Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MpoxSLDNet: A Novel CNN Model for Detecting Monkeypox Lesions and Performance Comparison with Pre-trained Models

Fatema Jannat Dihan, Saydul Akbar Murad

TL;DR

The paper presents MpoxSLDNet, a lightweight CNN designed to detect monkeypox skin lesions and distinguish them from non-monkeypox lesions using the Monkeypox Skin Lesion Dataset (1428 Monkeypox, 1764 Non-Monkeypox). It details a six-convolutional, six-pooling architecture with dense layers, trained over 20 epochs and compared against pretrained models DenseNet121, ResNet50, and VGG16, achieving a validation accuracy of 94.56% and an AUC of 0.94. The study demonstrates MpoxSLDNet’s superior precision, recall, F1, and overall AUC relative to baselines, highlighting its suitability for resource-constrained settings due to reduced storage and computation needs. It emphasizes the model’s potential for rapid, non-invasive screening to aid early treatment and outbreak control, while noting dataset limitations and the need for broader validation and dataset expansion.

Abstract

Monkeypox virus (MPXV) is a zoonotic virus that poses a significant threat to public health, particularly in remote parts of Central and West Africa. Early detection of monkeypox lesions is crucial for effective treatment. However, due to its similarity with other skin diseases, monkeypox lesion detection is a challenging task. To detect monkeypox, many researchers used various deep-learning models such as MobileNetv2, VGG16, ResNet50, InceptionV3, DenseNet121, EfficientNetB3, MobileNetV2, and Xception. However, these models often require high storage space due to their large size. This study aims to improve the existing challenges by introducing a CNN model named MpoxSLDNet (Monkeypox Skin Lesion Detector Network) to facilitate early detection and categorization of Monkeypox lesions and Non-Monkeypox lesions in digital images. Our model represents a significant advancement in the field of monkeypox lesion detection by offering superior performance metrics, including precision, recall, F1-score, accuracy, and AUC, compared to traditional pre-trained models such as VGG16, ResNet50, and DenseNet121. The key novelty of our approach lies in MpoxSLDNet's ability to achieve high detection accuracy while requiring significantly less storage space than existing models. By addressing the challenge of high storage requirements, MpoxSLDNet presents a practical solution for early detection and categorization of monkeypox lesions in resource-constrained healthcare settings. In this study, we have used "Monkeypox Skin Lesion Dataset" comprising 1428 skin images of monkeypox lesions and 1764 skin images of Non-Monkeypox lesions. Dataset's limitations could potentially impact the model's ability to generalize to unseen cases. However, the MpoxSLDNet model achieved a validation accuracy of 94.56%, compared to 86.25%, 84.38%, and 67.19% for VGG16, DenseNet121, and ResNet50, respectively.

MpoxSLDNet: A Novel CNN Model for Detecting Monkeypox Lesions and Performance Comparison with Pre-trained Models

TL;DR

The paper presents MpoxSLDNet, a lightweight CNN designed to detect monkeypox skin lesions and distinguish them from non-monkeypox lesions using the Monkeypox Skin Lesion Dataset (1428 Monkeypox, 1764 Non-Monkeypox). It details a six-convolutional, six-pooling architecture with dense layers, trained over 20 epochs and compared against pretrained models DenseNet121, ResNet50, and VGG16, achieving a validation accuracy of 94.56% and an AUC of 0.94. The study demonstrates MpoxSLDNet’s superior precision, recall, F1, and overall AUC relative to baselines, highlighting its suitability for resource-constrained settings due to reduced storage and computation needs. It emphasizes the model’s potential for rapid, non-invasive screening to aid early treatment and outbreak control, while noting dataset limitations and the need for broader validation and dataset expansion.

Abstract

Monkeypox virus (MPXV) is a zoonotic virus that poses a significant threat to public health, particularly in remote parts of Central and West Africa. Early detection of monkeypox lesions is crucial for effective treatment. However, due to its similarity with other skin diseases, monkeypox lesion detection is a challenging task. To detect monkeypox, many researchers used various deep-learning models such as MobileNetv2, VGG16, ResNet50, InceptionV3, DenseNet121, EfficientNetB3, MobileNetV2, and Xception. However, these models often require high storage space due to their large size. This study aims to improve the existing challenges by introducing a CNN model named MpoxSLDNet (Monkeypox Skin Lesion Detector Network) to facilitate early detection and categorization of Monkeypox lesions and Non-Monkeypox lesions in digital images. Our model represents a significant advancement in the field of monkeypox lesion detection by offering superior performance metrics, including precision, recall, F1-score, accuracy, and AUC, compared to traditional pre-trained models such as VGG16, ResNet50, and DenseNet121. The key novelty of our approach lies in MpoxSLDNet's ability to achieve high detection accuracy while requiring significantly less storage space than existing models. By addressing the challenge of high storage requirements, MpoxSLDNet presents a practical solution for early detection and categorization of monkeypox lesions in resource-constrained healthcare settings. In this study, we have used "Monkeypox Skin Lesion Dataset" comprising 1428 skin images of monkeypox lesions and 1764 skin images of Non-Monkeypox lesions. Dataset's limitations could potentially impact the model's ability to generalize to unseen cases. However, the MpoxSLDNet model achieved a validation accuracy of 94.56%, compared to 86.25%, 84.38%, and 67.19% for VGG16, DenseNet121, and ResNet50, respectively.
Paper Structure (24 sections, 6 equations, 17 figures, 10 tables)

This paper contains 24 sections, 6 equations, 17 figures, 10 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Methodology
  4. Data Collection
  5. Data Pre-processing
  6. Resizing Image
  7. Rescale:
  8. Data_format:
  9. Label Encoding:
  10. Data Augmentation (in-place):
  11. Data Visualization:
  12. Splitting Dataset
  13. Model Building
  14. Architecture of our Proposed MpoxSLDNet model :
  15. Result Analysis
  16. ...and 9 more sections

Figures (17)

  • Figure 1: Navigating Monkeypox Lesion Detection: An Introduction Overview.
  • Figure 2: Overall Workflow diagram.
  • Figure 3: Distribution of Monkeypox lesion and Non-Monkeypox lesion images in the dataset.
  • Figure 4: Sample set of images from dataset.
  • Figure 5: Image Augmentation (in-place ) process.
  • ...and 12 more figures