Automatic nerve segmentation of ultrasound images

Ultrasound guided regional anesthesia (UGRA) is the Regional Anesthesia (RA) technique of injecting the anesthetic in required amounts depending on the region to be covered using ultrasound images of patients. It is one of the fields which is steadfastly growing in medical imaging, providing advances in ultrasound imaging technology. However, nerve identification continues to remain as one of the most challenging tasks that anesthetists can encounter in the UGRA technique of injection. One of the main reasons of this include the low quality of the ultrasound images which are affected by the introduction of unwanted speckle noise. Since, Regional anesthesia (RA) is one of most frequently undertaken tasks in hospitals throughout the world to reduce or nullify the effect of the persistent pain in patients, any discrepancy in providing anesthesia can lead to severe damage to the respective region of the body or side effects to the rest of the body and also to the life of the patient. This enhances the need for correctly identifying the right nerve region to provide anesthesia. In this paper, we survey the conventional methods [1] used in biomedical image processing. For each stage of the processing, we have performed comparison with the different methodologies and we propose a combination of the methods which gives the best performance.