A CNN-based classifier for probe to skin contact in a robotic ultrasound system
MetadataShow full item record
Abstract: In an ultrasound examination the sonographer must be precise and apply the rightamount of pressure to get good quality images. An examination takes valuable timewhich could be reduced by introducing a robotic ultrasound system. In a roboticultrasound system the ultrasound transducer needs to maintain good contact with theskin of the subject. Automatic skin to probe contact detection may therefore be usefulin medical applications where it is desirable to use a robotic ultrasound device. In this thesis a convolutional neural network (CNN) algorithm was implementedand used to distinguish if the ultrasound probe had full contact, partial contact or nocontact with a person s skin. The main dataset consisted of ultrasound images of twobody parts from one subject. Different subsets of the complete dataset was also usedwhen training and testing the network. Four experiments with different subsets of thecomplete dataset were performed, with applied changes to the hyperparameters, inattempt to and the best overall CNN for the task. The CNN was developed and trained using MATLAB R2018a with the Neural NetworkToolbox. The results found indicated that a smaller CNN was sufficient for the task.The CNN which obtained the best results consisted of two convolutional layers andtwo fully connected layers. The experiment with this network was performed threetimes and reached an average accuracy of 100% and an average classification time of0.0430 seconds. The accuracy achieved indicated that similarities in the training set and test set could have affected the reliability of the results. It is therefore importantthat in future work, it is acquired images from more body parts and more subjects toget enough variation in the dataset. This is important to be able to further improvethe generalization of the CNN.