A A Study on the Effectiveness of Fine-Tuning VGG16 Versus Classical Machine Learning and Deep Learning Models for Skin Burn Detection

Authors

  • Muhammad Ashraf Gulab Devi Teaching Hospital Lahore
  • Hafiz MuhammadAfzaal Department of Computer Science & Information Technology, The Superior University Lahore
  • Muhammad Azam Department of Computer Science & Information Technology, The Superior University Lahore
  • Muhammad Ashraf IT Department Gulab Devi Teaching Hospital ,Lahore https://orcid.org/0009-0008-2567-9984
  • AbdulRasheed Department of Molecular Neuroscience University of Toyama, Japan
  • Tahir Akram Tech Delivery & Ops Excellence, Accenture Ltd, Saudi Arabia

DOI:

https://doi.org/10.56536/jicet.v4i2.125

Abstract

Abstract: Accurate skin burn classification is essential for successful medical diagnosis and treatment. This study focuses on the skin burn dataset of 1000 training images and 150 validation images where recognition performance of various classical machine learning algorithms as well as deep learning models has been analyzed. The classical models of deep learning that have been covered in this paper include, namely, Support Vector Machine (SVM), Random Forest (RF), K- Nearest Neighbors (KNN), and Gradient Boosting Machines (GBM). Also , we evaluate the performance of Convolutional Neural Networks (CNNs) and advanced transfer learning techniques using VGG16 for image recognition, classification and segmentation.

The  result  appear convincing although classical  machine learning methods offered valuable insights, their performance was undermined by deep learning approaches. In particular, the basic CNN model demonstrated average accuracy, but it was the fine-tuning of the VGG16 model that achieved the highest validation accuracy of 82.67%. The fine-tuned VGG16 model significantly outperformed classical models which is easy to fit  large dataset for image classification. The fine-tuned VGG16 model which had accuracy scores increase from 65.33% to 73.33%, and hybrid models showed different degrees of improvement.

The study also demonstrates how fine-tuned deep learning models are superior to some approaches in machine learning and offers detailed comparisons of various classification methodologies. The foundations of this study are laid down with the aim of increasing the efficacy of the automatic diagnosis of skin burn analysis and improving the efficiency of medical image analysis.

Author Biographies

Hafiz MuhammadAfzaal, Department of Computer Science & Information Technology, The Superior University Lahore

Student of MSCS  in Department of Computer Science & Information Technology,

 The Superior University Lahore

Muhammad Azam, Department of Computer Science & Information Technology, The Superior University Lahore

Dr Muhammad Azam

HOD  Department of Computer Science & Information Technology, The Superior University Lahore

 

Muhammad Ashraf, IT Department Gulab Devi Teaching Hospital ,Lahore

IT Incharge, Gulab Devi Teaching Hospital, Lahore

Student of MS CS in Superior University Lahore 

Tahir Akram, Tech Delivery & Ops Excellence, Accenture Ltd, Saudi Arabia

Manager 

Tech Delivery & Ops Excellence, Accenture Ltd, Saudi Arabia

Downloads

Published

2024-10-03