Kidney Stone Detection from Ultrasound Images of Kindney using Image Processing
Ultrasound imaging is one of the widely used imaging techniques used for diagnosis of kidney abnormalities especially renal calculi (kidney stones). During surgical processes it is vital to recognize the true and precise location of kidney stone. The detection of kidney stones using ultrasound imagi
2025-06-28 16:33:56 - Adil Khan
Kidney Stone Detection from Ultrasound Images of Kindney using Image Processing
Project Area of Specialization Artificial IntelligenceProject SummaryUltrasound imaging is one of the widely used imaging techniques used for diagnosis of kidney abnormalities especially renal calculi (kidney stones). During surgical processes it is vital to recognize the true and precise location of kidney stone. The detection of kidney stones using ultrasound imaging is a highly difficult task as they are of low contrast and contain speckle noise. This challenge is overcome by employing suitable image processing techniques. The ultrasound image is first pre-processed (restoration, smoothing and sharpening, and contrast enhancement), to get rid of speckle noise using the image restoration process. The restored image is then smoothened using Gabor filter and the subsequent image is enhanced by histogram equalization. The pre-processed image is achieved with level set segmentation to detect the stone region. Segmentation process is employed twice for getting better results; first to segment kidney portion and then to segment the stone portion, respectively. In this work, the level set segmentation uses two terms, namely, momentum and resilient propagation to detect the stone portion. Lastly, we perform refinement and crop the segmented kidney region from the original image.
Project ObjectivesProject objectives are included here as,
- To improve the performance of previously proposed kidney detection methods.
- To assist the doctors by improving and speeding-up the diagnosis process of kidney stone.
- To propose a cost-effective and non-invasive kidney stone detection system that requires ultrasound images instead of costly CT-Scan or X-ray images.
The project is purely based on deep learning and digital image processing. Tentatively, the following techniques can be used in the implementation
- Gathering the dataset.
- Speckle noise removal.
- Image enhancement.
- Training the classifier.
- Working on improvement of classifier.
- Segmentation of candidate regions of kidney stone.
Here are some benefits of our project.
- Assist the doctors by speeding up the detection process.
- Provide ease to patients to help them examine their own reports.
- Cost-effective solution to kidney stone detection.
Following are the technical deliverables of this project
- Classifier that classify whether the ultrasound image contain kidney stone or not.
- An algorithm that further process the ultrasound image that has been detected to have kidney stone to detect the precise location of the stone.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 78500 | |||
| GTX 1060 | Equipment | 1 | 30000 | 30000 |
| Power supply 850W | Equipment | 1 | 14000 | 14000 |
| Raspberry Pi 4 4GB | Equipment | 1 | 20000 | 20000 |
| Noir Raspberry Pi Camera | Equipment | 1 | 5500 | 5500 |
| Project Thesis Publication | Miscellaneous | 1 | 6000 | 6000 |
| Printing and Binding | Miscellaneous | 1 | 3000 | 3000 |