Different sequences of the algorithm were experimented as each algorithm was able to segment differently, leading to different segmentation in the final output. The performance of the segmentation algorithms was evaluated by calculating the number of correctly segmented images for each burn depth. The empirical results showed that the segmentation algorithm that was able to segment most of the burn depths had achieved 40.24%, 60.42% and 6.25% of correctly segmented image for SPTB, DPTB and FTB respectively. Most of the segmentation algorithms could not segment well for FTB images because of the different nature of the burn wounds as some of the FTB images contained dark brown and black colors. It can be concluded that a good segmentation algorithm is required to ensure that the representative features of each burn depth can be extracted to contribute to higher accuracy of classification of skin burn depth.

%U http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=10227 %R doi:10.18517/ijaseit.10.1.10227 %J International Journal on Advanced Science, Engineering and Information Technology %V 10 %N 1 %@ 2088-5334

IEEE

P.N. Kuan,S. Chua,E.B. Safawi and H.H. Wang,"A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth," International Journal on Advanced Science, Engineering and Information Technology, vol. 10, no. 1, pp. 145-150, 2020. [Online]. Available: http://dx.doi.org/10.18517/ijaseit.10.1.10227.

RefMan/ProCite (RIS)

TY  - JOUR
AU  - Kuan, P.N.
AU  - Chua, S.
AU  - Safawi, E.B.
AU  - Wang, H.H.
PY  - 2020
TI  - A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth
JF  - International Journal on Advanced Science, Engineering and Information Technology; Vol. 10 (2020) No. 1
Y2  - 2020
SP  - 145
EP  - 150
SN  - 2088-5334
PB  - INSIGHT - Indonesian Society for Knowledge and Human Development
KW  - skin burn depth; burn images; classification; segmentation; image mining approach.
N2  - 

A correct first assessment of a skin burn depth is essential as it determines a correct first burn treatment provided to the patients. The objective of this paper is to conduct a comparative study of the different segmentation algorithms for the classification of different burn depths. Eight different hybrid segmentation algorithms were studied on a skin burn dataset comprising skin burn images categorized into three burn classes by medical experts; superficial partial thickness burn (SPTB), deep partial thickness burn (DPTB) and full thickness burn (FTB). Different sequences of the algorithm were experimented as each algorithm was able to segment differently, leading to different segmentation in the final output. The performance of the segmentation algorithms was evaluated by calculating the number of correctly segmented images for each burn depth. The empirical results showed that the segmentation algorithm that was able to segment most of the burn depths had achieved 40.24%, 60.42% and 6.25% of correctly segmented image for SPTB, DPTB and FTB respectively. Most of the segmentation algorithms could not segment well for FTB images because of the different nature of the burn wounds as some of the FTB images contained dark brown and black colors. It can be concluded that a good segmentation algorithm is required to ensure that the representative features of each burn depth can be extracted to contribute to higher accuracy of classification of skin burn depth.

UR - http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=10227 DO - 10.18517/ijaseit.10.1.10227

RefWorks

RT Journal Article
ID 10227
A1 Kuan, P.N.
A1 Chua, S.
A1 Safawi, E.B.
A1 Wang, H.H.
T1 A Comparative Study of Segmentation Algorithms in the Classification of Human Skin Burn Depth
JF International Journal on Advanced Science, Engineering and Information Technology
VO 10
IS 1
YR 2020
SP 145
OP 150
SN 2088-5334
PB INSIGHT - Indonesian Society for Knowledge and Human Development
K1 skin burn depth; burn images; classification; segmentation; image mining approach.
AB 

A correct first assessment of a skin burn depth is essential as it determines a correct first burn treatment provided to the patients. The objective of this paper is to conduct a comparative study of the different segmentation algorithms for the classification of different burn depths. Eight different hybrid segmentation algorithms were studied on a skin burn dataset comprising skin burn images categorized into three burn classes by medical experts; superficial partial thickness burn (SPTB), deep partial thickness burn (DPTB) and full thickness burn (FTB). Different sequences of the algorithm were experimented as each algorithm was able to segment differently, leading to different segmentation in the final output. The performance of the segmentation algorithms was evaluated by calculating the number of correctly segmented images for each burn depth. The empirical results showed that the segmentation algorithm that was able to segment most of the burn depths had achieved 40.24%, 60.42% and 6.25% of correctly segmented image for SPTB, DPTB and FTB respectively. Most of the segmentation algorithms could not segment well for FTB images because of the different nature of the burn wounds as some of the FTB images contained dark brown and black colors. It can be concluded that a good segmentation algorithm is required to ensure that the representative features of each burn depth can be extracted to contribute to higher accuracy of classification of skin burn depth.

LK http://ijaseit.insightsociety.org/index.php?option=com_content&view=article&id=9&Itemid=1&article_id=10227 DO - 10.18517/ijaseit.10.1.10227