Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans
Burn is a devastating injury affecting over eleven million people worldwide and more than 265,000 affected individuals lost their lives every year. Low- and middle-income countries (LMICs) have surging cases of more than 90% of the total global incidences due to poor socioeconomic conditions, lack o...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Applied System Innovation |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2571-5577/3/4/43 |
| _version_ | 1827704341038891008 |
|---|---|
| author | Aliyu Abubakar |
| author_facet | Aliyu Abubakar |
| author_sort | Aliyu Abubakar |
| collection | DOAJ |
| description | Burn is a devastating injury affecting over eleven million people worldwide and more than 265,000 affected individuals lost their lives every year. Low- and middle-income countries (LMICs) have surging cases of more than 90% of the total global incidences due to poor socioeconomic conditions, lack of preventive measures, reliance on subjective and inaccurate assessment techniques and lack of access to nearby hospitals. These factors necessitate the need for a better objective and cost-effective assessment technique that can be easily deployed in remote areas and hospitals where expertise and reliable burn evaluation is lacking. Therefore, this study proposes the use of Convolutional Neural Network (CNN) features along with different classification algorithms to discriminate between burnt and healthy skin using dataset from Black-African patients. A pretrained CNN model (VGG16) is used to extract abstract discriminatory image features and this approach was due to limited burn images which made it infeasible to train a CNN model from scratch. Subsequently, decision tree, support vector machines (SVM), naïve Bayes, logistic regression, and <i>k</i>-nearest neighbour (KNN) are used to classify whether a given image is burnt or healthy based on the VGG16 features. The performances of these classification algorithms were extensively analysed using the VGG16 features from different layers. |
| first_indexed | 2024-03-10T15:38:21Z |
| format | Article |
| id | doaj.art-41cf5694936441e79f18ac4c17afb6cd |
| institution | Directory Open Access Journal |
| issn | 2571-5577 |
| language | English |
| last_indexed | 2024-03-10T15:38:21Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied System Innovation |
| spelling | doaj.art-41cf5694936441e79f18ac4c17afb6cd2023-11-20T17:00:27ZengMDPI AGApplied System Innovation2571-55772020-10-01344310.3390/asi3040043Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black AfricansAliyu Abubakar0Centre for Visual Computing, Faculty of Engineering and Informatics, University of Bradford, Bradford BD7 1DP, UKBurn is a devastating injury affecting over eleven million people worldwide and more than 265,000 affected individuals lost their lives every year. Low- and middle-income countries (LMICs) have surging cases of more than 90% of the total global incidences due to poor socioeconomic conditions, lack of preventive measures, reliance on subjective and inaccurate assessment techniques and lack of access to nearby hospitals. These factors necessitate the need for a better objective and cost-effective assessment technique that can be easily deployed in remote areas and hospitals where expertise and reliable burn evaluation is lacking. Therefore, this study proposes the use of Convolutional Neural Network (CNN) features along with different classification algorithms to discriminate between burnt and healthy skin using dataset from Black-African patients. A pretrained CNN model (VGG16) is used to extract abstract discriminatory image features and this approach was due to limited burn images which made it infeasible to train a CNN model from scratch. Subsequently, decision tree, support vector machines (SVM), naïve Bayes, logistic regression, and <i>k</i>-nearest neighbour (KNN) are used to classify whether a given image is burnt or healthy based on the VGG16 features. The performances of these classification algorithms were extensively analysed using the VGG16 features from different layers.https://www.mdpi.com/2571-5577/3/4/43burnstransfer learningdecision treeSVMnaïve Bayeslogistic regression |
| spellingShingle | Aliyu Abubakar Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans Applied System Innovation burns transfer learning decision tree SVM naïve Bayes logistic regression |
| title | Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans |
| title_full | Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans |
| title_fullStr | Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans |
| title_full_unstemmed | Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans |
| title_short | Comparative Analysis of Classification Algorithms Using CNN Transferable Features: A Case Study Using Burn Datasets from Black Africans |
| title_sort | comparative analysis of classification algorithms using cnn transferable features a case study using burn datasets from black africans |
| topic | burns transfer learning decision tree SVM naïve Bayes logistic regression |
| url | https://www.mdpi.com/2571-5577/3/4/43 |
| work_keys_str_mv | AT aliyuabubakar comparativeanalysisofclassificationalgorithmsusingcnntransferablefeaturesacasestudyusingburndatasetsfromblackafricans |