A deep learning-based study on automated CT diagnosis of kidney stones, hydronephrosis and pyonephrosis.

AI / informaticsBMC medical imaging· Jul 2026

A deep learning-based study on automated CT diagnosis of kidney stones, hydronephrosis and pyonephrosis.

Hanlin Gao, Hongyao Chen, Yihe Wang … Chenjing Guo

Read paper

Stones & EndourologyAI / informaticsQ1IF 4.4

Deep learning model identifies kidney stones, hydronephrosis, and pyonephrosis on CT with 94.40% accuracy

A deep learning-based study on automated CT diagnosis of kidney stones, hydronephrosis and pyonephrosis.

Hanlin Gao … Chenjing Guo

BMC medical imaging · Jul 2026

Background

Automated identification of urinary system diseases on non-contrast computed tomography (non-contrast computed tomography) can facilitate early diagnosis and inform treatment decisions.

Methods

Subsequently, we evaluated multiple image enhancement strategies, including histogram equalization, CLAHE, Laplacian sharpening, and brightness enhancement.

94.40%

Results

94.40%

correctly classified about 19 of every 20 CT scans

More results

Through adaptive weighted fusion, the network enhances discrimination and diagnostic performance when complex lesions coexist.

Ablation studies further confirm the contributions of PPM, ASPP, and Tri-Evidence Attention.

“

Conclusion

Overall, the proposed unified framework enables reliable identification of kidney stones, hydronephrosis, and pyonephrosis, supporting clinical decision-making.

A deep learning-based study on automated CT diagnosis of kidney stones, hydronephrosis and pyonephrosis.

A deep learning-based study on automated CT diagnosis of kidney stones, hydronephrosis and pyonephrosis.

Related papers