Laplacian sharpening achieves the best performance within the unified MSF-TEA Net framework, with a test accuracy of 94.40% ± 1.13% and an AUC of 99.30%, outperforming the other enhancement strategies.
Abstract
Automated identification of urinary system diseases on non-contrast computed tomography (NCCT) can facilitate early diagnosis and inform treatment decisions. However, kidney stones, hydronephrosis, and pyonephrosis share overlapping NCCT appearances and require diagnostic cues at different spatial scales, making accurate identification within a unified framework challenging. To address this, we constructed an integrated CT-based framework for automated identification of urinary system diseases. First, the collected CT data were split into training, validation, and test sets and processed using a unified preprocessing pipeline for resolution standardization and normalization. Subsequently, we evaluated multiple image enhancement strategies, including histogram equalization, CLAHE, Laplacian sharpening, and brightness enhancement. We conducted a systematic comparison of these enhancement methods to quantify their effects on disease identification performance. To support unified automated diagnosis of kidney stones, hydronephrosis, and pyonephrosis, we developed a multi-disease renal diagnostic network, MSF-TEA Net. Through multi-scale feature fusion and multi-evidence collaborative modeling, it effectively represents the image features of different pathological states. This model introduces Tri-Evidence Attention (TEA), which models the high-density features of small calculi targets, the global morphological changes of fluid accumulation, and the inflammatory texture features of pus accumulation. Through adaptive weighted fusion, the network enhances discrimination and diagnostic performance when complex lesions coexist. Experimental results show that, within the unified MSF-TEA Net framework, Laplacian sharpening achieves the best performance among the tested enhancement methods, with a test accuracy of 94.40% ± 1.13% and an AUC of 99.30%, outperforming the other enhancement strategies. Ablation studies further confirm the contributions of PPM, ASPP, and TEA. Overall, the proposed unified framework enables reliable identification of kidney stones, hydronephrosis, and pyonephrosis, supporting clinical decision-making.
Kidney stone disease is a common urological disease, which impacts a significant proportion of the population across the globe. It should be diagnosed early and properly because in the cases that are not properly dealt with, the complications may include obstruction of urine and permanent damage to the kidneys. Computed Tomography (CT) imaging is usually the method of choice among the existing diagnostic techniques because it is highly sensitive and can clearly display the arrangement of the stones. Nevertheless, the process of manual interpretation of medical images can be time-consuming and can be subject to variation, since the medical diagnosis can be determined by the experience of the radiologist. As computational methods developed, there has been an increase in interest in using Machine Learning (ML) and Deep Learning (DL) methods to automate the process of kidney stone detection. These techniques have demonstrated a possibility of enhancing uniformity and diagnostic capability. The review provides a summary of current approaches to kidney stones detection, including classic image processing algorithms, classical machine learning systems, and more modern deep learning models like Convolutional Neural Networks (CNNs) and object detection models like YOLO. Moreover, the paper addresses popular datasets, metrics of evaluation, and practical issues, such as limited data, the challenge of identifying stones of a small size, and the problem of model generalization in various clinical scenarios. Other recent directions, including hybrid modeling and explainable AI are also discussed, which could enhance the transparency and clinical adoption of automated systems. On the whole, this survey will equip a systematic knowledge of the existing trends and also pinpoint areas that need to be researched more in the context of automated kidney stone detection.
Vangapandu Venkata Kalyani, V. M. Moorthy· 2026 7th International Confe...· 0 citations
Nephrolithiasis (kidney stone disease) is a common urological disease that has a high clinical and economic impact. The early diagnosis is needed to avoid complications like obstruction of the ureter, infection, impaired kidney functioning. Traditional imaging modalities, such as ultrasonography, kidney-ureter-bladder radiography, and non-contrast computed tomography, are common but have a number of limitations, specifically their operator dependence, radiation, and low sensitivity to small or radiolucent stones. This review follows a PRISMA-based methodology to conduct a systematic review of studies published between 2015 and 2025 on the topic of computational intelligence methods such as artificial intelligence, machine learning, and deep learning to detect kidney stones based on medical images. Major scientific databases were considered in studies according to imaging modality, preprocessing method, model architecture and performance measures. Deep learning models, especially, Convolutional Neural Networks and U-Net-based frameworks, are highly effective in detection and segmentation tasks and have been reported to have accuracy of 86 to 99.9 percent, Dice coefficients over 0.85 and AUC of up to 0.99 in controlled data. Hybridization to combine ML classifiers, including Support Vector Machines, further improves the performance of classification. Yet, these outcomes are commonly limited through small datasets, class imbalance, external validation, and overfitting, which have an impact on real-life generalization. The use of computational intelligence has greatly improved the detection of kidney stones by enhancing automation, precision, and reproducibility. However, there are still major issues, such as the standardization of the dataset, interpretability of the models, and limitations to the clinical implementation. Explainable AI, federated learning, and 3D volumetric analysis should be prioritized in future research to create diagnostic systems.
Karthick P., Chiranji Lal Chowdhary· International Journal of Ima...· 0 citations
Kidney stone disease affects approximately 10% of the global population, underscoring the clinical importance of accurate CT-based stone classification and precise localization. This paper presents a leakage-controlled deep learning framework that delivers both pixel-level localization and transparent image-level prediction through two complementary model pipelines that are trained and evaluated independently. In the first pipeline, a U-Net is trained for grayscale-to-color reconstruction and stone localization; a binary stone mask is then obtained via post-processing to visually delineate suspected regions. In the second pipeline, an EfficientNet-B0 model is trained for binary classification (Normal vs. Stone) using patientdisjoint train/validation/test splits to mitigate data leakage and better reflect realistic deployment, and is benchmarked against standard Convolutional Neural Network (CNN) baselines. We further incorporate explainable AI to validate model reasoning: Gradient-weighted Class Activation Mapping (Grad-CAM) highlights discriminative regions, Local Interpretable Modelagnostic Explanations (LIME) provides superpixel-based local evidence, and SHapley Additive exPlanations (SHAP) assigns pixel-wise contribution scores. Across experiments, grayscale representations substantially outperformed color inputs, and augmentation improved generalization. Using two state-of-theart datasets, the proposed classifier achieved 99.76% and 99.65% test accuracy, respectively. For segmentation/localization, the U-Net component achieved Dice scores of 0.91-0.93 on the first dataset and 0.99 on the second dataset. Qualitative explainability results consistently focused on stone-relevant kidney regions, supporting interpretability and clinical plausibility of predictions. These findings indicate that combining robust training practices with multi-method explainability can yield highly accurate and clinically interpretable kidney stone analysis from CT images.
Abbas Rammal, Mageda A. A. Sharafeddin, Ranim Zreik et al.· 2026 6th International Confe...· 0 citations
This study presents an AI-driven computed tomography (CT) diagnostic system for the automated detection and classification of renal abnormalities in the Bangladeshi population. Renal abnormalities such as kidney cysts, stones, and tumors require timely and accurate diagnosis to reduce complications and improve treatment outcomes. In Bangladesh, the increasing burden of kidney-related diseases has created a strong need for efficient and intelligent diagnostic support systems. To assess the effectiveness of deep learning in multiclass renal abnormality diagnosis, three advanced convolutional neural network architectures were implemented and compared: Xception, VGG16, and ResNet152V2. The experimental results demonstrated outstanding classification performance across all models, with Xception achieving the highest accuracy of 99.84%, followed by ResNet152V2 at 99.76%, and VGG16 at 99.40%. Among the evaluated approaches, Xception showed the best overall performance, indicating its strong capability for reliable renal abnormality classification from CT images. The proposed system has significant potential to assist radiologists and healthcare professionals by providing fast, accurate, and automated diagnostic support. This work highlights the promise of artificial intelligence in medical imaging and contributes to the advancement of intelligent diagnostic solutions for kidney disease detection in Bangladesh.
Mithila Yeasmin Mitu· American Journal of Smart Te...· 0 citations
Background/Objectives: In resource-limited settings, where access to advanced imaging modalities such as magnetic resonance imaging (MRI) and histopathological confirmation may be limited, differentiating between hepatic cysts and metastatic lesions based solely on computed tomography (CT) remains challenging. This limitation may affect diagnostic confidence and increase the risk of misclassification, potentially impacting clinical decision making and patient management. In this study, we aimed to explore a more direct and automated approach for classifying hepatic lesions from CT images. Methods: We developed a deep learning-based framework combining transfer learning, decision fusion, and a stacking strategy by integrating five CNN architectures. The study included 100 patients, equally divided between metastatic liver tumors and pathological hepatic cysts. The dataset was built from both public data (LiTS) and internal clinical cases, and then split into training and testing sets. Results: The proposed stacking model provided the most consistent results, reaching an accuracy of 0.98, with high precision and sensitivity. The improvements in individual models, although moderate, were observed across all evaluation metrics. Conclusions: Overall, this approach offers a practical and reliable way to classify hepatic lesions with minimal manual intervention. It may help improve consistency in diagnosis and could be integrated into clinical workflows to support decision making in low-resource areas.
BACKGROUND AND OBJECTIVE
Contrast-enhanced ultrasound (CEUS) is widely used for evaluating thyroid nodule malignancy, but conventional time-intensity curve (TIC) analysis is labor-intensive and operator-dependent. This study proposes LSTAC, an automated framework for nodule segmentation and TIC analysis in CEUS videos.
METHODS
LSTAC integrates an improved YOLOv5-based segmentation network with a peak intensity frame (PIF) extraction algorithm to enable automatic nodule localization, TIC generation, and PIF identification. The framework was trained using CEUS data from 623 patients collected across three hospitals and evaluated on both internal and external validation cohorts.
RESULTS
LSTAC achieved 3-10× higher efficiency than VueBox in PIF extraction while maintaining strong temporal accuracy (0.94, 0.77, 0.79) and structural similarity (SSIM: 0.80, 0.60, 0.67). In malignancy prediction based on PIF features, LSTAC outperformed VueBox in two of three validation sets, with AUCs of 0.8279 vs. 0.8226 and 0.8000 vs. 0.7000.
CONCLUSION
LSTAC provides an efficient and accurate solution for automated CEUS analysis, reducing manual workload and improving consistency in thyroid nodule assessment, with potential for clinical application.
Aoxiang Yang, Liuyue Li, Ruifan He et al.· Artificial Intelligence in M...· 0 citations