Portable Ultrasound Research System for Use in Automated Bladder Monitoring with Machine-Learning-Based Segmentation

Abstract

We developed a new mobile ultrasound device for long-term and automated bladder monitoring without user interaction consisting of 32 transmit and receive electronics as well as a 32-element phased array 3 MHz transducer. The device architecture is based on data digitization and rapid transfer to a consumer electronics device (e.g., a tablet) for signal reconstruction (e.g., by means of plane wave compounding algorithms) and further image processing. All reconstruction algorithms are implemented in the GPU, allowing real-time reconstruction and imaging. The system and the beamforming algorithms were evaluated with respect to the imaging performance on standard sonographical phantoms (CIRS multipurpose ultrasound phantom) by analyzing the resolution, the SNR and the CNR. Furthermore, ML-based segmentation algorithms were developed and assessed with respect to their ability to reliably segment human bladders with different filling levels. A corresponding CNN was trained with 253 B-mode data sets and 20 B-mode images were evaluated. The quantitative and qualitative results of the bladder segmentation are presented and compared to the ground truth obtained by manual segmentation.

State of the Art of Non-Invasive Technologies for Bladder Monitoring: A Scoping Review

“…Over the last few decades, advancements in technology have led to the improvement and optimization of ultrasound devices for BUV monitoring [3,37,[45][46][47][48]. Van Leuteren et al [45] presented another wearable ultrasound device called SENS-U, which is positioned on the lower abdomen using a skin-friendly adhesive.…”

“…Table 1 presents a multidimensional comparison of the three commercially available portable ultrasound devices, UBM, SENS-U, and DFree, in terms of portability, their use for real-time analysis, participant groups used during testing, and their accuracy in detecting a full bladder. Fournelle et al [48] have introduced a novel low-cost, portable ultrasound device for research purposes, called MoUsE. It is used for long-term and automated BUV monitoring utilizing machine learning segmentation techniques.…”

Weakly supervised learning‐based 3D bladder reconstruction from 2D ultrasound images for bladder volume measurement

“…13,14 In recent years, deep learning has been a huge success in image segmentation tasks. [15][16][17][18] Some studies presented new methods for bladder ultrasound image segmentation based on U-Net convolutional neural network, [19][20][21][22] which showed higher segmentation accuracy. However, they just focused on the segmentation of the bladder in a single 2D ultrasound image and did not comprehensively utilize the information from multiple 2D ultrasound images of one patient.…”

Special Issue “Computer Aided Diagnosis Sensors”

“…Ahmad et al [ 16 ] provided proof-of-principle for an optical-based, quick, simple, and sensitive screening technology for the detection of SARS-CoV-2, utilizing antigen-antibody binding interactions. Fournelle et al [ 17 ] developed a new mobile ultrasound device for long-term and automated bladder monitoring without user interaction consisting of 32 transmit and receive electronic components as well as a 32-element, phased array, 3 MHz transducer. Khasawneh et al [ 18 ] customized and pre-trained deep learning models based on convolutional neural networks were used to detect pneumonia caused by COVID-19 respiratory complications.…”