[Paper Review] Classification of Human Monkeypox Disease Using Deep Learning Models and Attention Mechanisms

Md Enamul Haque, Md. Rayhan Ahmed+2 more|arXiv (Cornell University)|Nov 21, 2022

Poxvirus research and outbreaks27 citations

TL;DR

The study integrates deep transfer-learning models with CBAM attention for image-based monkeypox classification, identifying Xception-CBAM-Dense as the best architecture with 83.89% validation accuracy.

ABSTRACT

As the world is still trying to rebuild from the destruction caused by the widespread reach of the COVID-19 virus, and the recent alarming surge of human monkeypox disease outbreaks in numerous countries threatens to become a new global pandemic too. Human monkeypox disease syndromes are quite similar to chickenpox, and measles classic symptoms, with very intricate differences such as skin blisters, which come in diverse forms. Various deep-learning methods have shown promising performances in the image-based diagnosis of COVID-19, tumor cell, and skin disease classification tasks. In this paper, we try to integrate deep transfer-learning-based methods, along with a convolutional block attention module (CBAM), to focus on the relevant portion of the feature maps to conduct an image-based classification of human monkeypox disease. We implement five deep-learning models, VGG19, Xception, DenseNet121, EfficientNetB3, and MobileNetV2, along with integrated channel and spatial attention mechanisms, and perform a comparative analysis among them. An architecture consisting of Xception-CBAM-Dense layers performed better than the other models at classifying human monkeypox and other diseases with a validation accuracy of 83.89%.

Motivation & Objective

• Motivate image-based diagnosis of human monkeypox and differentiate it from similar syndromes.
• Evaluate transfer-learning models for monkeypox classification using attention mechanisms.
• Compare five models with integrated channel and spatial attention to identify the best architecture.
• Assess the performance of CBAM-augmented networks on monkeypox detection tasks.

Proposed method

• Implement five deep-learning models: VGG19, Xception, DenseNet121, EfficientNetB3, and MobileNetV2.
• Integrate Convolutional Block Attention Module (CBAM) to focus on relevant feature maps.
• Conduct comparative analysis across models for monkeypox and other diseases using image data.
• Report validation accuracy and identify the best-performing architecture.

Experimental results

Research questions

• RQ1 Does integrating CBAM improve image-based monkeypox classification performance across multiple models?
• RQ2 Which model augmented with CBAM yields the highest validation accuracy for monkeypox detection?
• RQ3 How does the Xception-CBAM-Dense architecture compare to other tested models?
• RQ4 What is the overall effectiveness of transfer-learning approaches for monkeypox versus similar dermatological conditions?

Key findings

• Xception-CBAM-Dense architecture achieved the best performance with a validation accuracy of 83.89%.
• Among the tested models, CBAM-augmented networks are evaluated and compared for monkeypox classification.
• The study demonstrates the potential of deep transfer-learning approaches combined with attention mechanisms for disease image classification.