A System for the Quantification of Fluid Intake - A Study on the Design of an Innovative Method to Improve Oral Hydration in Older Adults.
Lead Research Organisation:
King's College London
Department Name: Informatics
Abstract
Older adults are often poorly hydrated and therefore at higher risk of experiencing dehydration due to a diminished thirst sensation and a decrease in body water content, which naturally occurs with ageing [1][2]. In settings such as hospitals and care homes it has been observed that in the UK, 20% of the patients are dehydrated [3], due to the unreliability of the current standardised methods to assess hydration. Even if dehydration does not necessarily lead to death, it can lead to serious medical consequences such as delirium, infection of urinary tracts and cardiovascular symptoms [4]. Hence the concrete need to develop a tool which can improve monitoring and assessment of hydration in older adults.
The scope of this project is to develop a software that will be able to accurately quantify the daily fluid intake of older adults. To accomplish this, the idea is to use a combination of sensors such as EMGs, microphones and accelerometers to detect signals originated from swallowing activities. The main objectives are:
I. To determine the optimal anatomical position for the sensors in order to maximise the discriminative features of the recorded signals.
II. To develop methods for accurate identification of bolus type and quantification of the volume of the fluid being swallowed.
III. To test the developed methods in an asynchronous mode.
IV. To assess user acceptance of a device which can keep track of fluid intake.
The scope of this project is to develop a software that will be able to accurately quantify the daily fluid intake of older adults. To accomplish this, the idea is to use a combination of sensors such as EMGs, microphones and accelerometers to detect signals originated from swallowing activities. The main objectives are:
I. To determine the optimal anatomical position for the sensors in order to maximise the discriminative features of the recorded signals.
II. To develop methods for accurate identification of bolus type and quantification of the volume of the fluid being swallowed.
III. To test the developed methods in an asynchronous mode.
IV. To assess user acceptance of a device which can keep track of fluid intake.
Organisations
People |
ORCID iD |
| Ernest Kamavuako (Primary Supervisor) | |
| Carlotta Malvuccio (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/R513064/1 | 01/10/2018 | 30/09/2023 | |||
| 2338665 | Studentship | EP/R513064/1 | 07/01/2019 | 06/10/2022 | Carlotta Malvuccio |
| Description | The main aim of this project is to develop a device that can accurately monitor fluid intake in older adults living in healthcare settings. This device will mainly rely on on-body sensors, which are non-invasive and easy to use. Our first step was to assess user acceptance of our proposed device by potential end users, therefore, we organised a workshop where we submitted a set of questions to the participants, all aged 65 and over. All participants deemed the proposed device highly beneficial and considered it a step forward towards the quality of care for older adults. The full opinions and reflections gathered during the workshop are all documented in a paper for which we are currently seeking publication. Due to the positive response of potential end users, we proceeded with the selection of potential sensors able to harvest swallowing-related signals that can monitor fluid intake, with microphones and surface electromyography (sEMG) appearing as the best candidates. While there is evidence in the literature that microphones are able to detect fluid intake based on swallowing sounds, the data harvested is not enough to draw solid conclusions. Regarding sEMG, most studies only focused in harvesting data using these sensors to detect swallowing alterations in patients with dysphagia, however some studies suggested that with increasing volumes of swallowed liquid, certain features of sEMG present a variation. Therefore, we conducted an investigation on the ability of sEMG to classify between saliva and liquid swallows and estimating volume intake using four features, resulting in an accuracy of classification between saliva and liquid swallows of approximately 87%, and a mean Root Mean Square Error (RMSE) of approximately 2 ml for single sips. These results demonstrated the potential of sEMG to monitor fluid intake. The resulting paper was submitted and accepted at the IECBES IEEE-EMBS 2021 virtual conference on Biomedical Engineering and Sciences. Currently in press, and due to be published on IEEE Xplore, the paper received the Best Paper Award. After these results, we decided to conduct a more extensive investigation by using 46 features to verify if the results achieved on the aforementioned study can be improved and to select the features that will be used in the forthcoming experiments on sensor fusion, if the current pandemic situation allows. Our preliminary data for this study showed that sEMG features can detect swallowing events associated with fluid intake (versus combined saliva and silent period) with high sensitivity (96%), high specification (94%) and high precision (93%) and that the estimated volume per sip has an average mean squared error of 14% of the actual sip volume. Further results will be documented in the paper resulting from this study and for which we will seek publication once completed. |
| Exploitation Route | Delivering to the market a device that can accurately track fluid intake will immensely benefit hospitals and care homes by easing the workload of the healthcare personnel, as currently nurses and care home staff have to use manual recording methods to track fluid intake in older patients, who are more susceptible to dehydration and the risks it poses to their health. Especially in situations such as the current health emergency caused by covid-19, this tool would prove to be extremely beneficial and reduce the risk of infection between patients and the healthcare personnel. Furthermore, a possible and future expansion of this project could include groups which can also benefit from this device, such as the army and athletes. |
| Sectors | Aerospace, Defence and Marine,Healthcare,Leisure Activities, including Sports, Recreation and Tourism |
| Description | Workshop on a Fluid Intake Device with the University of Third Age (Sevenoaks) |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Public/other audiences |
| Results and Impact | A total of sixteen participants attended the workshop, of these, twelve were volunteers and all students at the University of Third Age, all aged 65 or over. The remaining four participants acted as moderators during the course of the workshop. Other than delving into user acceptance, the aims of this workshop were also to ascertain what the literature reports about the issue of dehydration in older adults and to gather unbiased perspectives and opinions from potential users. A set of questions was prepared to submit to the participants and let them express their experiences and opinions. The idea of developping a fluid intake device for older adults was widely supported and deemed as a potential significant improvement to quality of care, moreover, ideas on additional sensors and possible expansions to other fields were expressed by the participants. Furthermore, participants were eager to be involved in further studies and activities conducted by us. |
| Year(s) Of Engagement Activity | 2019 |