1 Introduction
Nocturnal enuresis (NE), i.e. bedwetting, is the involuntary discharge of urine at night in a child in the absence of congenital or acquired defects of the central nervous system or urinary tract [
1]. NE is the most common childhood complaint [
2] affecting 15 to 20% of 5-year-old children, 5% of 10-year-old children, and 1 to 2% of people aged 15 years and older [
3,
4]. Not only can it effect a child’s life in every aspect negatively, but it is also very stressful for the parents or carers [
5,
6]. Additionally, the cost for the family is estimated at £3000 per year. A comprehensive cost analysis along with the negative effects of NE on normal daily routines and social activities can be found in our previous study [
1].
Catheterisation remains the gold standard for bladder volume assessment, but it is invasive, uncomfortable to the patient, and introduces the risk of infection and trauma [
7]. Therefore, new non-invasive approaches are needed to monitor the bladder and to treat NE. Successful treatment of bedwetting tends to show increased self-esteem for children [
8]. A moisture detection alarm as a first-line treatment is recommended for up to 3 months, depending on the dry nights, by both The National Institute for Health and Care Excellence (NICE) [
9] and European Society for Paediatric Urology (ESPU) [
10]. A number of non-invasive products that can be categorised as (i) bedwetting alarms, (ii) pad and bell alarms, and (iii) bladder scanners currently available in the market to support the monitoring, diagnosis, and treatment of bed-wetting are yet to be found satisfactory in alleviating the predicaments of children with NE [
1]. Similarly, medications (e.g. Tricyclics, depression) are associated with potential side effects [
11] in addition to their poor performance [
12]. This suggests a need to explore alternative modalities. The initial request for innovation in this area, a more optimal device originated from clinicians, parents, and children who were unsatisfied with the performance of the traditional urinary post-void moisture alarms and medications [
1] where alarms are considered to be more effective than medications [
13].
1 In this regard, there have been several attempts to find a pre-void solution for NE sufferers [
14,
15]. An analysis of these attempts was presented in our previous study [
1]. The conclusion drawn is that there is no product in the market that can predict a pre-void occurrence so far.
There is an urgent need to develop an effective and comfortable device that (1) harbours artificial intelligence (AI) techniques to learn and evaluate the dynamic characteristics of the bladder and its surrounding tissues intelligently, (2) has customisable abilities for the children with various types of body morphologies, (3) determines imminent voiding need, and finally (4) provides pre-void alerts accordingly, in order to allow a patient to void in a dignified manner. We analysed the feasibility of such a system in our previous study [
1]. That study was developed to explore whether existing technologies could be synchronised, enhanced, and modulated to form an intelligent alarm system that could provide a pre-void warning system for sufferers. More explicitly, in that study, the viability of building, refining, and evaluating a new, safe, comfortable, and non-invasive wearable autonomous intelligent electronic device to monitor the bladder using a single element low-powered low-frequency ultrasound (US) with the help of artificial intelligence (AI) and machine learning (ML) techniques was carried out. The results indicated that customised imminent voiding need, based on the expansion of the bladder, could be determined by applying a single element transducer onto a bladder in an intermittent manner and the acquired results could be improved further with a comfortable non-invasive device by adding several other features. The approaches and techniques determined to cure the NE were patented nationally and internationally [
4,
16‐
20].
The aim in Advanced Mechatronics Systems (AMSs) development is to produce high-quality intelligent autonomous products and maintain a competitive edge through better product performance by forging effective sensing, self-learning, self-optimisation, self-configuration, self-diagnosis, and precise autonomous decision making and actuation. This is performed with no or less human intervention using effective location-independent monitoring, control, and management applications with products [
21]. With the advanced wireless communication techniques and improved battery technologies, AMSs are capable of becoming independent and working with other massive AMSs to construct robust, customisable, energy-efficient, autonomous, intelligent, and immersive platforms [
21]. Miniaturisation of components and consequently devices using MEMS (Micro-Electro-Mechanical Systems) technology is imperative for ergonomic and functional use. This technology is effectively used in almost every field such as automotive, electronics, medicine, communications, and defense, (e.g. airbag, intelligent tyres, disk drive heads). In this regard, the main objective of this study is to develop a robust, cheaper, more reliable, more flexible, customisable, and effective dry alarm to treat NE and manage bedwetting, which would be more acceptable than any currently available moisture alarms until the child has learned to control the bladder. More explicitly, this study has been carried out to explore whether existing technologies could be synchronised, enhanced, miniaturised, and modulated to form an AMS that could provide a pre-void warning, minimising bedwetting, reaching stable dryness through learning bladder control, and enhancing quality of life for children and families. To clarify the novelty of this paper, the contributions are outlined as follows.
1.To our knowledge, this study is the most comprehensive study in the literature to find an effective solution— i.e. a pre-void alarm system to treat NE involving a cross-disciplinary team, Patient and Public Involvement and Engagement (PPI) in cooperation with various prominent organisations, expert companies, and the AMS technologies.
2.This is the first attempt that explicitly studies a pre-void AMS alarm system to treat NE by forging the features of the edge and cloud platforms, communication technologies, and AI (particularly, reinforcement learning (RL)) and ML to enable the implementation of an automated, multi-functional, autonomous customisable system through self-learning by using transdisciplinary knowledge to address the challenges involved.
3.Bespoke US MEMS sensors specific to this application area have been built and incorporated into the system, and their viability and usability is tested both on phantoms and volunteers, in order to build a more comfortable wearable device.
4.Characteristics of the bladder with respect to liquid consumed and time are analysed in order to create a bespoke application specific to this organ, which can also direct other studies related to the bladder.
5.Bladder volume calculation with respect to the height and width of the bladder using 2D images acquired from a conventional US device has been analysed, which can help other studies related to the volume calculation of the bladder.
6.Morphologies of children are analysed in order to design the comfortable wearable undergarment and the other components, which can direct other studies related to wearable devices designed for children.
The paper is organised as follows. Section
1 outlines the background of the study including a comprehensive state-of-the-art literature on the evolution of NE, and previous and current attempts in order to cure NE. The approaches and techniques performed and the components used in this study are explored in Section
2. The results are presented in Section
3 along with the discussions. Finally, outlining the limitations in Section
4, Section
5 draws conclusions and Section
6 provides directions for potential future works.
3 Results and discussion
Development of a safe, comfortable, and non-invasive pre-void wearable alarm and associated technology using advanced mechatronics to treat bedwetting is the main objective of this study. In this manner, this study was carried out to explore whether existing technologies could be synchronised, enhanced, and modulated to form an intelligent alarm system with AMS components that could provide a pre-void warning, minimising bedwetting, reaching stable dryness through learning bladder control and enhancing quality of life for children who wet the bed. The results suggest that the samples acquired from the a single element US device and conventional US device correspond to the same status of the bladder. The maximum ROI for the pulse generator and the sensors should be 15 cm for 7–9-year-old children to ensure that all related area is covered. During the tests, the children consumed liquid during the data acquisition phases in order to both observe the characteristics of the expansion with respect to the consumed liquid and reduce the total test time to be able to observe scenario with the fastest filling of the bladder. The posterior wall can be detected between the second trial and third trial, in around 30 min where the expansion of the bladder starts above the 50-ml urine volume and the expansion increases almost in a linear form. The bladder fills in 100 min in four of the children when they were drinking as much as possible. The voiding need starts in approximately 65 min (i.e. 2/3 full bladder status) for 7–9-year-old children and 80 min for 14-year-old children that can change slightly from person to person regarding the bladder size and the specific characteristics of individuals. Please note that the children coded as MP3 and MP4 ended the trials earlier since their full bladder measurements are significantly lower than the normal ranges as explained in Section
2.3. Bladder volumes vary between individuals and have large standard deviations when working on the dynamic environment of the bladder as presented in Table
2. Therefore, the techniques employed in this study are designed to customise the MyPAD device for the specific characteristics of individuals through continuous learning supported by RL techniques. More explicitly, the techniques are developed to analyse and learn the patterns of all acquired echoed pulses within a dedicated ROI (e.g. 15 cm for 7–9-year-old and 20 cm for 14-year-old children) with respect to the specific features of the individuals rather than calculating the urine volume, in particular, patterns of echoed pulses that correspond to the voiding need. These patterns were found successful to discern the bladder volume status of empty, 1/2 (half), 3/4 (three quarters), and full [
1]. Further tests on more children using recently developed MEMS sensors printed on a flexible thin film (Figs.
18 and
19) are required to determine if the patterns indicate a specific volume of urine, which is planned and explained in Section
6. We are keen to perform this process in comparison with the voided urine of individuals since the conventional US modality is an approximation of the real volume,
4 and the gold standard catheterisation is invasive, uncomfortable, and introduces the risk of infection and trauma as explained in Section
1.
Se observed using the conventional sensors and gel as mentioned in our previous paper [
1] was 0.89 resulting in 11 false alarm out of 100 and causing sleep interruption with no reason. We are targeting to achieve a
Se value of over 0.95 with the new design explored throughout this paper, which is intended to be improved with autonomous feedback and child input as the device is used.
Regarding the workshop mentioned in Section
2.2 , (1) The families are desperate about the current moisture alarms and medicine they are using, (2) all these 4 children are suffering during day-time as well and they need to visit the toilet regularly to avoid any involuntary daytime voiding, (3) they would prefer using a pre-void alarm system, (4) The prominent features of such a system emphasised by the families and children are comfortable design and easy-to-use abilities regarding the hardware, SW, and undergarment, and most importantly (5) The children do not want to be different from their peers in appearance while using the device.
The solution for treating the bedwetting in this study is to design and develop an intelligent autonomous AMS to trigger a pre-void warning that can be customised to the user’s specific physical characteristics by combining several measurement attributes of the bladder when it is full, expanded, or empty. In this manner, we aim to deliver a compact AMS device with easy-to-use interfaces in a compact package that can be used by children without needing any engineer or parent. This device will be unique in that it recognises the warning signs of a pending emptying of the bladder via tracking expansion of the bladder volume over time, and will wake the patient up in time to prevent it. This process is customised or tuned to an individual patient’s bladder volume trigger point. This more accurate advanced warning system will help the children to alter their behaviour over time, reducing the frequency of NE [
30] through learning bladder control over time. The main advantage of our techniques using single element MEMS sensors is its simplicity, safe, low cost, and most importantly comfortable use. In the long run, larger collection of data samples from different sexes, age groups, and morphology types using the cloud platform as disclosed in Section
6 can be processed to improve the performance of the MyPAD device and train a stand-alone system that can be employed for larger range of NE patients with a very short customisation period.
Beyond this study, there are numerous other areas of application i.e. elder care (geriatric) settings, stroke patients [
31], diagnosis of urinary retention, and veterinary science in which My-PAD can be of potential benefit.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Disclaimer
The funding agreement ensured the authors’ independence in designing the study, interpreting the data, writing, and publishing the report. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health.