PoXeptionNet: Accelerating Pox Disease Identification Using a Dual Neural Network Fusing Xception and EfficientNetB0 Architectures
Objective
This project aims to develop two hybrid deep learning models, PoXeptionNet-Attn and PoxNetPlus_DensSwi, by combining Xception, EfficientNetB0, DenseNet121, and Swin Transformer architectures to extract and refine features from skin lesion images for accurate disease classification. The project involves preparing a dataset of Chickenpox, Cowpox, HFMD, Measles, Monkeypox, and Healthy skin lesions through preprocessing steps like resizing, normalization, and augmentation. The models will be trained and optimized using techniques such as learning rate adjustment, early stopping, and model checkpointing to ensure high performance and prevent overfitting. After training, the models will be evaluated using accuracy, confusion matrix, and classification report to measure their classification ability. A web-based application will be developed using the Flask framework, allowing healthcare professionals to upload skin lesion images for disease classification. The system will be scalable to handle large datasets and user-friendly for healthcare professionals to interact with easily. Future work will focus on improving the models by integrating additional disease classification models, fine-tuning existing ones, and incorporating explainable AI techniques for better interpretability of predictions.
Abstract
This project focuses on automating the identification and classification of pox diseases from skin lesion images using deep learning. It integrates two powerful models, PoXeptionNet-Attn and PoxNetPlus_DensSwi, to improve the accuracy and efficiency of disease detection. PoXeptionNet-Attn fuses the Xception and EfficientNetB0 architectures with an attention mechanism, enabling the model to focus on important features in skin lesions. PoxNetPlus_DensSwi, on the other hand, combines DenseNet121 and Swin Transformer to capture both global and local features, refining the model's feature extraction process. The dataset used for this project includes images of Chickenpox, Cowpox, HFMD, Measles, Monkeypox, and Healthy skin lesions. The system is developed using the Flask framework, allowing it to be deployed as a web-based application for healthcare professionals. By automating the disease identification process, this project aims to reduce the reliance on manual examination, improving diagnosis speed and accuracy. The results of this system are evaluated using standard metrics such as accuracy, confusion matrix, and classification report.
Keywords: PoXeptionNet, deep learning, Xception, EfficientNetB0, DenseNet121, Swin Transformer, disease classification, skin lesions, machine learning, healthcare automation.
NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.
Block Diagram
Specifications
HARDWARE REQUIREMENTS
• Processor - I5/Intel Processor
• RAM - 8GB (min)
• Hard Disk - 160 GB
• Key Board - Standard Windows Keyboard
• Mouse - Two or Three Button Mouse
• Monitor - Any
SOFTWARE REQUIREMENS
• Operating System : Windows 7/8/10
• Server side Script : HTML, CSS, Bootstrap & JS
• Programming Language : Python
• Libraries : Flask, Pandas, Mysql.connector, Os, Numpy,
Scikit-learn.
• IDE/Workbench : VS-Code
• Technology : Python 3.10+
• Server Deployment : Xampp Server
• Database : MySQL
Paper Publishing
Request Call Back
Would you like to receive a free callback now?
Choose the best time for callback:
Leave your message and we will contact you as soon as possible
6-2-85/B, Old Maternity Hospital Road, Thyagaraja Nagar, Tirupati, Andhra Pradesh – 517501
+91 9030333433
+91 9393939065
0877-2261612
Disclaimer - Takeoff Edu Group Projects are not associated or affiliated with IEEE in any way. The IEEE Projects mentioned here are mentioned in the context of student projects, whose ideas are derived from IEEE publications, not projects of or by IEEE.
