Search

Achieving safe collaboration between humans and robots in an industrial work-cell requires effective communication. This can be achieved through a robot perception system developed using data-driven machine learning. The challenge for human–robot communication is the availability of extensive, labelled datasets for training. Due to the variations in human behaviour and the impact of environmental conditions on the performance of perception models, models trained on standard, publicly available datasets fail to generalize well to domain and application-specific scenarios. Thus, model personalization involving the adaptation of such models to the individual humans involved in the task in the given environment would lead to better model performance. A novel framework is presented that leverages robust modes of communication and gathers feedback from the human partner to auto-label the mode with the sparse dataset. The strength of the contribution lies in using in-commensurable multimodes of inputs for personalizing models with user-specific data. The personalization through feedback-enabled human–robot communication (PF-HRCom) framework is implemented on the use of facial expression recognition as a safety feature to ensure that the human partner is engaged in the collaborative task with the robot. Additionally, PF-HRCom has been applied to a real-time human–robot handover task with a robotic manipulator. The perception module of the manipulator adapts to the user's facial expressions and personalizes the model using feedback. Having said that, the framework is applicable to other combinations of multimodal inputs in human–robot collaboration applications. [ABSTRACT FROM AUTHOR]

This paper presents an on-device semi-supervised human activity detection system that can learn and predict human activity patterns in real time. The clinical objective is to monitor and detect the unhealthy sedentary lifestyle of a user. The proposed semi-supervised learning (SSL) framework uses sparsely labelled user activity events acquired from Inertial Measurement Unit sensors installed as wearable devices. The proposed cluster-based learning model in this approach is trained with data from the same target user, thus preserving data privacy while providing personalized activity detection services. Two different cluster labelling strategies, namely, population-based and distance-based strategies, are employed to achieve the desired classification performance. The proposed system is shown to be highly accurate and computationally efficient for different algorithmic parameters, which is relevant in the context of limited computing resources on typical wearable devices. Extensive experimentation and simulation study have been conducted on multi-user human activity data from the public domain in order to analyze the trade-off between classification accuracy and computation complexity of the proposed learning paradigm with different algorithmic hyper-parameters. With 4.17 h of training time for 8000 activity episodes, the proposed SSL approach consumes at most 20 KB of CPU memory space, while providing a maximum accuracy of 90% and 100% classification rates. [ABSTRACT FROM AUTHOR]

IntroductionInsomnia causes serious adverse health effects and is estimated to affect 10–30% of the worldwide population. This study leverages personalized fine-tuned machine learning algorithms to detect insomnia risk based on questionnaire and longitudinal objective sleep data collected by a smart bed platform.MethodsUsers of the Sleep Number smart bed were invited to participate in an IRB approved study which required them to respond to four questionnaires (which included the Insomnia Severity Index; ISI) administered 6 weeks apart from each other in the period from November 2021 to March 2022. For 1,489 participants who completed at least 3 questionnaires, objective data (which includes sleep/wake and cardio-respiratory metrics) collected by the platform were queried for analysis. An incremental, passive-aggressive machine learning model was used to detect insomnia risk which was defined by the ISI exceeding a given threshold. Three ISI thresholds (8, 10, and 15) were considered. The incremental model is advantageous because it allows personalized fine-tuning by adding individual training data to a generic model.ResultsThe generic model, without personalizing, resulted in an area under the receiving-operating curve (AUC) of about 0.5 for each ISI threshold. The personalized fine-tuning with the data of just five sleep sessions from the individual for whom the model is being personalized resulted in AUCs exceeding 0.8 for all ISI thresholds. Interestingly, no further AUC enhancements resulted by adding personalized data exceeding ten sessions.DiscussionThese are encouraging results motivating further investigation into the application of personalized fine tuning machine learning to detect insomnia risk based on longitudinal sleep data and the extension of this paradigm to sleep medicine.

Stress is widely recognized as a major contributor to a variety of health issues. Stress prediction using biosignal data recorded by wearables is a key area of study in mobile sensing research because real-time stress prediction can enable digital interventions to immediately react at the onset of stress, helping to avoid many psychological and physiological symptoms such as heart rhythm irregularities. Electrodermal activity (EDA) is often used to measure stress. However, major challenges with the prediction of stress using machine learning include the subjectivity and sparseness of the labels, a large feature space, relatively few labels, and a complex nonlinear and subjective relationship between the features and outcomes. To tackle these issues, we examined the use of model personalization: training a separate stress prediction model for each user. To allow the neural network to learn the temporal dynamics of each individual's baseline biosignal patterns, thus enabling personalization with very few labels, we pre-trained a one-dimensional convolutional neural network (1D CNN) using self-supervised learning (SSL). We evaluated our method using the Wearable Stress and Affect Detection(WESAD) dataset. We fine-tuned the pre-trained networks to the stress-prediction task and compared against equivalent models without any self-supervised pre-training. We discovered that embeddings learned using our pre-training method outperformed the supervised baselines with significantly fewer labeled data points: the models trained with SSL required less than 30% of the labels to reach equivalent performance without personalized SSL. This personalized learning method can enable precision health systems that are tailored to each subject and require few annotations by the end user, thus allowing for the mobile sensing of increasingly complex, heterogeneous, and subjective outcomes such as stress. [ABSTRACT FROM AUTHOR]

Stress is widely recognized as a major contributor to a variety of health issues. Stress prediction using biosignal data recorded by wearables is a key area of study in mobile sensing research because real-time stress prediction can enable digital interventions to immediately react at the onset of stress, helping to avoid many psychological and physiological symptoms such as heart rhythm irregularities. Electrodermal activity (EDA) is often used to measure stress. However, major challenges with the prediction of stress using machine learning include the subjectivity and sparseness of the labels, a large feature space, relatively few labels, and a complex nonlinear and subjective relationship between the features and outcomes. To tackle these issues, we examined the use of model personalization: training a separate stress prediction model for each user. To allow the neural network to learn the temporal dynamics of each individual’s baseline biosignal patterns, thus enabling personalization with very few labels, we pre-trained a one-dimensional convolutional neural network (1D CNN) using self-supervised learning (SSL). We evaluated our method using the Wearable Stress and Affect Detection(WESAD) dataset. We fine-tuned the pre-trained networks to the stress-prediction task and compared against equivalent models without any self-supervised pre-training. We discovered that embeddings learned using our pre-training method outperformed the supervised baselines with significantly fewer labeled data points: the models trained with SSL required less than 30% of the labels to reach equivalent performance without personalized SSL. This personalized learning method can enable precision health systems that are tailored to each subject and require few annotations by the end user, thus allowing for the mobile sensing of increasingly complex, heterogeneous, and subjective outcomes such as stress.

Affect-aware socially assistive robotics (SAR) has shown great potential for augmenting interventions for children with autism spectrum disorders (ASD). However, current SAR cannot yet perceive the unique and diverse set of atypical cognitive-affective behaviors from children with ASD in an automatic and personalized fashion in long-term (multi-session) real-world interactions. To bridge this gap, this work designed and validated personalized models of arousal and valence for children with ASD using a multi-session in-home dataset of SAR interventions. By training machine learning (ML) algorithms with supervised domain adaptation (s-DA), the personalized models were able to tradeoff between the limited individual data and the more abundant less personal data pooled from other study participants. We evaluated the effects of personalization on a long-term multimodal dataset consisting of four children with ASD with a total of 19 sessions, and derived inter-rater reliability (IR) scores for binary arousal (IR = 83%) and valence (IR = 81%) labels between human annotators. Our results show that personalized Gradient Boosted Decision Trees (XGBoost) models with s-DA outperformed two non-personalized individualized and generic model baselines not only on the weighted average of all sessions, but also statistically (p <.05) across individual sessions. This work paves the way for the development of personalized autonomous SAR systems tailored toward individuals with atypical cognitive-affective and socio-emotional needs. [ABSTRACT FROM AUTHOR]

Smartphone-based approaches for Human Activity Recognition have become prevalent in recent years. Despite the amount of research undertaken in the field, issues such as cross-subject variability are still posing an obstacle to the deployment of solutions in large scale, free-living settings. Personalized methods (i.e. aiming to adapt a generic classifier to a specific target user) attempt to solve this problem. The lack of labeled data for training purposes, however, represents a major barrier. This is especially the case when taking into consideration that personalization generally requires labeled data to be user-specific. This paper presents a novel personalization method combining a semi-population based approach with user adaptation. Personalization is achieved through the following. Firstly, the proposed method identifies a subset of users from the available population as best candidates for initializing the classifier to the target user. Subsequently, a semi-population Neural Network classifier is trained using data from this subset of users. The classifier's network weights are then updated using a small amount of labeled data from the target user subsequently implementing personalization. This approach was validated on a large publicly available dataset collected in a free-living scenario. The personalized approach using the proposed method has shown to improve the overall F-score to 74.4% compared to 70.9% when using a generic non-personalized approach. Results obtained, with statistical significance being confirmed on a set of 57 users, indicate that model initialization using the semi-population approach can reduce the amount of labeled data required for personalization. As such, the proposed method for model initialization could facilitate the real-world deployment of systems implementing personalization by reducing the amount of data needed for personalization.

The accurate measure of fetal fraction is important to ensure the results of noninvasive prenatal testing. However, measuring fetal fraction could require a substantial amount of data and additional costs. Therefore, this study proposes an alternative method of measuring fetal fraction with a limited sample size and low sequencing reads. Adaptive machine-learning algorithms customized to each laboratory's environment, were used to measure fetal fraction. Pregnant women with female fetuses were tested to exclude the bias caused by training data of women carrying male fetuses. The accuracy of fetal DNA fraction prediction was enhanced by increasing the training sample size. When trained with 1,000 samples (males) and tested with 45 samples (females), the optimal bin sizes using the read count and size features were 300 kb and 800 kb, respectively. Comparing the new 300-kb bin to the 50-kb bin used by SeqFF with 4,000-5,000 training samples, the correlation was approximately 3-5% higher with the 300-kb bin. Therefore, we propose an effective and tailored method to measure fetal fraction in individual laboratories with limited sample collecting conditions and relatively low-coverage sequencing data.

A wearable system that can continuously track the fluctuation of blood pressure (BP) based on pulse signals is highly desirable for the treatments of cardiovascular diseases, yet the sensitivity, reliability, and accuracy remain challenging. Since the correlations of pulse waveforms to BP are highly individualized due to the diversity of the patients' physiological characteristics, wearable sensors based on universal designs and algorithms often fail to derive BP accurately when applied on individual patients. Herein, a wearable triboelectric pulse sensor based on a biomimetic nanopillar layer was developed and coupled with Personalized Machine Learning (ML) to provide accurate and continuous monitoring of BP. Flexible conductive nanopillars as the triboelectric layer were fabricated through soft lithography replication of a cicada wing, which could effectively enhance the sensor's output performance to detect weak signal characteristics of pulse waveform for BP derivation. The sensors were coupled with a personalized Partial Least-Squares Regression (PLSR) ML to derive unknown BP based on individual pulse characteristics with reasonable accuracy, avoiding the issue of individual variability that was encountered by General PLSR ML or formula algorithms. The cuffless and intelligent design endow this ML-sensor as a highly promising platform for the care and treatments of hypertensive patients.