Related Experiment Video
Updated: Jun 4, 2025
09:47
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
Published on: December 15, 2023
949
Classification of cervical cancer using Dense CapsNet with Seg-UNet and denoising autoencoders
Hui Yang1, Walid Aydi2,3, Nisreen Innab4
1Department of Critical Medicine, Baoshan People's Hospital, Baoshan, 678000, Yunnan Province, China. huiyangscientif@outlook.com.
Scientific Reports
|December 31, 2024
Summary
This study introduces a novel deep learning approach for cervical cancer detection, achieving 99.65% accuracy. The method enhances traditional Pap smear analysis by improving image segmentation and classification for early diagnosis.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Cervical cancer necessitates early detection through methods like Pap smear testing, which can be prone to human error.
- Existing machine learning and deep learning models struggle with accurate cervical cell segmentation and skewed data distributions.
Purpose of the Study:
- To develop an advanced deep learning framework for improved cervical cancer classification.
- To overcome limitations in current diagnostic methods, particularly segmentation difficulties and data imbalance.
Main Methods:
- A five-phase approach: pre-processing (contrast maximization), data augmentation (m-GAN), segmentation (Seg-UNet), feature extraction (denoising autoencoders), and classification (Dense CapsNet).
- Utilized the SIPaKMeD dataset for training and validation.
Main Results:
- The proposed system achieved a high classification accuracy of 99.65%.
- Demonstrated superior performance compared to existing literature methods.
Conclusions:
- The integrated deep learning model effectively addresses segmentation challenges and data skewness in cervical cancer classification.
- This approach offers a promising advancement for accurate and reliable early detection of cervical abnormalities.
