Abstract:The challenge for ultrasonic (US) power transfer systems, in implanted/wearable medical devices, is to determine when misalignment occurs (e.g., due to body motion) and apply directional correction accordingly. In this study, a number of machine learning algorithms were evaluated to classify US transducer misalignment, based on data signal transmissions between the transmitter and receiver. Over seven hundred US signals were acquired across a range of transducer misalignments. Signal envelopes and spectrograms were used to train and evaluate machine learning (ML) algorithms, classifying misalignment extent. The algorithms included an autoencoder, convolutional neural network (CNN) and neural network (NN). The best performing algorithm, was deployed onto a TinyML device for evaluation. Such systems exploit low power microcontrollers developed specifically around edge device applications, where algorithms were configured to run on low power, restricted memory systems. TensorFlow Lite and Edge Impulse, were used to deploy trained models onto the edge device, to classify signals according to transducer misalignment extent. TinyML deployment, demonstrated near real-time ( 99%. This opens the possibility to apply such ML alignment algorithms to US arrays (capacitive micro-machined ultrasonic transducer (CMUT), piezoelectric micro-machined ultrasonic transducer (PMUT) devices) capable of beam-steering, significantly enhancing power delivery in implanted and body worn systems.

What problem does this paper attempt to address?

### Problems Addressed by the Paper The paper aims to address the alignment issue in ultrasonic (US) energy transfer systems for implantable or wearable medical devices. Specifically, it focuses on how to detect misalignment and perform corresponding directional correction when the ultrasonic transmitter and receiver are misaligned due to human movement. ### Summary of the Paper's Main Content - **Research Background**: Ultrasonic energy transfer technology can charge batteries or directly power implants from shallow to deep layers. Compared to traditional electromagnetic energy transfer, ultrasonic transmitters have advantages such as small size, low tissue attenuation, high safe operating power, and the ability to achieve energy focusing. However, human movement can lead to significant losses in ultrasonic energy transfer, with even the smallest misalignment (<4 mm) potentially reducing acoustic energy by up to 40%. - **Research Methods**: Researchers evaluated several machine learning algorithms to classify the misalignment of ultrasonic transmitters. By collecting over 700 ultrasonic signals under different degrees of misalignment and using signal envelopes and spectrograms to train and evaluate the machine learning algorithms. The algorithms used include autoencoders, convolutional neural networks (CNN), and neural networks (NN). These algorithms can be directly applied to ultrasonic signals without extracting specific parameters. - **Experimental Design**: Signal acquisition was performed using low-cost commercial ultrasonic transmission modules (JSNSR-04T), and signals were recorded by misaligning at different horizontal positions. Each position was assigned a label for training and testing the classification algorithms. - **Data Processing**: The raw signals were processed using the Hilbert transform to extract the positive signal envelope, which was used as input for autoencoders and neural networks. For CNN, spectrograms generated from the raw signals were used as input. - **Algorithm Evaluation**: The performance of the algorithms was evaluated using metrics such as accuracy, precision, and recall. Ultimately, the best-performing algorithm was deployed on a TinyML device for real-time signal classification, achieving >99% accuracy and completing classification in <350 milliseconds. ### Conclusion The study successfully addressed the misalignment detection issue of ultrasonic transmitters in implantable or wearable medical devices using machine learning algorithms. It demonstrated the feasibility of achieving high-accuracy classification on low-power, memory-constrained edge devices. This provides a new direction for the future application of ultrasonic arrays (such as CMUT, PMUT devices), especially in scenarios requiring beam-steering to enhance energy transfer.

Evaluation of a Machine Learning Algorithm to Classify Ultrasonic Transducer Misalignment and Deployment Using TinyML

Tiny Machine Learning for High Accuracy Product Quality Inspection

Machine learning for medical ultrasound: status, methods, and future opportunities

Machine Learning Approaches for Defect Detection in a Microwell-based Medical Device

Tiny Machine Learning: Progress and Futures

Moving Healthcare AI-Support Systems for Visually Detectable Diseases onto Constrained Devices

A Self-Attention-Assisted TinyML With Effective Representation for UWB NLOS Identification

Performance Evaluation of Embedded Image Classification Models Using Edge Impulse for Application on Medical Images

Comparison of Tiny Machine Learning Techniques for Embedded Acoustic Emission Analysis

Machine Learning in Robotic Ultrasound Imaging: Challenges and Perspectives

A Machine Learning-Oriented Survey on Tiny Machine Learning

Machine Learning in Nano-Scale Biomedical Engineering

An Ultra-low Power TinyML System for Real-time Visual Processing at Edge

Addressing Gap between Training Data and Deployed Environment by On-Device Learning

Machine Learning Algorithms for Processing and Classifying Unsegmented Phonocardiographic Signals: An Efficient Edge Computing Solution Suitable for Wearable Devices

Machine learning used for simulation of MitraClip intervention: A proof-of-concept study

ML-MCU: A Framework to Train ML Classifiers on MCU-based IoT Edge Devices

Edge deep learning for neural implants: a case study of seizure detection and prediction

Improving Ultrasonic Testing by Using Machine Learning Framework Based on Model Interpretation Strategy