CARE: Context-awareness for elderly care

The purpose of this paper is to identify features of space planning guidelines for elderly care environments from a holistic health perspective. This was undertaken with the expected outcome of promoting a better understanding of elderly care environment as a health resource. The major method used was a content analysis technique that employed physical, psychological, and social health criteria. Data for analysis were collected through care facility administrators using a checklist of 114 items related to spatial planning, developed through analyzing design guidelines from Japan, the UK, Germany, and the USA. As a supplementary method, a literature review was conducted on the current status of Korean healthcare facilities and recent trends in design theories for elderly care facilities. To measure results, a matrix was developed in which each guideline item was matched to each of the three health dimensions, thereby delineating the relationship between them. The elderly care environment is a vital resource to provide and support holistic health, and it is. Thus, important to understand how these factors relate to and utilize one another in order to increase their congruence. The resulting matrix is expected to function as a foundation for providing better planning services, which could develop further into “elderly care facility design guidelines for holistic health”.

The global aged population is expected to exceed 2.1 billion, representing 21.65% of the total population by 2050.This demographic shift underscores an urgent need for efficient elderly care, particularly in home settings.AI advancements have made sensor technology, including wearable biosensors, environmental monitors, and biochemical sensors, essential for elderly care by enabling the collection of physiological and activity data.Current systems overwhelm caregivers with complex data analysis and personalized recommendations.Large language models (LLMs) address this by offering insights through natural language interfaces, using extensive medical data.While some studies have integrated sensor data with LLMs for health monitoring applications, a comprehensive framework for seamlessly combining diverse sensor data with LLMs in elderly care is still missing.In this study, we propose a novel methodological framework that addresses the challenges of integrating heterogeneous sensor data with LLMs to provide real-time healthcare insights for caregivers of the elderly using sensor technologies.Our framework employs few-shot learning on Generative Pre-trained Transformer (GPT-4) and GPT-3.5 to process structured sensor data from wearable and environmental devices.The LLMpowered application then generates insightful responses based on the user's input, providing actionable and personalized recommendations.The GPT-4 model outperformed GPT-3.5 in Structured Query Language (SQL) query generation for sensor data retrieval and processing, achieving a semantic similarity score of 0.95, precision of 88.5%, recall of 98.92%, and an F1score of 93.40%.In this study, we explore how integrating sensor data with LLMs enhances usability and reduces complexity in health monitoring systems.Our framework sets a new benchmark for advancing elderly care through innovative LLM-powered applications and sensor technology.Recent advances in healthcare sensor technology enable real-time diagnostics and various functions, including the continuous monitoring of physiological parameters.AI evolution has made sensor technologies crucial in elderly care, generating signal-based time series data.However, current health monitoring systems (31)(32)(33)(34)(35)(36)(37)(38)(39)(40)(41)(42)(43) often overwhelm caregivers for lack of personalized insights owing to the complexity of data captured by sensor technologies.LLM services can bridge this gap by providing personalized, actionable recommendations in natural language, making sensor data interpretation more accessible and effective for elderly care.Various studies (44)(45)(46)(47)(48)(49) have integrated LLMs with specific sensor data to provide health insights and personalized recommendations for patients and healthcare professionals.However, no methodological approach that seamlessly combines LLM capabilities with heterogeneous sensor data in elderly care has been proposed.Such integration would provide actionable insights and personalized recommendations for caregivers while expanding sensor technology applications by conveying health insights in natural language.To address this gap, in this study, we propose a comprehensive methodological framework for integrating diverse sensor data types into LLMs.This unified approach aims to enhance elderly care by leveraging sensor technologies to provide caregivers with actionable insights and personalized recommendations in real time.By demonstrating how sensor data such as physiological parameters (e.g., heart rate, blood oxygen saturation, steps, heart rate variability) can be effectively processed and interpreted using LLM, in this framework, we establish a scalable solution for personalized healthcare.Furthermore, we set the foundation for future studies on LLM integration with sensor technologies in diverse health monitoring applications for older adults, taking a significant step forward in the evolution of sensor technologies in healthcare.This study is carried out with the following specific objectives.a.To identify sensor types, their data, and processing techniques in elderly care and to define data sources and processing requirements for LLM integration b.To propose a framework for integrating diverse sensor data with LLMs for personalized elderly care and to evaluate LLMs' potentials in sensor data interpretation The rest of the paper is organized as follows.In Sect.2, we review existing work, and the materials and methods are described in Sect.3. In Sect.4, we present the results, whereas in Sect.5, we introduce the proposed framework.Experiment details are described in Sect.6, the discussion in Sect.7, and challenges and opportunities in Sect.8. Finally, in Sect.9, we conclude this study and outline future work. Related WorkSeveral LLMs such as Med-PaLM2, (19) HuatuoGPT, (50) DISC-MedLLM, (51) ChatDoctor, (52) and Baize-HealthCare (53) have been developed for medical question answering, using large medical datasets for diagnostic dialogues.With supervised fine-tuning, these models outperform Generative Pre-trained Transformer (GPT-4) and Llama2 on benchmark datasets such as USMLE questions, MedQuAD, and online medical consultation datasets. (54)ang et al. (55) suggested models using multiple LLMs, each specialized in a medical area, for automated diagnosis.Li et al. (56) enhanced LLM diagnostic abilities by incorporating clinical

The need for privacy in elderly care is crucial, especially where constant monitoring can intrude on personal dignity. This research introduces the development of a unique camera-based monitoring system designed to address the dual objectives of elderly care: privacy and safety. At its core, the system employs an AI-driven technique for real-time subject anonymization. Unlike traditional methods such as pixelization or blurring, our proposed approach effectively removes the subject under monitoring from the scene, replacing them with a two-dimensional avatar. This is achieved through the use of YOLOv8, which facilitates accurate real-time person detection and pose estimation. Furthermore, the proposed system incorporates a fall detection algorithm that utilizes a residual causal convolutional network together with motion features of persons to identify emergency situations and promptly notify caregivers in the event of a fall. The effectiveness of the system is evaluated to emphasize its advanced privacy protection technique and fall detection capabilities using several metrics. This evaluation demonstrates the system’s proficiency in real-world applications and its potential to enhance both safety and privacy in elderly care environments.

Evaluation of safety and efficacy of desmopressin/tolterodine combination therapy in women. This double-blind, randomized, proof-of-concept study enrolled 106 patients (≥18 years), with overactive bladder (OAB) and nocturia, with ≥2 nocturnal voids, receiving a 3-month once-daily combination (desmopressin 25 µg, orally-disintegrating tablets [ODT]/tolterodine 4 mg [Detrol® LA]; n = 49) or monotherapy (tolterodine 4 mg/placebo ODT; n = 57). Primary endpoint was change from baseline in mean number of nocturnal voids. Secondary endpoints were change from baseline in nocturnal voided volume, time to first nocturnal void, and quality-of-life. Post-hoc exploratory analysis were performed for patients with and without baseline nocturnal polyuria (NP, n = 47 each). Overall population showed a non-significant reduction in mean number of nocturnal voids with combination versus monotherapy (full analysis set: adjusted treatment contrast [TC], -0.34; P = 0.112). Change in mean nocturnal void volume (TC, -64.16 mL; P = 0.103), mean time to first nocturnal void (TC, 18.00 min; P = 0.385) and Nocturia Impact (NI) Diary© scores were comparable. In post-hoc analysis, NP patients showed a benefit with combination versus monotherapy for nocturnal void volume (P = 0.034) and time to first nocturnal void (P = 0.045), and a non-significant improvement in NI Diary© scores. Safety profile was comparable between treatments. A single transient event of asymptomatic clinically significant hyponatremia in combination group resolved subsequently. Low-dose desmopressin could be safely combined with tolterodine for treating nocturia in women with OAB, with a significant benefit in women with NP. Further, prospective validation studies of combination therapy are warranted in mixed NP/OAB population, based on this favorable proof-of-concept finding.

To report a study of the relationship between person-centred care and ability to perform activities of daily living, quality of life, levels of pain, depressive symptoms, and agitated behaviours among residents with dementia in residential care facilities. Standardized measurements of person-centred care have not previously been used to investigate the relationship between person-centred care and well-being for residents with dementia in residential aged care units. This study had a cross-sectional design. Staff and resident surveys were used in a sample of 1261 residents with dementia and 1169 staff from 151 residential care units throughout Sweden. Valid and reliable scales were used to measure person-centredness and ability to perform activities of daily living, quality of life, levels of pain, depressive symptoms, and agitated behaviours in residents. All data were collected in May 2010. Person-centred care was correlated with residents' ability to perform activities of daily living. Furthermore, residents in units with higher levels of person-centred care were rated as having higher quality of life and better ability to perform activities of daily living compared with residents in units with lower levels of person-centred care. There seems to be a relationship between person-centredness, residents' ability to perform activities of daily living, and residents' quality of life. Further studies are needed to explain the variation of person-centredness between units and the extent and ways this might impact on the quality of life and well-being of frail older residents with cognitive impairments in clinical practice.

The lack of healthcare staff and increasing proportions of elderly population is alarming. The traditional means to look after elderly has resulted in 255,000 reported falls (only within UK). This not only resulted in extensive aftercare needs and surgeries (summing up to £4.4 billion) but also in added suffering and increased mortality. In such circumstances, the technology can greatly assist by offering automated solutions for the problem at hand. The proposed work offers an Internet of things (IoT) based patient bed-exit monitoring system in clinical settings, capable of generating a timely response to alert the healthcare workers and elderly by analyzing the wireless data streams, acquired through wearable sensors. This work analyzes two different datasets obtained from divergent families of sensing technologies, i.e., smartphone-based accelerometer and radio frequency identification (RFID) based accelerometer. The findings of the proposed system show good efficacy in monitoring the bed-exit and discriminate other ambulating activities. Furthermore, the proposed work manages to keep the average end-to-end system delay (i.e., communications of sensed data to Data Sink (DS)/Control Center (CC) + machine-based feature extraction and class identification + feedback communications to a relevant healthcare worker/elderly) below of a second.

In Japan, humanoid robots has been introduced in the medical and elderly care environment. The application program of Care Prevention Gymnastics Exercises for Pepper (Pepper-CPGE) made by Xing Company, Japan is a body-brain gymnastics recreation program for 40 minutes tailored to the elderly's functional level. It consists of moving, watching/healing, and playing. "Move the body" exercise and other active range of motion activities are done according to the music. Pepper-CPGE was introduced as a clinical trial at the Mifune hospital, beginning in May 2018. At the units where clinical trials are done, 80% of the patients are with mental illness diagnosis with decreased physical functions often moving by wheelchair only. When Pepper-CPGE was introduced, the following changes were observed : (1) communication between patients and nurses during rehabilitation care using Pepper-CPGE was increased ; (2) patients wereinteractive, engaged, and actively participated in the Care Prevention Gymnastic Exercises using Pepper-CPGE ; (3) patients had fun and enjoyed talking to Pepper-CPGE. Interventions using Pepper-CPGE appear to be an effective rehabilitation strategy to increase engagement and participation of elderly patients who require long-term care and rehabilitation. J. Med. Invest. 66 : 50-53, February, 2019.

With an aging global population, the need for innovative solutions in elderly care is more critical than ever. Robotics offers significant potential to improve mobility, enhance independence, and reduce the physical burden on caregivers. This paper explores the design, development, challenges, and impact of robotic technologies for elderly care, with a specific focus on assistive devices aimed at improving patient mobility. Key examples of exoskeletons, robotic canes, and mobility aids are discussed, alongside their market adoption, evolution, and impact. Through a deeper understanding of the formulation, design, and challenges, this paper highlights the current state of robotic systems in elderly care and outlines a future trajectory for their widespread implementation. Keywords Elderly care, assistive robotics, patient mobility, exoskeletons, robotic cane, caregiving support, human-robot interaction, aging population, rehabilitation robotics, smart mobility, fall prevention, caregiver assistance, healthcare robotics, AI in robotics, rehabilitation exoskeletons, autonomous robots, robotics for aging, aging-in-place technology, wearable robotics, aging population, robotic exoskeletons.

The convergence of artificial intelligence and robotics presents unprecedented opportunities to address growing challenges in long-term care settings, particularly skilled nursing facilities. As global populations age and healthcare worker shortages intensify, AI-driven robotic systems offer promising solutions by augmenting human caregiving capabilities. This article examines the current applications, benefits, implementation challenges, and future directions of AI-powered robotic assistance in eldercare environments. Current applications include mobility assistance and physical support systems that reduce caregiver injuries while increasing resident autonomy, companion robots that address loneliness and social isolation through engaging interactions, and health monitoring systems that enhance emergency response capabilities while enabling predictive healthcare approaches. Benefits include enhanced operational efficiency through task automation and resource optimization, improved quality of care through consistent protocol execution and personalization, and addressing workforce challenges by extending staff capabilities and reducing physical strain. Despite these benefits, implementation faces significant ethical concerns regarding human contact reduction, privacy, and equitable access; technical limitations in language processing, emotional recognition, and physical capabilities; and adoption barriers, including costs, staff acceptance, and regulatory uncertainties. Future directions suggest technological advancements in AI capabilities and physical design, emerging research in personalization, cultural adaptation, and hybrid care models; and necessary policy developments regarding standards, data management, reimbursement structures, and certification requirements. The integration of robotic caregiving represents a transformative approach that, when thoughtfully implemented, can enhance both caregiver effectiveness and resident quality of life.

We present a unique occlusion and foreground overlap detection technique from depth sensor data using a fuzzy rule-based system. Features such as bounding box parameters and skeletonization were extracted from the foreground images and then input to the Fuzzy Inference System. Overlap and occlusion confidence measures were taken for each frame in the image sequence and compared against the extracted ground truth. This technique can help filter out occluded regions in the image sequence which, in an Eldercare environment, can then be used to compute accurate estimates of fall risk parameters such as stride time, stride length, and walking speed on a daily basis in in order to monitor the well-being of older adults in an ambient assisted living facility.

In this paper, we analyze the changes in family structure and explore the changes in detail, based on which we construct a neural network model of smart aging. Based on the gender perspective, the individual growth model in the multilayer linear model is used to examine the effects of family structure changes on the elderly in terms of economic exchange, daily care, and emotional support. The results show that there is no significant gender difference in the family structure changes on the elderly in terms of economic exchange and daily care, but there is a significant gender difference in terms of emotional support. To solve the problem of data imbalance in the daily activity categories of the elderly, this paper resamples the data and uses different neural network models for activity recognition of the sensor data generated from the daily activities of the elderly. In this paper, the daily behavior patterns of the elderly over a while are studied by correlating three conditions of time distance, optimal path, and sensor distance to discover the daily behavior patterns of the elderly, while the abnormal behavior patterns can be well separated by EM clustering algorithm. The daily behavior of the elderly is a coarse-grained representation of their daily activities. It is not limited to a specific activity and does not require the sensor ID, trigger time, and location triggered by the activity to be consistent, but in long-term daily activity data, it abstracts the general behavior rules of the elderly activities. Through the research of this paper, the existing system is improved, and the multifaceted needs of the elderly are fully considered, from housing needs to spiritual needs, to face the current elderly care problems with a positive attitude, create a good social elderly care environment for the elderly, and realize the real elderly care.

Urban environment is a key focus area for tourism activities, where local and micro-scale climatic conditions significantly influence the health, well-being, and behaviour of both tourists and residents. Climate change can alter the spatial distribution of urban areas in terms of their suitability for tourist activities. Consequently, monitoring the impact of climate system transformations on urban tourist destinations is crucial for engaging tourists and stakeholders in effective climate adaptation. However, current regional climate models lack the spatial resolution and detailed representation of local physical processes necessary to adequately capture urban-scale climatic dynamics. For meteorological variables such as air temperature, humidity, and wind speed, higher-resolution outputs derived from land surface models offer more representative and granular insights for impact assessments in urban contexts. In this study, the SURFEX land surface model is employed to assess the impacts of climate change on tourism in Szeged, Hungary. Specifically, the ALADIN-Climate regional model outputs are downscaled with SURFEX to a horizontal resolution of 1 km. Tourism climate conditions are evaluated for the periods 2041–2070 and 2071–2100 using the urban Holiday Climate Index and a modified form of the Tourism Climate Index. Future anthropogenic activities are described under the RCP4.5 and RCP8.5 emission scenarios. The uncertainties in the simulations are addressed through comparative analyses of the different indices, emission scenarios, and periods. The results of this study can contribute to the broader discourse on bioclimatic resilience, emphasizing the need for sustainable urban design and climate adaptation measures that mitigate adverse effects on health and well-being of tourists and residents.Project no. 142335 has been implemented with the support provided by the Ministry of Culture and Innovation of Hungary from the National Research, Development and Innovation Fund, financed under the PD_22 funding scheme.

Currently, immuno-checkpoint inhibitor (ICI) has been widely used for unresectable advanced/recurrent esophageal squamous cell carcinoma (ESCC) as 2nd or later line treatment, however clinical evidence regarding efficacy or safety of the 1st line ICI treatment for ESCC remains limited. The global phase 3 CheckMate 648 (ONO-4538-50) demonstrated significant overall survival (OS) benefit of nivolumab + chemotherapy and nivolumab+ ipilimumab versus chemotherapy (5FU/CDDP) in previously untreated patients with advanced ESCC with an acceptable safety profile. A total 970 ESCC patients with unresectable advanced/recurrent disease were randomized globally and this post-hoc exploratory subgroup analysis presented here includes only the 34 patients randomized and treated at this hospital during Dec, 2017-Dec, 2019. Treatment efficacy and adverse events were assessed among the three groups; nivolumab (3 mg/kg every 2 weeks) + ipilimumab (1 mg/kg every 6 weeks) (Group A, n = 11), nivolumab (240 mg every 2 weeks) + 5FU/CDDP (5FU800 mg/m2 day1–5/CDDP80 mg/m2 day 1、every 4 weeks) (Group B, n = 13) and 5FU/CDDP (5FU800 mg/m2 day1–5/CDDP80 mg/m2 day 1、every 4 weeks) (Group C, n = 10). Patient characteristics of the group A/B/C were as follows; age = 69 (52–82)/70 (52–85)/70 (55–75), gender (male/female) = 11:0/10:3/9:1. Regarding survival analysis (median follow-up time; 19.5 months) of Jan 24th, 2022, median OS were not reached in the group A, 15 months (95%CI = 5–28) in the group B and 12 (95%CI = 2–25) months in the group C. The most common treatment-related adverse events (≥ grade3) were pruritus in 2 cases, rash in 2 (Group A), neutropenia in 5 cases, leukopenia in 3, and anorexia in 3 (Group B) and no cases (Group C). No treatment-related death was identified in any groups. The results from this Japanese single-center exploratory subgroup analysis are consistent with the global study results, and support nivolumab + ipilimumab and nivolumab + chemotherapy as promising first-line treatments for patients with unresectable advanced/recurrent ESCC. Also, establishment of useful biomarker in distinguishing use of nivolumab + chemotherapy and nivolumab+ ipilimumab would be urgent.

Exacerbator subtypes in the tiotropium safety and performance in respimat (TIOSPIR™) trial

A major goal of human activity and behavior recognition (HAR) is to recognize activities and behaviors from a series of action data for different subjects under different environmental conditions. The use of wearables, smart devices, and vision-based systems enables the collection of human action and behavior data for greater health benefits, rehabilitation, elderly care, and monitoring. This chapter overviews the state-of-the-art in activity and behavior recognition based on sensor data. First, we provided a general architecture of HAR describing primary data sources and pre-processing techniques. Then, we describe robust feature extraction and selection strategies with a comparative analysis of statistical and deep learning-based models. Additionally, we provide information about more than 100 benchmark datasets and repositories in this domain. The paper concludes with a discussion of major issues and challenges, highlighting open issues and scopes that need to be approached in prospective research.