Real-time Needle Tip Localization in 2D Ultrasound Images using Kalman Filter
| Autor: | M. Felix Orlando, Nisha Agarwal, Arjun Gupta, Ankur Kumar Yadav |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0301 basic medicine
Modality (human–computer interaction) medicine.diagnostic_test business.industry Computer science 030106 microbiology Ultrasound Image processing Kalman filter Tracking (particle physics) 03 medical and health sciences 030104 developmental biology Medical imaging Canny edge detector medicine Fluoroscopy Computer vision Artificial intelligence business |
| Zdroj: | AIM |
| DOI: | 10.1109/aim.2019.8868799 |
| Popis: | In percutaneous interventions, there are various medical imaging modalities such as MRI, Fluoroscopy, CT scan and Ultrasound. Among them, Ultrasound imaging modality is the safest and thus most popular to be useful for online analysis. However, US images are noisy and contains large number of artifacts. Thus it is a not an easy task to track the needle tip in a real time US imaging. In this paper, we have proposed a method to track the brachytherapy needle tip position in a real-time ultrasound medical image during percutaneous interventional procedure. In the present methodology, use of Image Processing technique along with the Kalman filter increases the stability and accuracy of the algorithm and realize an application in a dynamic situation. The Canny edge detection is used to estimate the position of the tip. Then the Kalman filter uses the results given by the above mentioned method as the measurements to increase the certainty. The main objective of the study is to identify needle losses during tracking and recover the tip position in an automated needle insertion. The results obtained show that in both stationary and dynamic situations, the proposed method gives a more stable and accurate result that can be accepted for clinical applications. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|