Ravikumar Ch.; Satyanarayana Nimmala; P. Vasanthsena; Nenavath Chander; R. Sahith

Attention-Based Spatial-Temporal Learning for Enhanced Kidney Abnormality Diagnosis from Ultrasound Images

PDF (649KB), PP.474-485

Views: 0 Downloads: 0

Author(s)

Ravikumar Ch. ^1,* Satyanarayana Nimmala ² P. Vasanthsena ³ Nenavath Chander ⁴ R. Sahith ⁵

1. Department of CSE, Sreenidhi University, Hyderabad, Telangana, 500031, India

2. Department of CSE (Data Science), CVR College of Engineering, Hyderabad, India

3. G. Narayanamma Institute of Technology and Science (for women), Hyderabad, India

4. Department of CSE (cyber security), Malla Reddy University, Hyderabad, Telangana, India

5. Department of CSE, CVR College of Engineering, Hyderabad, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijem.2026.05.25

Received: 3 Jul. 2026 / Revised: 30 Jul. 2026 / Accepted: 25 Aug. 2026 / Published: 8 Oct. 2026

Index Terms

Kidney abnormalities, Ultrasound imaging, Attention-Driven CNN-LSTM, Feature fusion, Explainable AI

Abstract

Kidney abnormalities, when detected late, can develop into chronic conditions such as kidney tumors and cardiorenal syndrome. Recently, there has been a move toward automation, especially as it pertains to developing a framework for advanced preprocessing, feature fusion, and hybrid spatial-temporal learning for the detection of the abnormalities in question. For example, Non-Local Means (NLM) filtering for noise reduction, image quality improvements through U-Net kidney segmentation and deep learning-based text annotation removal, as well as robust training through data augmentation and preprocessing ends up facilitating the process. The combination of classical texture and shape descriptors and deep embeddings from a Vision Transformer (ViT) through an attention mechanism allows the model to pinpoint where clinically relevant focus should be. A hybrid CNN-LSTM framework attends to spatial and temporal feature extraction, while attention modules receive the classification output to refine it. The incorporation of weighted binary cross-entropy for handling class imbalance, and of explainable AI, is demonstrated through Integrated Gradients. Experimental results for VATLA show it used advanced algorithms and automation principles in developing systems for all previous performance indicators, while providing evidence of scoring 0.94 in accuracy, 0.92 in precision, 0.95 in recall, and 0.96 in AUC, or area under the curve.

Cite This Paper

Ravikumar Ch., Satyanarayana Nimmala, P. Vasanthsena, Nenavath Chander, R. Sahith, "Attention-Based Spatial-Temporal Learning for Enhanced Kidney Abnormality Diagnosis from Ultrasound Images", International Journal of Engineering and Manufacturing (IJEM), Vol.16, No.5, pp. 474-485, 2026. DOI:10.5815/ijem.2026.05.25

Reference

[1]M. Di Serafino, F. Iacobellis, M. L. Schillirò, et al., "Common and uncommon errors in emergency ultrasound," Diagnostics, vol. 12, no. 3, Art. no. 631, 2022. doi: 10.3390/diagnostics12030631.
[2]D. A. Drew, D. E. Weiner, and M. J. Sarnak, "Cognitive impairment in CKD: Pathophysiology, management, and prevention," American Journal of Kidney Diseases, vol. 74, no. 6, pp. 782–790, 2019. doi: 10.1053/j.ajkd.2019.05.017.
[3]W. R. Zhang and C. R. Parikh, "Biomarkers of acute and chronic kidney disease," Annual Review of Physiology, vol. 81, pp. 309–333, 2019. doi: 10.1146/annurev-physiol-020518-114605.
[4]S. Cammarasana, P. Nicolardi, and G. Patanè, "Real-time denoising of ultrasound images based on deep learning," Medical & Biological Engineering & Computing, vol. 60, no. 8, pp. 2229–2244, 2022. doi: 10.1007/s11517-022-02573-5.
[5]H. Chen, M. Ma, G. Liu, et al., "Breast tumor classification in ultrasound images by fusion of deep convolutional neural network and shallow LBP feature," Journal of Digital Imaging, vol. 36, pp. 932–946, 2023. doi: 10.1007/s10278-022-00711-x.
[6]R. Arora and B. Raman, "BUS-Net: Breast tumor detection network for ultrasound images using bi-directional ConvLSTM and dense residual connections," Journal of Digital Imaging, vol. 36, no. 2, pp. 627–646, 2023. doi: 10.1007/s10278-022-00733-5.
[7]K. Velayuthapandian, M. Kishore, and S. Sajin, "End-to-end light-weighted deep-learning model for abnormality classification in kidney CT images," International Journal of Imaging Systems and Technology, vol. 34, 2024. doi: 10.1002/ima.23022.
[8]S. K. Pande and R. Agarwal, "Multi-class kidney abnormalities detecting novel system through computed tomography," IEEE Access, Early Access, 2024, doi: 10.1109/ACCESS.2024.3351181.
[9]E. Krispin, E. Dreyfuss, O. Fischer, A. Wiznitzer, E. Hadar, and R. Bardin, "Significant deviations in sonographic fetal weight estimation: Causes and implications," Archives of Gynecology and Obstetrics, vol. 302, pp. 1339–1344, 2020. doi: 10.1007/s00404-020-05732-x.
[10]N. Blau, E. Klang, N. Kiryati, M. Amitai, O. Portnoy, and A. Mayer, "Fully automatic detection of renal cysts in abdominal CT scans," International Journal of Computer Assisted Radiology and Surgery, vol. 13, no. 7, pp. 957–966, 2018. doi: 10.1007/s11548-018-1726-6.
[11]E. Kaczmarek, O. X. Miguel, A. C. Bowie, R. Ducharme, A. L. J. Dingwall-Harvey, S. Hawken, et al., "CAManim: Animating end-to-end network activation maps," PLoS ONE, vol. 19, no. 6, Art. no. e0296985, 2024, doi: 10.1371/journal.pone.0296985.
[12]C. P. Kovesdy, "Epidemiology of chronic kidney disease: An update 2022," Kidney International Reports, vol. 12, no. 1, pp. 7–11, Apr. 2022. doi: 10.1016/j.kisu.2021.11.003.
[13]I. Aksakallı, S. Kaçdıoğlu, and Y. S. Hanay, "Kidney X-ray images classification using machine learning and deep learning methods," Balkan Journal of Electrical and Computer Engineering, vol. 9, no. 2, pp. 144–151, Apr. 2021. doi: 10.17694/bajece.878116.
[14]N. Bhaskar, M. Suchetha, and N. Y. Philip, "Time Series Classification-Based Correlational Neural Network With Bidirectional LSTM for Automated Detection of Kidney Disease," IEEE Sensors Journal, vol. 21, no. 4, pp. 4811–4818, 2021. doi: 10.1109/JSEN.2020.3028738.
[15]F. Ma, T. Sun, L. Liu, and H. Jing, "Detection and diagnosis of chronic kidney disease using deep learning-based heterogeneous modified artificial neural network," Future Generation Computer Systems, vol. 111, pp. 17–26, 2020. doi: 10.1016/j.future.2020.04.036.
[16]M. Patrício, J. Pereira, J. Crisóstomo, et al., "Using resistin, glucose, age and BMI to predict the presence of breast cancer," BMC Cancer, vol. 18, Art. no. 29, 2018. doi: 10.1186/s12885-017-3877-1.
[17]T. D. K. S. C. Gunasekara, P. M. C. S. De Silva, E. M. D. V. Ekanayake, et al., "Urinary biomarkers indicate pediatric renal injury among rural farming communities in Sri Lanka," Scientific Reports, vol. 12, Art. no. 8478, 2022. doi: 10.1038/s41598-022-10874-w.
[18]N. Park, Y. Choi, H. Lee, et al., "Predicting acute kidney injury in cancer patients using heterogeneous and irregular data," PLOS ONE, vol. 13, no. 7, Art. no. e0199839, Jul. 2018. doi: 10.1371/journal.pone.0199839.
[19]J. Qin, L. Chen, Y. Liu, C. Liu, C. Feng, and B. Chen, "A machine learning methodology for diagnosing chronic kidney disease," IEEE Access, vol. 8, pp. 20991–21002, 2020. doi: 10.1109/ACCESS.2019.2963053.
[20]R. J. Kate, X. Perez, K. Mazumdar, M. J. Pasupathy, and A. Nilakantan, "Prediction and detection models for acute kidney injury in hospitalized older adults," BMC Medical Informatics and Decision Making, vol. 16, Art. no. 39, 2016. doi: 10.1186/s12911-016-0277-4.
[21]K.-H. Uhm, S.-W. Jung, M. H. Choi, et al., "Deep learning for end-to-end kidney cancer diagnosis on multi-phase abdominal computed tomography," NPJ Precision Oncology, vol. 5, no. 1, p. 54, 2021. doi: 10.1038/s41698-021-00195-y.
[22]S. Shi, "A novel hybrid deep learning architecture for predicting acute kidney injury using patient record data and ultrasound kidney images," Applied Artificial Intelligence, vol. 35, no. 15, pp. 1329–1345, 2021. doi: 10.1080/08839514.2021.1976908.
[23]V. Singh, V. K. Asari, and R. Rajasekaran, "A deep neural network for early detection and prediction of chronic kidney disease," Diagnostics, vol. 12, no. 1, Art. no. 116, 2022. doi: 10.3390/diagnostics12010116.
[24]P. Gogoi and J. A. Valan, "Chronic kidney disease prediction using machine learning techniques: A comparative study of feature selection methods with SMOTE and SHAP," Multiscale and Multidiscip. Model. Exp. and Des., vol. 8, Art. no. 216, 2025, doi: 10.1007/s41939-025-00806-2.
[25]Q. Zheng, S. L. Furth, G. E. Tasian, and Y. Fan, "Computer-aided diagnosis of congenital abnormalities of the kidney and urinary tract in children based on ultrasound imaging data by integrating texture image features and deep transfer learning image features," Journal of Pediatric Urology, vol. 15, no. 1, pp. 75.e1–75.e7, Feb. 2019, doi: 10.1016/j.jpurol.2018.10.020., vol. 15, no. 1, pp. 75.e1–75.e7, Feb. 2019. doi: 10.1016/j.jpurol.2018.10.020.
[26]K. Yildirim, P. G. Bozdag, M. Talo, O. Yildirim, M. Karabatak, and U. R. Acharya, "Deep learning model for automated kidney stone detection using coronal CT images," Computers in Biology and Medicine, vol. 135, Art. no. 104569, 2021. doi: 10.1016/j.compbiomed.2021.104569.
[27]S. Sudharson and P. Kokil, "An ensemble of deep neural networks for kidney ultrasound image classification," Computer Methods and Programs in Biomedicine, vol. 197, Art. no. 105709, 2020. doi: 10.1016/j.cmpb.2020.105709.
[28]Y. Bi, X. Shi, J. Ren, M. Yi, and X. Han, "Transarterial chemoembolization of unresectable renal cell carcinoma with doxorubicin-loaded CalliSpheres drug-eluting beads," Scientific Reports, vol. 12, Art. no. 7866, 2022. doi: 10.1038/s41598-022-12334-x.
[29]X. Wang, X. Ning, L. Xia, Q. Guo, and G. Cheng, "A new effective machine learning framework for sepsis diagnosis," IEEE Access, vol. 6, pp. 48300–48310, 2018. doi: 10.1109/ACCESS.2018.2867728.

International Journal of Engineering and Manufacturing (IJEM)