The role of human activity recognition in healthcare : a study focusing on patients suffering from chronic illnesses

Abstract

Human activity recognition substantiates itself as an appropriate advocate for primary care. As a matter of fact, it is an indispensable element offering an inherent possibility for medical centres to improve their operational degree of efficiency and effectiveness. Within the field of pervasive sensing, in the context of health, there are still ineffectual standards for individuals grieving debilitating maladies. This dissertation aimed to traverse the quintessential clinical pathways obliged to take advantage of the rapid technological progressions of ubiquity. Primarily, the investigation intended to delineate chronic maladies based on disease criticality regarding the benefits gained using omnipresent sensing. Secondly, this exposition planned to decipher the accumulated information utilising the Cumulative Illness Rating Scale and the RAND/UCLA Appropriateness Method. Finally, this inquiry endeavoured to explore how the subject matter can be implemented prospectively in business processes to support debilitated patients. The research concentrated on understanding specific procedures in the setting of pervasive electronic monitoring. Saint Vincent de Paul Residence, a local nursing home and hospital for the elderly presented a pivotal context in this study. Sequentially, the compilation of data transpired following a series of interviews and two meticulous focus groups constituted of domain experts in the medical field. After the retrieval of erudition, the Friedman's test distinguished every mean rating scores rendered. Moreover, the analysis of variance test identified the benchmark for the corresponding mean appropriateness and necessity scores. These entanglements exhibit the advantages of human activity recognition in comparison to seven elected lifelong illnesses. Consequently, the rating towards pervasiveness is significantly higher in dementia, depression, atrial fibrillation, hypertension, chronic obstructive pulmonary disease and congestive heart failure. Inversely, however, the difference between the remaining experimented maladies does not deviate significantly. As a matter of fact, on the antithesis, diabetes is the only tested illness that did not exceed the 0.05 level of significance. In conclusion, the thesis focused on acknowledging specific mechanisms about ubiquitous computing in healthcare to inscribe the predicament of insufficient data employing wearable devices. The scope of this commitment is to aid further researchers and practitioners operating on hospital systems to revamp the designs and the requirements of substantial managerial and medical problems.