Enhanced Feature Extraction for Detection and Classification of Kidney Abnormalities.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Nephrology

Background:

• Kidney abnormalities (cysts, stones, tumors) present significant health risks and can lead to chronic kidney disease if not diagnosed promptly.
• Early and accurate diagnosis is crucial for effective patient management and improved clinical outcomes.

Purpose of the Study:

• To propose a deep learning-based diagnostic framework for the automatic detection and classification of multiple kidney conditions using CT scan images.
• To introduce the Kidney Transformer Network (KTNET) with an enhanced feature extraction strategy for improved diagnostic accuracy.

Main Methods:

• Development of the Kidney Transformer Network (KTNET), a novel deep learning model utilizing transformer-based architecture.
• Application of KTNET for feature extraction and classification of kidney abnormalities (Normal, Cyst, Tumor, Stone) from CT scan images.

Main Results:

• The proposed KTNET model achieved outstanding diagnostic performance: 99.7% accuracy, 99.4% precision, 99.3% recall, and 99.6% F1-score.
• KTNET significantly outperformed traditional image processing methods and existing deep learning models in classifying kidney conditions.
• The model demonstrated high adaptability and efficiency across diverse CT scan datasets.

Conclusions:

• The KTNET framework offers an intelligent, reliable, and accurate solution for early detection and classification of kidney abnormalities.
• This research advances medical imaging analysis, with strong potential for practical integration into clinical workflows for enhanced patient diagnosis and decision-making.