Abstract:
"Monkeypox is a rare but potentially fatal viral disease that affects both humans and animals. The
lack of an accurate and efficient detection system has made it difficult for public health
professionals to diagnose and treat the disease in a timely manner. The study will explore the
potential of using images of skin lesions, which are a key diagnostic feature of monkeypox, to
develop predictive models that can aid in early diagnosis and treatment of the disease. To address
this problem, we propose a deep learning-based approach that utilizes Convolutional Neural
Networks (CNNs) to classify Monkeypox images.
Our proposed Monkeypox detection system consists of a custom CNN architecture that
consists of multiple convolutional and pooling layers to extract and learn relevant features from
the input images. We also employed transfer learning to improve the model's accuracy and reduce
the risk of overfitting. The CNN was trained on a large dataset of Monkeypox images, which were
preprocessed using various techniques, such as image normalization and data augmentation.
In our experiments, we evaluated the performance of our proposed Monkeypox detection
system using various data science metrics, such as accuracy, precision, recall, and F1 score. Our
results demonstrate that the proposed CNN-based approach achieved high levels of accuracy and
outperformed existing state-of-the-art methods. These results demonstrate that the CNN model
achieved high accuracy and sensitivity in predicting monkeypox from skin lesion images.
Specifically, the CNN model achieved an overall accuracy of 93%, sensitivity of 93.3%, and
specificity of 93%. Overall, our proposed Monkeypox detection system has the potential to help
public health professionals diagnose and treat Monkeypox in a timely and accurate manner,
particularly in resource-limited settings where traditional diagnostic methods may be unavailable
or costly."
