Memhans: an Assistive Device for Dementia Patients

Superior voice recognition in a patient with acquired prosopagnosia and object agnosia

A palliative approach to care aims to meet the needs of patients and caregivers throughout a chronic disease trajectory and can be delivered by non-palliative specialists. There is an important gap in understanding the perspectives and experiences of primary care providers on an integrated palliative approach in dementia care and the impact of existing programs and models to this end. To address these, we undertook a scoping review. We searched five databases; and used descriptive numerical summary and narrative synthesizing approaches for data analysis. We found that: (1) difficulty with prognostication and a lack of interdisciplinary and intersectoral collaboration are obstacles to using a palliative approach in primary care; and (2) a palliative approach results in statistically and clinically significant impacts on community-dwelling individuals, specifically those with later stages of dementia. There is a need for high-quality research studies examining the integrated palliative approach models and initiation of these models sooner in the care trajectory for persons living with mild and moderate stages of dementia in the community.

The Providing Enhanced Access to Child Health Services (PEACH) initiative at the Sydney Children’s Hospitals Network (SCHN) aims to reduce inequity in healthcare access and outcomes for priority population children and young people (including Aboriginal and/or Torres Strait Islander, culturally and linguistically diverse, from refugee or asylum seeker backgrounds, living in out-of-home care or recipients of the National Disability Insurance Scheme). Central to PEACH is active children and caregivers engagement through co-design, recognising the importance of understanding their experiences to facilitate enhanced healthcare provision. This qualitative study explored the experiences of priority population children and caregivers at SCHN before the implementation of PEACH initiatives, with the aim of identifying strengths, limitations, and recommendations for enhancing healthcare services tailored to priority population needs. Guided by an experience-based co-design methodology, 28 qualitative semi-structured interviews were carried out with a total of 38 participants from priority population group(s) who had accessed care at SCHN. The data were analysed via inductive thematic analysis. The findings are presented through the lens of the concept of identity, delineated into personal and social identity. Personal identity is related to person before problem and demography, while social identity is linked to cultural responsiveness. Participants emphasised both the significance of social identities in health service design, and the importance of an individualised approach that recognises and respects their unique personal identities within the broader social context. This research advocates for a person-centred approach to healthcare delivery that recognises and responds to the dynamic interconnected relationship between personal and social identities.

PurposeThe purpose of this paper is to explore a broader understanding of the role of Cypriot school principals’ personal identities, through a values system perspective, when exercising their leadership.Design/methodology/approachA multicase study methodology was followed with five school principals, representing five different leadership styles. In each case, an in-depth investigation of the school principal’s personal identity was undertaken. School principals’ personal values were explored during interviews, staff meetings and daily activities observations, as well as through the use of the think-aloud protocol method. This study utilized the Schwartz Theory of Basic Human Values, as well as the Pashiardis–Brauckmann Holistic Leadership Framework, as the guiding theoretical framework.FindingsSchool principals’ personal identities in Cyprus seem to influence, to some extent, their daily leadership practice. However, particular factors associated with the context in which they live (social identity) and work (professional identity), seemed to be affecting the personal values embedded throughout their personal identities.Research limitations/implicationsFive school principals are not enough to make generalizations on the relationships between leadership styles and values. However, through this paper, the authors sought to provide examples on how school principals’ personal identities influence their leadership practice.Practical implicationsThe findings highlight the important role and attention to school principals’ personal identities, beyond the core management and leadership courses. The findings also shed light on the importance of looking more closely at contextual elements “outside” and “inside” the school and to what extent these could influence school principals’ personal identities.Originality/valueThis paper offers insights into school principals’ personal identities, through a values system perspective, and how these personal identities influence their leadership practice.

In this paper we investigate multimodal biometric person and object identification. We model this process of multimodal identification as multimodal multiple hypothesis testing with independent modalities. We analyze theoretical performance limits that can be attained in such a multimodal protocol in terms of exponents of average error probability. Furthermore, we address a privacy related issues in this paper. In particular, we consider performance/privacy trade-off due to the indirect multimodal identification performed in a secret subspace and approximate the obtained performance limits using properties of random projections. Finally, a set of experiments exemplifies our findings.

The increasing average human age and a growing number of old age population over the globe have emerged the need for automation in the health care domain. Eventually, health care robots will contribute to manage the workload of the care providers in the future health care domain. Elderly people face a major challenge to identify and pronounce names of daily use objects and different family and friends due to dementia-related issues. One of the many tasks by AI-Enabled elderly care humanoid robots will be communication with such patients. They will aid them to remember objects and persons they find difficult to identify and recognize due to dementia. This study has investigated the potential of the JD humanoid robot for object identification in such a scenario. The study has presented a prototype AI-Enabled elderly care robot that can aid elder persons to identify different objects. The robot is trained on ImageNet data set for object identification. Further, a personalized data set comprising of faces of different persons labeled by names of family and friends are also used to train robots for family recognition aid. The JD Humanoid robot captures an image of an object/person and also tracks it as the object/face moves. It identifies input images and outputs in audio format the name of the object/person to guide the elderly person. The third feature of the robot is the notifications of daily routine tasks for elder people like reminders for prayer, exercise walk, and medicine intake. Testing results have shown an object and person identification with 95% accuracy.

Nurses Have Special Strategies for Dementia

As Alzheimer's advances, it develops into a degenerative neurological condition which significantly affects brain connectivity and progressively decreases memory and cognitive abilities. For prompt intervention and better patient care, early diagnosis is essential. By employing brain MRI scans in combination with convolutional neural networks (CNN), this study suggests an automated method for recognizing and categorizing the successive phases of Alzheimer's progression. W ith the help of Python & TensorFlow, the model is trained on four types of MRI images: Normal cognition, very mild dementia stage, mild dementia stage, and moderate dementia stage. To improve the effectiveness of the algorithm, the approach utilizes preprocessing methods including over-sampling and augmentation. To facilitate real-time categorization through straightforward MRI picture uploads, an intuitive Flask-based web application was also created. Primary performance metrics were employed to assess the model, and for the nondemented class, an AUC of 0.94 was recorded for the model, with corresponding F1-score, precision, and recall values of 0.90, 0.89, and 0.92 respectively. The F1- scores for the remaining classes are obtained between 0.69 and 0.79, suggesting consistent performance in closely related categories with room for improvement. This work demonstrates the promise of AI in improving early-stage Alzheimer's detection and acts as a proof-of-concept for incorporating deep learning into healthcare diagnostics.

The elderly population in the United States is increasing exponentially in tandem with risk for frailty. Frailty is described by a clinically significant state where a patient is at risk for developing complications requiring increased assistance in daily activities. Frailty syndrome studied in geriatric patients is responsible for an increased risk for falls, and increased mortality. In efforts to prepare for and to intervene in perioperative complications and general frailty, a universal scale to measure frailty is necessary. Many methods for determining frailty have been developed, yet there remains a need to define clinical frailty and, therefore, the most effective way to measure it. This article reviews six popular scales for measuring frailty and evaluates their clinical effectiveness demonstrated in previous studies. By identifying the most time-efficient, criteria comprehensive, and clinically effective scale, a universal scale can be implemented into standard of care and reduce complications from frailty in both non-surgical and surgical settings, especially applied to the perioperative surgical home model. We suggest further evaluation of the Edmonton Frailty Scale for inclusion in patient care.

High expansion of urban population and infrastructure complemented by recent geopolitical world events, triggered an increasingly alarming number of threats. Law enforcement authorities are now challenged to redesign societal security concepts. Although current technological advances eased the information access, e.g., via video surveillance cameras, satellite data, drones, wearables, manual analysis of such big and diverse data to extract strategic knowledge is not a solution anymore. There is the critical need for automatic solutions. Artificial Intelligence (AI) and the breakthrough of deep neural networks opened a new perspective for providing such solutions with a human-grade accuracy. Here, we provide a snapshot of our AI research for counter terrorism on automatic person and object identification, retrieval of speech intelligence, and dissimulated behavior analysis. The research was carried out during the UEFISCDI SPIA-VA research project at CAMPUS Research Center, University “Politehnica” of Bucharest, with the participation of UTI Grup and of the Military Equipment and Technologies Research Agency (ACTTM), having as public beneficiary the Protection Guard Service, Romania (SPP).

This research investigates the language choices and digital identities represented in text messages exchanged by high school students in the New Capital City of Indonesia (Ibu Kota Nusantara/IKN) and the implications of these linguistic practices for language teaching. Using a corpus of messages transmitted via WhatsApp collected from 100 high school students, this study produced 2.1 million tokens of 83,414 word types. This study uses corpus linguistic analysis to investigate the distinctive features of language usage, lexical changes, and communication trends in the digital discourse of IKN high school students. Data analysis utilizes features from AntConc corpus tools, including word lists, collocations, concordances, and N-grams. Concerning the features uncovered the subtle nuances that contribute to the unique linguistic identity of the student authors. This research highlights how digital language patterns attribute individual identity within the context of IKN high school students. The data analysis discovers three linguistic patterns in the electronic texts produced by IKN High School students, reflecting their general digital identity as language users in IKN, namely (1) lexical choice, (2) orthographic selection, and (3) lexical bundles. Furthermore, the study reveals the complex construction of digital identity through language, with students negotiating social positions, connections, and personal identities via their linguistic choices in electronic texts. The research findings contribute to the implications of language teaching in the IKN area and a deeper understanding of how students within the IKN area express themselves through language in the virtual realm, thus shaping their digital identities.

BackgroundWe aimed to describe specific changes in brain perfusion in patients with dementia with Lewy bodies (DLB) at both the prodromal (also called mild cognitive impairment) and mild dementia stages, relative to patients with Alzheimer’s disease (AD) and controls.MethodsAltogether, 96 participants in five groups (prodromal DLB, prodromal AD, DLB with mild dementia, AD with mild dementia, and healthy elderly controls) took part in an arterial spin labeling MRI study. Three analyses were performed: a global perfusion value comparison, a voxel-wise analysis of both absolute and relative perfusion, and a linear discriminant analysis. These were used to assess the global decrease in perfusion, regional changes, and the sensitivity and specificity of these changes.ResultsPatterns of perfusion in DLB differed from AD and controls in both the prodromal stage and dementia, DLB having more deficits in frontal, insular, and temporal cortices whereas AD showed reduced perfusion in parietal and parietotemporal cortices. Decreases but also increases of perfusion in DLB relative to controls were observed in both absolute and relative measurements. All these regional changes of perfusion classified DLB patients with respect to either healthy controls or AD with sensitivity from 87 to 100 % and specificity from 90 to 96 % depending on the stage of the disease.ConclusionsOur results are consistent with previous studies. We extend the scope of those studies by integrating prodromal DLB patients and by describing both hypo- and hyperperfusion in DLB. While decreases in perfusion may relate to functional impairments, increases might suggest a functional compensation of some brain areas.Electronic supplementary materialThe online version of this article (doi:10.1186/s13195-016-0196-8) contains supplementary material, which is available to authorized users.

Truth that Matters

With the rapid increase of interest in independent living along with the continual growth of the global aging population, there is a significant need for developing technological solutions that are trustworthy, functional, and scalable. Towards this vision, we developed a novel framework that provides an organic, evolving, and immersive assistive interaction between human, robot, and their environment. Our contribution is in the system design and the computing aspect of the proposed solution. At the system level, we are utilizing the concept of the Internet of Robotic Things (IoRT) to integrate smart personal devices and smart space automation technologies with social companion robot(s) and mixed/augmented reality smart glasses. At the computing level, we propose utilizing advancements in the field of Artificial Intelligence (AI) and Machine Learning (ML) towards providing task assistance in daily activities, smart space control, and improving emotion and mood through Affective Computing. We provide details of our IoRT system integration and quantitative experimental results on Facial Emotion Recognition (FER) which is one of the core components for Affective Computing. We improved FER utilizing Deep Learning on the edge and then deployed it on our Social Companion Robot (SCR) called Miro. We integrated the SCR with a Mixed Reality smart glass (Microsoft Hololens 2) and smart home automaton hub (Google). While the smart automation controller is directly in charge of controlling the environment (e.g., lighting, temperature), the robot can increase the control level with a more context-aware decision providing a socially embodied control.KeywordsHuman-robot-interactionInternet of Robotic Things (IoRT)Social companion robotsSmart home automationSocial embodimentAffective computingMixed/augmented realityIndependent living

Theoretical accounts have offered a general understanding of the social significance and importance of dignity and suggest that older age may threaten dignity by structuring and limiting the opportunities for participation and/or social recognition. Micro-sociological research has shown how older people negotiate their identity, in the face of its erosion by the aging body and disability and the domination of health and social care workers. These theoretical approaches inform the study presented here, which explores the meaning and experience of dignity for older people in their daily lives. Older people's accounts were elicited through a series of focus groups with participants living in England and Wales. Participants were selected according to predefined criteria to represent a range of different socio-economic and ethnic backgrounds, levels of fitness, and home circumstances. Personal identity and autonomy were the aspects of dignity most meaningful to the participants, and discussions tended to focus mainly on when personal identity and autonomy were threatened or violated in the context of the provision of health and social care. The authors discuss the extent to which older people's discourse on dignity resonates with the theoretical discourse.