Fast-Unet++

This paper introduces Fast-Unet++, a convolutional neural network designed for precise kidney segmentation in sagittal and axial ultrasound images to estimate key biomarkers like kidney volume and dimensions. The model was trained and tested on both public datasets and patient scans, achieving high segmentation performance with Dice scores up to 0.97 and 0.95. Evaluation showed strong accuracy across multiple metrics, demonstrating the method’s effectiveness in supporting kidney disorder diagnosis and measurement.