Detection of Brachial Plexus on Ultrasound Image

Type: Machine Learning Algorithm
Technologies: NumPy, Pandas, Pyhton, TensorFlow, Skimage
Team1 developer and 1 QA specialist,

This project is in the field of medical surgery, specifically shoulder surgery. Surgery is a medical specialty that uses operative manual and instrumental techniques on a person to investigate or treat a pathological condition such as a disease or injury, to help improve bodily function or appearance or to repair unwanted ruptured areas. Post-surgery effects are associated with discomfort and complications, like pain and soreness.

Currently, the patient’s pain is frequently managed by means of drugs, which causes plenty of undesirable side effects. One of the methods to reduce the patient’s pain instead of using drugs is implanting catheters that block or mitigate the pain at the source. Pain management catheters reduce drug addiction and speed up the patient’s recovery. So the catheter should be implanted into the correct area of the body and affect the nerves directly.

Project Objectives

Detection of Brachial Plexus and Location

To allow users to detect the nerve structure called the brachial plexus on the ultrasound picture and the exact area in the patient’s body to implant the catheter.


The main challenge of the project was to train and set up the neural network to provide users with a high segmentation accuracy. This task involved a lot of time to spend on setting the neural network learning rate, its size and defining the optimal learning deviation (optimal loss function).


To solve this problem, we chose UNet architecture. It is a special neural network architecture developed for biomedical images segmentation. The hyper-parameters tuning has shown that the maximum accuracy in recognition is achieved through the combination of Dice + Binary Cross-entropy as a loss function.

Shows the highlighted brachial plexus and its location

highlighted brachial plexus and its location
highlighted brachial plexus and its location


As a result, a machine learning script connected with the neural network was developed. The ultrasound image of the shoulder joint comes as an input. As an output, there is an image with the highlighted brachial plexus. Since it is a script, it can be used in any type of app (web, desktop, mobile). The proper neural network training and setting has allowed getting ~70% accuracy in recognition.

Areas of our expertise

Deep Learning
Neural Networks
Image Processing

Got a project in mind?

Just give us a holler