This research demonstrates significant correlations between pathomic features and disease progression in focal segmental glomerulosclerosis and minimal change disease.
Key Points
Using computational methods, significant correlations were found between visual scores of pathology and arterial characteristics in patients with focal segmental glomerulosclerosis and minimal change disease.
Pathomic features showed improved predictions of disease progression, enhancing clinical assessment metrics from 0.70 to 0.75 for arterioles and from 0.69 to 0.74 for arcuate arteries.
A total of 2,316 arterial structures, including arterioles and arteries, were analyzed using advanced deep learning models to accurately segment and quantify vascular features.
This innovative computational pipeline improves the reliability and standardization of arterial assessments, potentially transforming clinical practices in nephrology.