An opportune early diagnosis of Alzheimer’s disease (AD) would help to overcome symptoms and improve the quality of life for AD patients. Research studies have identified early manifestations of AD that occur years before the diagnosis. For instance, eye movements of people with AD in different tasks differ from eye movements of control subjects. In this review, we present a summary and evolution of research approaches that use eye tracking technology and computational analysis to measure and compare eye movements under different tasks and experiments. Furthermore, this review is targeted to the feasibility of pioneer work on developing computational tools and techniques to analyze eye movements under naturalistic scenarios. We describe the progress in technology that can enhance the analysis of eye movements everywhere while subjects perform their daily activities and give future research directions to develop tools to support early AD diagnosis through analysis of eye movements.

Activity recognition, a key component in pervasive healthcare monitoring, relies on classification algorithms that require labeled data of individuals performing the activity of interest to train accurate models. Labeling data can be performed in a lab setting where an individual enacts the activity under controlled conditions. The ubiquity of mobile and wearable sensors allows the collection of large datasets from individuals performing activities in naturalistic conditions. Gathering accurate data labels for activity recognition is typically an expensive and time-consuming process. In this paper we present two novel approaches for semi-automated online data labeling performed by the individual executing the activity of interest. The approaches have been designed to address two of the limitations of self-annotation: (i) The burden on the user performing and annotating the activity, and (ii) the lack of accuracy due to the user labeling the data minutes or hours after the completion of an activity. The first approach is based on the recognition of subtle finger gestures performed in response to a data-labeling query. The second approach focuses on labeling activities that have an auditory manifestation and uses a classifier to have an initial estimation of the activity, and a conversational agent to ask the participant for clarification or for additional data. Both approaches are described, evaluated in controlled experiments to assess their feasibility and their advantages and limitations are discussed. Results show that while both studies have limitations, they achieve 80% to 90% precision.

Driver distraction is one of the major causes of accidents. Most methods for inferring distracted driving behaviors are vision-based systems that determine the head's orientation. One of the significant challenges of this approach is to develop robust algorithms that detect face and eye features under various lighting conditions. Our approach is based on inferring the vehicle's cabin spot drawing the driver's attention through head-mounted inertial sensors. To achieve this aim, we collected accelerometer, gyroscope, and magnetometer data from ten participants who drove under semi-naturalistic conditions. We generated classifiers by using the Support Vector Machine (SVM linear and RBF), k-nearest neighbor (k-NN), and Random Forest (RF) machine learning techniques. These techniques, except SVM linear, produced an accuracy, precision and recall higher than 96%. Our results demonstrate that raw signals collected from the inertial sensors provide enough information about the head posture associated with the car's cabin spot.

In this work, we aim at identifying the speaker in interactions between mothers and children with Down syndrome (DS) using audio. We collected audio from a session in which children with DS solved puzzles, and their mothers were by their side. We generated a dataset by manually annotating human speech activity and non-speech. We used machine learning to perform four experiments, including individual and generalized models achieving on average F1-scores of 0.74 (five-class model) and 0.84 (two-class) for individual models and 0.69 (five-class) and 0.82 (two-class) for the generalized models. Our results can be helpful to behavioral researchers, therapists, and those interested in better understanding how mother–child interactions unfold in naturalistic settings.