RadDenseNet: An Explainable AI Model for Detection of Hepatic Steatosis from Abdominal CT Images
Journal of Electronics, Electromedical Engineering, and Medical Informatics · 17.10.2026 · 8 (4) · 1555-1570
DOI: 10.35882/jeeemi.v8i4.1688 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Electronics, Electromedical Engineering, and Medical Informatics, online erschienen am 17.10.2026. Verfasst von Rahul LR und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Hepatic steatosis affects approximately 30.2% of the global population and, if undetected, can silently progress to non-alcoholic steatohepatitis, cirrhosis, and hepatocellular carcinoma. Manual CT interpretation for fatty liver assessment is time-consuming and prone to inter-observer variability, motivating research into automated screening tools. We propose RadDenseNet, a DenseNet121 model initialized with RadImageNet weights pretrained on 1.35 million medical images spanning CT, MRI, and ultrasound, and evaluate it as a research prototype. The first 397 of 427 backbone layers were frozen, and the remaining 30 layers were fine-tuned alongside a classification head comprising global average pooling, a dense layer of 256 units, batch normalization, and dropout of 0.5. The model was trained on the Kaggle Liver CT Image Dataset - 3,557 abdominal CT images across Healthy and Hepatic Steatosis classes - using a stratified 70/20/10 split with 5-fold cross-validation. RadDenseNet was compared against six baseline models spanning ImageNet and RadImageNet pretraining across ResNet50, InceptionV3, and DenseNet121 architectures. The decision threshold was optimized on the validation set, yielding τ = 0.48. On the held-out test set, RadDenseNet achieved 96.35% accuracy, AUC of 0.9966, sensitivity of 0.9538, specificity of 0.9752, precision of 0.9789, and F1-score of 0.9662, with 5-fold cross-validation confirming stable performance (mean accuracy 96.46% ± 0.68%, mean AUC 0.9962 ± 0.0014). RadImageNet pretraining consistently outperformed ImageNet pretraining across all architectures, with accuracy improvements of 2.81 to 5.90 percentage points. Grad-CAM analysis covering all four prediction categories revealed that false positives arose from extra-hepatic attention and false negatives from weak parenchymal activation - clinically meaningful failure modes. External validation on multi-center datasets, extension to multi-class steatosis grading, and liver ROI segmentation as preprocessing are recommended as future directions
Zitieren
APA
LR, R., & Gaddamedi, S. P. (2026). RadDenseNet: An Explainable AI Model for Detection of Hepatic Steatosis from Abdominal CT Images. Journal of Electronics, Electromedical Engineering, and Medical Informatics, 8(4), 1555-1570. https://doi.org/10.35882/jeeemi.v8i4.1688
Vancouver
LR R, Gaddamedi SP. RadDenseNet: An Explainable AI Model for Detection of Hepatic Steatosis from Abdominal CT Images. Journal of Electronics, Electromedical Engineering, and Medical Informatics. 2026;8(4):1555-1570. doi:10.35882/jeeemi.v8i4.1688
BibTeX
@article{LR2026,
author = {LR, Rahul and Gaddamedi, Satya Prasad},
title = {RadDenseNet: An Explainable AI Model for Detection of Hepatic Steatosis from Abdominal CT Images},
journal = {Journal of Electronics, Electromedical Engineering, and Medical Informatics},
year = {2026},
volume = {8},
number = {4},
pages = {1555--1570},
doi = {10.35882/jeeemi.v8i4.1688}
}
EUVIMED