Proximal detection of guide wire perforation using feature extraction from bispectral audio signal analysis combined with machine learning

• Published in: Computers in biology and medicine Vol. 107; pp. 10 - 17
• Main Authors: Mahmoodian, Naghmeh, Schaufler, Anna, Pashazadeh, Ali, Boese, Axel, Friebe, Michael, Illanes, Alfredo
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.04.2019; Elsevier Limited
• Subjects: Accuracy; Acoustic emission; Adapters; Algorithms; Artificial neural networks; Audio signals; Bispectral analysis; Catheterization; Data processing; Feature extraction; Feedback; Flow velocity; Friction; Guide wire perforation; Guide wires; Learning algorithms; Learning theory; Machine learning; Medical instruments; Medical personnel; Neural networks; Perforation; Programmable logic controllers; Sensors; Signal analysis; Signal classification; Signal processing; Support vector machines; Surgeons; Three dimensional printing; Ultrasonic imaging; Veins & arteries; Wire; Acoustic emission; Bispectral analysis; Guide wire perforation
• ISSN: 0010-4825; 1879-0534
• DOI: 10.1016/j.compbiomed.2019.02.001

Artery perforation during a vascular catheterization procedure is a potentially life threatening event. It is of particular importance for the surgeons to be aware of hidden or non-obvious events. To minimize the impact it is crucial for the surgeon to detect such a perforation very early. We propose a novel approach to identify perforations based on the acquisition and analysis of audio signals on the outside proximal end of a guide wire. The signals were acquired using a stethoscope equipped with a microphone and attached to the proximal end of the guide wire via a 3D printed adapter. Bispectral analysis was employed to extract acoustic signatures in the signal and several features were extracted from the bispectrum of the signal. Finally, three machine learning algorithms - K-nearest Neighbor, Support Vector Machine (SVM), and Artificial Neural Network (ANN)- were used to classify a signal as a perforation or as an artifact. The bispectrum-based features resulted in valuable features allowing a perforation to be clearly identifiable from other occurring events. A perforation leaves a clear audio signal trace in the time-frequency domain. The recordings were classified as perforation, friction or guide wire bump using SVM with 97% (polykernel) and 98.62% (RBF) accuracy, k-nearest Neighbor an accuracy of 98.28% and ANN with accuracy of 98.73% was obtained. The presented approach shows that interactions starting at the tip of a guide wire can be picked up at its proximal end providing a valuable additional information that could be used during a guide wire procedure. •Acoustic emission from the distal tip of a guidewire are measurable proximally.•Audio profiles allow clinical observation of dynamics during guidewire procedures.•Bispectral feature extraction plus ANN allow to classify guide wire events.•Proximal sensor placement provides significant advantages for device developments.•Classification results of the guidewire events was possible with an accuracy of 95%.