UNET-Based Deep Learning Model for Liver Cirrhosis Stage Classification and Tumor Size Detection

DOI

10.2991/978-94-6463-738-0_26How to use a DOI?

Keywords

UNET-based models; deep learning; hybrid models; liver medical images; image segmentation; liver stage classification; tumor size prediction; CT images

Abstract

The liver is one of the vital organs in the human body, playing a crucial role in metabolism, maintaining blood sugar levels, aiding digestion, and storing vitamins and minerals. It is the largest organ, weighing about 3 pounds and roughly the size of a football. However, liver diseases such as hepatitis, fibrosis, cirrhosis, and liver failure pose significant health challenges. This study focuses on liver disease stage prediction using machine learning and hybrid deep learning models. Techniques like supervised learning, clustering, and random forest were compared, with random forest achieving 98% accuracy. The dataset, sourced from platforms like Kaggle, includes numerical data and CT images for tumor detection and chronic liver disease analysis. The proposed system integrates UNET-based models for precise liver segmentation and hybrid techniques combining CNN and LSTM algorithms. These hybrid models accurately classify liver disease stages and measure tumor size. The system categorizes liver stages into four levels using clinical datasets and achieves a tumor detection accuracy of 96%.While the CNN+LSTM integration is computationally complex, it significantly enhances early detection of cirrhosis, a critical stage of liver damage, enabling timely intervention. The hybrid model demonstrates a precision of 91%, recall of 90%, and an F1 score of 90%, providing robust results for chronic liver disease management. This approach ensures earlier diagnosis, improved patient outcomes, and enhanced confidence in individual health assessments.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - K. Kalaiselvi
AU  - S. Priyadharshini
AU  - B. Hariram
PY  - 2025
DA  - 2025/06/22
TI  - UNET-Based Deep Learning Model for Liver Cirrhosis Stage Classification and Tumor Size Detection
BT  - Proceedings of the International Conference on Advances and Applications in Artificial Intelligence (ICAAAI 2025)
PB  - Atlantis Press
SP  - 309
EP  - 318
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-738-0_26
DO  - 10.2991/978-94-6463-738-0_26
ID  - Kalaiselvi2025
ER  -