SmartFABER: Recognizing fine-grained abnormal behaviors for early detection of mild cognitive impairment

Riboni, Daniele; Bettini, Cláudio; Civitarese, Gabriele; Janjua, Zaffar Haider; Helaoui, Rim

doi:10.1016/j.artmed.2015.12.001

Cited by 88 publications

(55 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data-driven methods rely on a training set of sensor data, labeled with executed activities, and machine learning algorithms to build the activities' model. Observations regarding the user's surrounding environment (in particular, objects' use), possibly coupled with body-worn sensor data, are the basis of those activity recognition systems [3,55]. In [56] the authors propose a time series data analysis method to segment sequences of sensor events in order to recognize ADLs.…”

Section: Data-driven Methodsmentioning

confidence: 99%

“…In Chapter 3, we propose a hybrid method to recognize ADLs which is based on a combination of supervised learning and knowledge-based conditions to refine the statistical predictions [3]. The proposed technique combines data-driven and knowledge-driven methods in order to exploit the strong points of both approaches.…”

Section: Hybrid Techniques To Recognize Adlsmentioning

confidence: 99%

“…In Chapter 6, we propose a novel technique for fine-grained abnormal behavior recognition [31,3,17]. Our method relies on medical models describing abnormal activity routines that may indicate the onset of early symptoms of MCI.…”

Section: Fine-grained and Long-term Anomalies Recognitionmentioning

confidence: 99%

“…In this chapter we will only focus on the evaluation of activity recognition. Results on anomaly recognition can be found in [3]. The following ADLs have been selected to validate our method:…”

Section: A Smart Lab Datasetmentioning

confidence: 99%

“…The majority of older adults prefer to stay in the comfort of their own homes, and given the costs of nursing home care [2], it is imperative to develop technologies that help older adults to age in place. For these reasons, independent living and pro-active health-care are becoming strategic application areas for major research programmes all over the world [3], considering that the senior population is projected to double as a percentage over the whole population in the next decades [4]. Indeed, many research groups are studying pervasive solutions to continuously and unobtrusively monitor fragile subjects at their homes, reducing health-care costs and supporting the medical diagnosis.…”

Section: Introduction 11 Motivation and Problem Descriptionmentioning

confidence: 99%

See 4 more Smart Citations

Human Activity Recognition in Smart-Home Environments for Health-Care Applications

Civitarese¹

2019

2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)

Self Cite

View full text Add to dashboard Cite

With a growing population of elderly people, the number of subjects at risk of cognitive disorders is rapidly increasing. Many research groups are studying pervasive solutions to continuously and unobtrusively monitor fragile subjects in their homes, reducing health-care costs and supporting the medical diagnosis. Clinicians are interested in monitoring several behavioral aspects for a wide variety of applications: early diagnosis, emergency monitoring, assessment of cognitive disorders, etcetera. Among the several behavioral aspects of interest, anomalous behaviors while performing activities of daily living (ADLs) are of great importance. Indeed, these anomalies can be indicators of serious cognitive diseases like Mild Cognitive Impairment. The recognition of such abnormal behaviors relies on robust and accurate ADLs recognition systems. Moreover, in order to enable unobtrusive and privacy-aware monitoring, environmental sensors in charge of unobtrusively capturing the interaction of the subject with the home infrastructure should be preferred. This thesis presents several contributions on this topic. The major ones are two novel hybrid ADLs recognition algorithms. The former is supervised while the latter is unsupervised. Preliminary results, which still need to be confirmed, show that the recognition rate of the unsupervised method is comparable to the one obtained by the supervised one, with the great advantage of not requiring the acquisition of an annotated dataset. Beyond ADLs recognition, other contributions on smart sensing and anomaly recognition are presented. Regarding unobtrusive sensing, we propose a machine learning technique to detect fine-grained manipulations performed by the inhabitant on household objects instrumented with tiny accelerometer sensors. Finally, a novel rule-based framework for the recognition of fine-grained abnormal behaviors is presented. Experimental results on several datasets show the effectiveness of all the proposed techniques.ii

show abstract

Section: Data-driven Methodsmentioning

confidence: 99%

Section: Hybrid Techniques To Recognize Adlsmentioning

confidence: 99%