Kidney stone disease is a prevalent urological condition that requires timely and accurate diagnosis to prevent serious health complications. Ultrasound imaging is commonly used for kidney examination due to its safety, affordability, and non-invasive nature. However, the presence of speckle noise, low contrast, and structural ambiguities in ultrasound images makes manual interpretation challenging and subjective. To address these limitations, this paper presents an automated kidney stone detection system based on ultrasound image analysis.The proposed approach integrates image preprocessing, region-of-interest segmentation, hybrid texture feature extraction, and artificial neural network classification. Median filtering is employed to reduce speckle noise, followed by threshold-based segmentation to isolate potential stone regions. Discriminative texture features are extracted using a combination of Gray Level Co-occurrence Matrix, Local Binary Patterns, statistical descriptors, and wavelet coefficients. The extracted hybrid feature vector is then classified using a multi-layer perceptron artificial neural network to determine the presence of kidney stones.Experimental evaluation demonstrates that the proposed system achieves high accuracy, sensitivity, and specificity while maintaining low computational complexity.