Browsing by Person "Vassallo, Michael"

Now showing 1 - 7 of 7

Advancing real-world applications: A scoping review on emerging wearable technologies for recognizing activities of daily living
(Elsevier, 2025-03-25) Ahmed, Mustafa Elhadi; Yu, Hongnian; Vassallo, Michael; Koufaki, Pelagia
Wearable technologies for Activities of Daily Living (ADL) recognition have emerged as a crucial area of research, driven by the global rise in aging populations and the increase in chronic diseases. These technologies offer significant benefits for healthcare by enabling continuous monitoring and early detection of health issues. However, the field of ADL recognition with wearables remains under-explored in key areas such as user variability and data acquisition methodologies. This review aims to provide a comprehensive overview of recent advancements in ADL recognition using wearable devices, with a particular focus on commercially available devices. We systematically analyzed 157 studies from six databases following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, narrowing our focus to 77 articles that utilized proprietary datasets. These studies revealed three main categories of wearables: prototype devices (40 %), commercial research-grade devices (32 %), and consumer-grade devices (28 %) adapted for ADL recognition. Additionally, various detection algorithms were identified, with 31 % of studies utilizing basic machine learning techniques, 40 % employing advanced deep learning methods, and the remainder exploring ensemble learning and transfer learning approaches. Our findings underscore the growing adoption of accessible, commercial devices for both research and clinical applications. Furthermore, we identified two key areas for future research: the development of user-centered data preparation techniques to account for variability in ADL performance, and the enhancement of wearable technologies to better align with the practical needs of healthcare systems. These advancements are expected to enhance the usability and efficiency of wearables in improving patient care and healthcare management.
Can a systems approach reduce adverse outcomes in patients with dementia in acute settings? (innovative practice)
(SAGE, 2017-11-03) Duah-Owusu White, Mary; Vassallo, Michael; Kelly, Fiona; Nyman, Samuel
People with dementia experience adverse outcomes such as pressure sores during their stay in acute hospitals. The application of a systems approach in an acute setting places an emphasis on the patient's journey in addition to the organisational factors that are present within a hospital context. This article draws upon principles obtained from a theoretical model, which was extracted from the work of Edwards (1972), Hawkins (1987) and Zecevic et al. (2007), in order to illustrate how the application of a novel systems approach (human interaction, environment, equipment and policy) could be used in acute hospital settings to reduce adverse health outcomes by using an imaginary patient with dementia.
Enhanced LDA with SVM and Ensemble Learning for Advanced ADL Classification Using Smartwatch Data
(IEEE, 2024-08-28) Ahmed, Mustafa Elhadi; Yu, Hongnian; Vassallo, Michael; Koufaki, Pelagia
The demographic shift towards an older population introduces significant challenges for healthcare, necessitating advancements in predictive, dynamic care models. Central to this transformation are Activities of Daily Living (ADLs) and Instrumental Activities of Daily Living (IADLs), which are crucial for evaluating an individual's functional status. This study leverages Machine Learning (ML) with smartwatch technology for continuous, non-intrusive monitoring of ADLs /IADLs. By integrating Support Vector Machine (SVM) and Ensemble Learning (EL) with enhanced Linear Discriminant Analysis (LDA), we aim to accurately classify a broad spectrum of daily activities. Data from 30 participants, captured through a smartwatch, underwent analysis to assess these methodologies' efficacy in real-life applications. Results indicate that LDA enhanced classification accuracy, with EL models demonstrating superior performance in recognizing complex activities—achieving over 90% accuracy. The study also examines the hardware performance of these models, noting variations in RAM usage, inference times, CPU usage, and network sizes. EL models, especially when coupled with LDA, offer an optimal balance between computational demands and classification precision. This investigation underlines the potential of integrating sophisticated data processing techniques with wearable technologies for health monitoring, presenting a viable approach to overcoming real-life deployment challenges and advancing personalized healthcare for the aging demographic.
Integrating Sensors Data and Machine Learning for Enhanced ADL and Fall Detection Systems
(IEEE, 2025-09-06) Pathan, Nazia; Yu, Hongnian; Vassallo, Michael; Koufaki, Pelagia
Developing a fall-proof environment presents various challenges. Statistics show that falls are the leading cause of injury-related death among people aged more than 80. Fall detection and prediction systems can increase stability and safety to reduce the risk. One of the best ways to develop fall detection/ prediction systems (FDPS) is with multi-sensor (accelerometer, gyroscope, magnetometer) data by using machine learning algorithms. However, the challenges for getting this system are a selection of optimal sensors, a fusion of sensors and machine learning (ML) algorithms for the Fall Detection and Prediction System. This study analyses the impact of sensor integration, ML algorithms performance on integrated sensors, and individual data for the FDPS. To achieve the objective of this research, five publicly available multi-sensor datasets are used; a proposed feature selection algorithm is applied for sensor data selection and ML algorithms such as Decision Tree, Naive Bayes, K-Nearest Neighbour, Support Vector Machine, Adaboost, Gradient Boosting, and Random Forest are applied for fall prediction. The ML algorithm’s performance is significantly enhanced by using integrated sensors, achieving an accuracy of 100% on dataset d3. Among the algorithms tested, the Random Forest algorithm emerged as the best-performing model across all five datasets, d1 to d5, with an accuracy of 85%, 98%, 100%, 97%, and 41%, respectively. Additionally, evaluating the performance of individual sensors for predictions, the accelerometer was found to be the most effective sensor.
Two factors that can increase the length of hospital stay of patients with dementia
(Elsevier, 2022-12-02) Duah-Owusu White, Mary; Vassallo, Michael; Kelly, Fiona; Nyman, Samuel
Objectives Patients with dementia are at greater risk of a long hospital stay and this is associated with adverse outcomes. The aim of this service evaluation was to identify variables most predictive of increased length of hospital stay amongst patients with dementia. Methods/Design We conducted a retrospective analysis on a cross-sectional hospital dataset for the period January–December 2016. Excluding length of stay less than 24h and readmissions, the sample comprised of 1133 patients who had a dementia diagnosis on record. Results The highest incidence rate ratio for length of stay in the dementia sample was: (a) discharge to a care home (IRR: 2.443, 95% CI 1.778–3.357), (b) falls without harm (IRR: 2.486, 95% CI 2.029–3.045). Conclusions Based on this dataset, we conclude that improvements made to falls prevention strategies in hospitals and discharge planning procedures can help to reduce the length of stay for patients with dementia.
Understanding the hospital discharge planning process for medical patients with dementia
(Informa UK Limited, 2023-10-21) Duah-Owusu White, Mary; Kelly, Fiona; Vassallo, Michael; Nyman, Samuel R.
Background: Poor hospital discharge processes can result in the readmission of patients and potentially increase the stress levels of carers. Therefore, this study sought to understand the factors related to the discharge planning process for patients with dementia. Methods: The researchers interviewed 32 carers of patients with dementia and 20 hospital staff who worked on medical wards in a United Kingdom (UK) hospital. The semi-structured interviews were analysed thematically using a systems theory (patient–carer–staff relationships, hospital equipment and policies). Results: The findings indicated that the following factors could either have a positive or negative impact on discharge planning: patient (e.g. cognitive capacity), carer (e.g. preconceived ideas about care homes), staff (e.g. communication skills), policy (e.g. procedures such as discharge meetings), equipment (e.g. type of service provider delivering the equipment) and the wider social context (e.g. availability of specialist dementia beds in care homes). Conclusion: It is important for hospital staff to adopt a systems perspective and to integrate the different elements of the hospital system when planning for patients’ discharge.
Using a systems perspective to understand hospital falls among patients with dementia
(Elsevier, 2022-09-26) Duah-Owusu White, Mary; Kelly, Fiona; Vassallo, Michael; Nyman, Samuel R.
Background Falls are a frequent event among older adults with dementia during their hospital stay. This qualitative study explores the factors contributing to falls in this population using a systems perspective. Methods Semi-structured interviews were conducted with 32 carers of patients with dementia and 20 hospital staff who worked on medical wards. Interview transcripts were analysed thematically using a systems framework. Results The themes generated from this falls research were factors related to the: patient (e.g. their physical health), carer (e.g. their ability to re-call a patient's past medical history), staff (e.g. teamwork), hospital policies (e.g. transfer of patients between wards), the hospital environment (e.g. lack of observation side rooms for infectious patients who are at risk of falls on some wards) and the use of hospital equipment (e.g. walking aid). Conclusion We recommend that future hospital falls intervention programmes need to be supported by a credible systems approach aiming to improve patient outcomes in relation to falls prevention.