Bladder volume estimation from electrical impedance tomography

Application of Machine Learning Algorithms to the Discretization Problem in Wearable Electrical Tomography Imaging for Bladder Tracking

The article presents the implementation of artificial intelligence algorithms for the problem of discretization in Electrical Impedance Tomography (EIT) adapted for urinary tract monitoring. The primary objective of discretization is to create a finite element mesh (FEM) classifier that will separate the inclusion elements from the background. In general, the classifier is designed to detect the area of elements belonging to an inclusion revealing the shape of that object. We show the adaptation of supervised learning methods such as logistic regression, decision trees, linear and quadratic discriminant analysis to the problem of tracking the urinary bladder using EIT. Our study focuses on developing and comparing various algorithms for discretization, which perfectly supplement methods for an inverse problem. The innovation of the presented solutions lies in the originally adapted algorithms for EIT allowing for the tracking of the bladder. We claim that a robust measurement solution with sensors and statistical methods can track the placement and shape change of the bladder, leading to effective information about the studied object. This article also shows the developed device, its functions and working principle. The development of such a device and accompanying information technology came about in response to particularly strong market demand for modern technical solutions for urinary tract rehabilitation.

Urinary Bladder Volume Monitoring Using Magnetic Induction Tomography: A Rotational Simulation Model for Anatomical Slices within the Pelvic Region

Urinary bladder volume monitoring can benefit from contactless measurements, as alternative to the traditional medical methods of transurethral catheterization or ultrasound examination. The emerging modality of Magnetic Induction Tomography (MIT) offers the possibility for estimation of the intravesical volume in the physiological and pathological states using conductivity map reconstructions of the tissues present in the pelvic region. Within MIT, eddy currents originating from the conductive urine can produce their own magnetic field in response to an external magnetic source that is susceptible of being detected outside the body by means of a static ring of sensing coils. However, the ill-conditioned and ill-posed nature of the MIT Inverse Problem make the numerical implementation and conductivity estimation highly laborious. In this paper, we present a rotational frame model based on the MIT principles with application in urodynamic studies, which allows to extend the number of contactless measurements without increasing the overall dimension of the simulation domain, at the expense of solving multiple MIT Forward Problems. On the inversion process, the single-step Gauss-Newton method with Laplacian regularizer is recruited to estimate the bladder volume non-invasively and remotely (estimation error of 19%), paving the way for this technique to surpass the current limitations found in intravesical volume monitoring in quasi-real time.

A comprehensive survey on non-invasive wearable bladder volume monitoring systems

Measuring the volume of urine in the bladder is a significant issue in patients who suffer from the lack of bladder fullness sensation or have problems with timeliness getting to the restroom, such as spinal cord injury patients and some of the elderlies. Real-time monitoring of the bladder, therefore, can be highly helpful for urinary incontinence. Bladder volume monitoring technologies can be divided into two distinct categories of invasive and non-invasive. In invasive techniques, a catheter is directly inserted into the urethra to measure the amount of urine accurately. However, it is painful, limits the user's ordinary movements, and may hurt the urinary tract. Current non-invasive techniques measure the volume of the bladder from the skin using different stationary or portable apparatus at health centers. Both techniques have difficulties and are not cost-effective to use for a long period. Recently, both invasive and non-invasive methods have been attempted to be produced in the form of wearable devices utilizing different sensing and communication technologies. Wearable bladder monitoring devices can be easily used by patients with no or few clinical steps, making them much more affordable than non-wearable devices. While wearable devices seem to be a highly convenient and effective solution, they suffer from few drawbacks, such as relatively low precision. Hence, a great number of studies have been conducted to address these issues. In this article, we review and discuss non-invasive and minimally invasive methods for monitoring the bladder volume. We focus on the most practical and state-of-the-art methods employed in wearable devices, classify them by engineering and medical characteristics, and investigate their specifications, architectures, and measurement algorithms. This study aims to introduce the latest advances in this field to practitioners while comparing the advantages and disadvantages of existing approaches. Our study concludes with open problems and future trends in the area of bladder monitoring and measurement systems. Graphical abstract Wearable bladder monitoring system.

Advances in Ambulatory Urodynamics

PURPOSE OF REVIEW

This manuscript reviews recent technological advances in ambulatory urodynamics.

RECENT FINDINGS

Ambulatory urodynamics is currently recommended by the International Continence Society as a second-line diagnostic tool in patients with nondiagnostic traditional urodynamics. Novel techniques involving telemetric monitoring are in development, which utilize catheter-free wireless systems to address several recognized shortcomings of inoffice urodynamic studies. Current research in catheter-free bladder pressure measurements involves either an intravesical, intradetrusor, or transdetrusor approach. Real-time bladder volume estimation may be performed using ultrasonography, near-infrared spectroscopy, or bladder volume conductance measurement. Ambulatory urodynamics can measure bladder function in the "real world" setting, capturing physiological bladder filling and emptying and allowing patients to reproduce the activities that may trigger their symptoms. Telemetric devices being developed represent further advances in this field and focus upon improving diagnostic capabilities, evaluating patient response to treatment, and facilitating closed-loop bladder control with neuroprosthetic integration.

Bladder Volume Monitoring Using Electrical Impedance Tomography With Simultaneous Multi-Tone Tissue Stimulation and DFT-Based Impedance Calculation Inside an FPGA

In this article, a novel method for measuring the volume of the urinary bladder non-invasively is presented that relies on the principles dictated by Electrical Impedance Tomography (EIT). The electronic prototype responsible for injecting innocuous electrical currents to the lower abdominal region and measuring the developed voltage levels is fully described, as well as the computational models for resolution of the so-called Forward and Inverse Problems in Imaging. The simultaneous multi-tone injection of current provided by a high performance Field Programmable Gate Array (FPGA), combined with impedance estimation by the Discrete Fourier Transform (DFT) constitutes a novelty in Urodynamics with potential to monitor continuously the intravesical volume of patients in a much faster and comfortable way than traditional transurethral catheterization methods. The resolution of the Inverse Problem is performed by the Gauss-Newton method with Laplacian regularization, allowing to obtain a sectional representation of the volume of urine encompassed by the bladder and surrounding body tissues. Experimentation has been carried out with synthetic phantoms and human subjects with results showing a good correlation between the levels of abdominal admittivity acquired by the EIT system and the volume of ingested water.

Recent advances in urodynamics in women

Urodynamics is the study of the storage and evacuation of urine from the urinary tract. The aim is to reproduce the patient’s symptoms and provide a pathophysiological explanation for them by identifying all factors that contribute to the lower urinary tract dysfunction, including those that are asymptomatic. Urodynamics consists of various tests, each of which is designed to assess a different aspect of lower urinary tract function. There is a lack of evidence regarding when urodynamics should be used in the non-neurogenic bladder. Some small randomised controlled trials suggest that urodynamics does not alter the outcome of surgery for stress urinary incontinence when compared with office evaluation alone. However, this is widely felt to be inaccurate and many health-care professionals still advocate the use of urodynamics prior to any invasive treatment, especially surgery on the lower urinary tract. There have been few technological advances in urodynamics in recent years. Air-charged rather than fluid-filled catheters were thought to help reduce artefact, but the evidence is unclear, and there is doubt over their accuracy. Ambulatory urodynamics is carried out over a longer period of time, enabling physiological bladder filling, but it remains invasive and artificial. To attempt to replicate symptoms more accurately, there have been efforts to develop wireless devices to measure detrusor pressure directly. These may be promising but are far from suitable in humans at present. Urodynamics continues to provide useful information for assessing lower urinary tract function, but further large studies are required to assess its value and develop innovations to improve the accuracy of the tests and acceptability to patients.

Boxer Underwear Incorporating Textile Moisture Sensor to Prevent Nocturnal Enuresis

Around 15% of children still wet their bed after five years old. Although bedwetting alarms have proven to be effective to achieve nighttime dryness, they are cumbersome so children could be reluctant to use them. Therefore, the moisture sensor and wire were made unobtrusive by seamlessly integrated them into fully textile underwear by using conductive yarns. Consequently, the alarm acceptability should be enhanced by improving children’s comfort. Three conductive textile metallic yarns, made of silver or stainless steel, were considered to fabricate the urine leakage sensor. Silver-plated-nylon yarn, which showed the highest electrical conductivity, outperformed the stainless-steel yarns regarding its ability to detect urine leakage as well as its detection speed. Furthermore, it was proven to withstand multiple urine soakings and the following machine-washings, even at high temperature (60 °C). However, the electrical current, necessary to detect the leakage, tends to corrode the silver. Therefore, the detection circuit was adapted. Eventually, the designed leakage sensor was seamlessly integrated into a child’s trunk underwear, into which a miniaturized alarm can be plugged. The resulting textile underwear aims at replacing the rigid alarm system currently available, hence improving the quality of life of enuretic children and help them achieving nighttime dryness.

Robustness of focused and global impedance estimates of bladder volumes against uncertainty of urine conductivity

Bioimpedance measurements are currently used to monitor various biological processes and are potentially useful for studies of urodynamics. Global impedance (GI) and focused impedance measurements (FIM) can be used to monitor bladder volumes, but these are subject to varying conductivity of urine. To address this, we emulated a human bladder using an agar phantom filled with saline solutions of varying conductivities and estimated volumes using a modified FIM-based approach. Using this novel strategy, electrical potentials did not change significantly with constant liquid volumes, even when the conductivity of the saline solutions was varied between 1.027 to 1.877 and 2.610 S/m. Conversely, GI and classic FIM measurements of constant liquid volumes varied with conductivity. These observations suggest that the proposed FIM approach is suitable for bladder volume estimation due to its robustness against uncertainties of conductivity. The bioimpedance hardware used in our experiments comprised 8 electrodes and a a small and low cost impedance measurement system based on an AFE4300 direct impedance measurement device.

Ambulatory urodynamic monitoring: state of the art and future directions

Urodynamic studies are a key component of the clinical evaluation of lower urinary tract dysfunction and include filling cystometry, pressure-flow studies, uroflowmetry, urethral function tests and electromyography. However, pitfalls of traditional urodynamics include physical and emotional discomfort, artificial test conditions with catheters and rapid retrograde filling of the bladder, which result in variable diagnostic accuracy. Ambulatory urodynamic monitoring (AUM) uses physiological anterograde filling and, therefore, offers a longer and more physiologically relevant evaluation. However, AUM methods rely on traditional catheters and pressure transducers and do not measure volume continuously, which is required to provide context for pressure changes. Novel telemetric AUM (TAUM) methods that use wireless, catheter-free, battery-powered devices to monitor bladder pressure and volume while patients carry out their daily activities are currently being investigated. TAUM devices under current development are innovating in the areas of remote monitoring, rechargeable energy sources, device deployment and retrieval and materials engineering to provide increased diagnostic accuracy and improved comfort for patients with incontinence or voiding dysfunction. These devices hold promise for improving the diagnosis and management of patients with lower urinary tract disorders.

Analysis of measurement electrode location in bladder urine monitoring using electrical impedance

Purpose

The aim of this study was to document more appropriate electrode location of a four-electrode-based electrical impedance technology in the monitoring of bladder filling, and to characterize the relationship between bladder filling duration and the measured electrical impedances.

Methods

A simulation study, based on a 2-dimension computational model, was conducted to determine the preferable locations of excitation and measurement electrodes in a conventional four-electrode setup. A human observation study was subsequently performed on eight healthy volunteers during natural bladder urine accumulation to validate the result of the simulation study. The correlation between the bladder filling time and the measured electrical impedance values was evaluated.

Results

The preferable location of measurement electrodes was successively validated by the model simulation study and human observation study. Result obtained via the selected electrodes location revealed a significant negative correlation (R = 0.916 ± 0.059, P < 0.001) between the measured electrical impedance and the urine accumulation time, which was consistent with the result of simulation study.

Conclusions

The findings in this study not only documented the desirable electrodes location to monitor the process of bladder urine accumulation using four-electrode measurement, but also validated the feasibility of utilizing electrical impedance technique to monitor and estimate the bladder urine volume for those with urological disorders.

EIT Imaging of Upper Airway to Estimate Its Size and Shape Changes During Obstructive Sleep Apnea

Noninvasive continuous imaging of the upper airway during natural sleep was conducted for patients with obstructive sleep apnea (OSA) using the electrical impedance tomography (EIT) technique. A safe amount of alternating current (AC) was injected into the lower head through multiple surface electrodes. Since the air is an electrical insulator, upper airway narrowing during OSA altered internal current pathways and changed the induced voltage distribution. Since the measured voltage data from the surface of the lower head were influenced not only by upper airway narrowing but respiratory motions, head motions, and blood flows, we developed a pre-processing algorithm to extract the voltage component originated from upper airway closing and opening. Using an EIT image reconstruction algorithm, time-series of EIT images of the upper airway were produced with a temporal resolution of 50 frames per second. Applying a postprocessing algorithm to the reconstructed EIT images, we could extract quantitative information about changes in the size and shape during upper airway closing and opening. Results of the clinical studies with seven normal subjects and ten OSA patients show the feasibility of the new method for OSA phenotyping and treatment planning.

Image-based classification of bladder state using electrical impedance tomography

OBJECTIVE

In this study, we examine the potential of using machine learning classification to determine the bladder state ('not full', 'full') with electrical impedance tomography (EIT) images of the pelvic region. Accurate classification of these states would enable urinary incontinence (UI) monitoring to alert the patient, before involuntary voiding occurs, in a low-cost and discrete manner.

APPROACH

Using both numerical and experimental data, we form datasets that contain diverse observations with varying clinical parameters such as bladder volume, urine conductivity, and the reference used for time-difference imaging. We then classify the bladder state using both pixel-wise and feature extraction-based classification techniques. We employ principal component analysis, wavelets, and image segmentation to help create features.

MAIN RESULTS

The performance was compared across several classifier algorithms. The minimum accuracy was 77.50%. The highest accuracy observed was 100%, and was found by combining principal component analysis and the Gaussian radial based function kernel support vector machine. This combination also offered the best trade-off between classification performance and the costs of training time and memory space. The biggest challenge in bladder state classification is classifying volumes near the separation volume of not full and full, in which choosing the most suitable classifier combination can minimize this error.

SIGNIFICANCE

We performed the first machine learning classification of bladder EIT images, achieving high classification accuracies with both numerical and experimental data. This work highlights the potential of using image-based machine learning with an EIT device to support bladder monitoring for those suffering from UI.

Brain haemorrhage detection using a SVM classifier with electrical impedance tomography measurement frames

Brain haemorrhages often require urgent treatment with a consequent need for quick and accurate diagnosis. Therefore, in this study, we investigate Support Vector Machine (SVM) classifiers for detecting brain haemorrhages using Electrical Impedance Tomography (EIT) measurement frames. A 2-layer model of the head, along with a series of haemorrhages, is designed as both numerical models and physical phantoms. EIT measurement frames, taken from an electrode array placed on the head surface, are used to train and test linear SVM classifiers. Various scenarios are implemented on both platforms to examine the impact of variables such as noise level, lesion location, lesion size, variation in electrode positioning, and variation in anatomy, on the classifier performance. The classifier performed well in numerical models (sensitivity and specificity of 90%+) with signal-to-noise ratios of 60 dB+, was independent of lesion location, and could detect lesions reliably down to the tested minimum volume of 5 ml. Slight variations in electrode layout did not affect performance. Performance was affected by variations in anatomy however, emphasising the need for large training sets covering different anatomies. The phantom models proved more challenging, with maximal sensitivity and specificity of 75% when used with the linear SVM. Finally, the performance of two more complex classifiers is briefly examined and compared to the linear SVM classifier. These results demonstrate that a radial basis function (RBF) SVM classifier and a neural network classifier can improve detection accuracy. Classifiers applied to EIT measurement frames is a novel approach for lesion detection and may offer an effective diagnostic tool clinically. A challenge is to translate the strong results from numerical models into real world phantoms and ultimately human patients, as well as the selection and development of optimal classifiers for this application.

Continuous bladder volume monitoring system for wearable applications

In this research, we propose a bladder volume monitoring system that can be effectively applied for various voiding dysfunctions. Whereas conventional systems lack consecutive measurements, the proposed system can continuously monitor a user's status even during unconscious sleep. For the convenience, we design a simple and comfortable waist-belt-type device by using the body impedance analysis (BIA) technique. To support various measurement scenarios, we develop applications by connecting the device to a smartphone. To minimize motion noises, which are inevitable when monitoring over an extended period, we propose a motion artifact reduction algorithm that exploits multiple frequency sources. The experimental results show a strong relationship between the impedance variation and the bladder volume; this confirms the feasibility of our system.

Supervised Learning Classifiers for Electrical Impedance-based Bladder State Detection

Urinary Incontinence affects over 200 million people worldwide, severely impacting the quality of life of individuals. Bladder state detection technology has the potential to improve the lives of people with urinary incontinence by alerting the user before voiding occurs. To this end, the objective of this study is to investigate the feasibility of using supervised machine learning classifiers to determine the bladder state of ‘full’ or ‘not full’ from electrical impedance measurements. Electrical impedance data was obtained from computational models and a realistic experimental pelvic phantom. Multiple datasets with increasing complexity were formed for varying noise levels in simulation. 10-Fold testing was performed on each dataset to classify ‘full’ and ‘not full’ bladder states, including phantom measurement data. Support vector machines and k-Nearest-Neighbours classifiers were compared in terms of accuracy, sensitivity, and specificity. The minimum and maximum accuracies across all datasets were 73.16% and 100%, respectively. Factors that contributed most to misclassification were the noise level and bladder volumes near the threshold of ‘full’ or ‘not full’. This paper represents the first study to use machine learning for bladder state detection with electrical impedance measurements. The results show promise for impedance-based bladder state detection to support those living with urinary incontinence.

A realistic pelvic phantom for electrical impedance measurement

OBJECTIVE

To design and fabricate an anatomically and conductively accurate phantom for electrical impedance studies of non-invasive bladder volume monitoring.

APPROACH

A modular pelvic phantom was designed and fabricated, consisting of a mechanically and conductively stable boundary wall, a background medium, and bladder phantoms. The wall and bladders are made of conductive polyurethane. The background material is an ultrasound gel-based mixture, with conductivity matched to a weighted average of the pelvic cavity organs, bone, muscle and fat. The phantom boundary is developed using a computer tomography model of a male human pelvis. The bladder phantoms were designed to correlate with human bladder dimensions. Electrical impedance measurements of the phantom were recorded, and images produced using six different bladder phantoms and a realistic finite element model.

MAIN RESULTS

Five different bladder volumes were successfully imaged using an empty bladder as a reference. The average conductivity index from the reconstructed images showed a strong positive correlation with the bladder phantom volumes.

SIGNIFICANCE

A conductively and anatomically accurate pelvic phantom was developed for non-invasive bladder volume monitoring using electrical impedance measurements. Several bladders were designed to correlate with actual human bladder volumes, allowing for accurate volume estimation. The conductivity of the phantom is accurate over 50-250 kHz. This phantom can allow changeable electrode location, contact and size; multi-layer electrodes configurations; increased complexity by addition of other organ or bone phantoms; and electrode movement and deformation. Overall, the pelvic phantom enables greater scope for experimentation and system refinement as a precursor to in-man clinical studies.

A prospective study of examining physiological signals for estimating occurrence of nocturnal enuresis

The nocturnal enuresis is challenging due to the increased social activities of the children. This disorder significantly bothers both the children and their parents in psychological, behavioral, social, and financial manners. However, the primary treatments have limitations and further are not able to completely cure the disorder. In order to reduce pain and burdens of patients and their parents, it is important to accurately estimate when the enuretic incident occurs in advance. For the estimation, we have comprehensively investigated various studies of the nocturnal enuresis in the diverse fields. Through the investigations, we have summarized four hypotheses of the physiological signals related to the enuretic moment. In order to conquer the nocturnal enuresis, we design a preliminary framework sensing and investigating the physiological signals with the sensors. Our synthesized approach to understand and estimate the moments of the enuretic incidents can establish a foothold to complete the promising prediction system.

Design and evaluation of potentiometric principles for bladder volume monitoring: a preliminary study

Recent advances in microelectronics and wireless transmission technology have led to the development of various implantable sensors for real-time monitoring of bladder conditions. Although various sensing approaches for monitoring bladder conditions were reported, most such sensors have remained at the laboratory stage due to the existence of vital drawbacks. In the present study, we explored a new concept for monitoring the bladder capacity on the basis of potentiometric principles. A prototype of a potentiometer module was designed and fabricated and integrated with a commercial wireless transmission module and power unit. A series of in vitro pig bladder experiments was conducted to determine the best design parameters for implementing the prototype potentiometric device and to prove its feasibility. We successfully implemented the potentiometric module in a pig bladder model in vitro, and the error of the accuracy of bladder volume detection was <±3%. Although the proposed potentiometric device was built using a commercial wireless module, the design principles and animal experience gathered from this research can serve as a basis for developing new implantable bladder sensors in the future.