In the context of dietary management, accurate monitoring of eating habits is receiving increased attention. Wearable sensors, combined with the connectivity and processing of modern smartphones, can be used to robustly extract objective and real-time measurements of human behavior. In particular, for the task of chewing detection, several approaches based on an in-ear microphone can be found in the literature, while other types of sensors have also been reported, such as strain sensors. In this paper, performed in the context of the SPLENDID project, we propose to combine an in-ear microphone with a photoplethysmography (PPG) sensor placed in the ear concha, in a new high accuracy and low sampling rate prototype chewing detection system. We propose a pipeline that initially processes each sensor signal separately, and then fuses both to perform the final detection. Features are extracted from each modality, and support vector machine (SVM) classifiers are used separately to perform snacking detection. Finally, we combine the SVM scores from both signals in a late-fusion scheme, which leads to increased eating detection accuracy. We evaluate the proposed eating monitoring system on a challenging, semifree living dataset of 14 subjects, which includes more than 60 h of audio and PPG signal recordings. Results show that fusing the audio and PPG signals significantly improves the effectiveness of eating event detection, achieving accuracy up to 0.938 and class-weighted accuracy up to 0.892.

Overweight and obesity are both associated with in-meal eating parameters such as eating speed. Recently, the plethora of available wearable devices in the market ignited the interest of both the scientific community and the industry toward unobtrusive solutions for eating behavior monitoring. In this paper, we present an algorithm for automatically detecting the in-meal food intake cycles using the inertial signals (acceleration and orientation velocity) from an off-the-shelf smartwatch. We use five specific wrist micromovements to model the series of actions leading to and following an intake event (i.e., bite). Food intake detection is performed in two steps. In the first step, we process windows of raw sensor streams and estimate their micromovement probability distributions by means of a convolutional neural network. In the second step, we use a long short-term memory network to capture the temporal evolution and classify sequences of windows as food intake cycles. Evaluation is performed using a challenging dataset of 21 meals from 12 subjects. In our experiments, we compare the performance of our algorithm against three state-of-theart approaches, where our approach achieves the highest F1 detection score (0.913 in the leave-one-subject-out experiment). The dataset used in the experiments is available at https://mug.ee.auth.gr/intake-cycle-detection/.

The increased worldwide prevalence of obesity has sparked the interest of the scientific community towards tools that objectively and automatically monitor eating behavior. Despite the study of obesity being in the spotlight, such tools can also be used to study eating disorders (e.g. anorexia nervosa) or provide a personalized monitoring platform for patients or athletes. This paper presents a complete framework towards the automated i) modeling of in-meal eating behavior and ii) temporal localization of meals, from raw inertial data collected in-the-wild using commercially available smartwatches. Initially, we present an end-to-end Neural Network which detects food intake events (i.e. bites). The proposed network uses both convolutional and recurrent layers that are trained simultaneously. Subsequently, we show how the distribution of the detected bites throughout the day can be used to estimate the start and end points of meals, using signal processing algorithms. We perform extensive evaluation on each framework part individually. Leave-one-subject-out (LOSO) evaluation shows that our bite detection approach outperforms four state-of-the-art algorithms towards the detection of bites during the course of a meal (0.923 F1 score). Furthermore, LOSO and held-out set experiments regarding the estimation of meal start/end points reveal that the proposed approach outperforms a relevant approach found in the literature (Jaccard Index of 0.820 and 0.821 for the LOSO and held-out experiments, respectively). Experiments are performed using our publicly available FIC and the newly introduced FreeFIC datasets.

Despite their immense success in numerous fields, machine and deep learning systems have not yet been able to firmly establish themselves in mission-critical applications in healthcare. One of the main reasons lies in the fact that when models are presented with previously unseen, Out-of-Distribution samples, their performance deteriorates significantly. This is known as the Domain Generalization (DG) problem. Our objective in this work is to propose a benchmark for evaluating DG algorithms, in addition to introducing a novel architecture for tackling DG in biosignal classification. In this article, we describe the Domain Generalization problem for biosignals, focusing on electrocardiograms (ECG) and electroencephalograms (EEG) and propose and implement an open-source biosignal DG evaluation benchmark. Furthermore, we adapt state-of-the-art DG algorithms from computer vision to the problem of 1D biosignal classification and evaluate their effectiveness. Finally, we also introduce a novel neural network architecture that leverages multi-layer representations for improved model generalizability. By implementing the above DG setup we are able to experimentally demonstrate the presence of the DG problem in ECG and EEG datasets. In addition, our proposed model demonstrates improved effectiveness compared to the baseline algorithms, exceeding the state-of-the-art in both datasets. Recognizing the significance of the distribution shift present in biosignal datasets, the presented benchmark aims at urging further research into the field of biomedical DG by simplifying the evaluation process of proposed algorithms. To our knowledge, this is the first attempt at developing an open-source framework for evaluating ECG and EEG DG algorithms.

Automatic objective monitoring of eating behavior using inertial sensors is a research problem that has received a lot of attention recently, mainly due to the mass availability of IMUs and the evidence on the importance of quantifying and monitoring eating patterns. In this paper we propose a method for detecting food intake cycles during the course of a meal using a commercially available wristband. We first model micro-movements that are part of the intake cycle and then use HMMs to model the sequences of micro-movements leading to mouthfuls. Evaluation is carried out on an annotated dataset of 8 subjects where the proposed method achieves 0:78 precision and 0:77 recall. The evaluation dataset is publicly available at http://mug.ee.auth.gr/intake-cycle-detection/.

Detecting chewing sounds from a microphone placed inside the outer ear for eating behaviour monitoring still remains a challenging task. This is mainly due the difficulty in discriminating non-chewing sounds (e.g. speech or sounds caused by walking) from chews, as well as due to to the high variability of the chewing sounds of different food types. Most approaches rely on detecting distictive structures on the sound wave, or on extracting a set of features and using a classifier to detect chews. In this work, we propose to use feature-learning in the time domain with 1-dimensional convolutional neural networks for for chewing detection. We apply a network of convolutional layers followed by fully connected layers directly on windows of the audio samples to detect chewing activity, and then aggregate individual chews to eating events. Experimental results on a large, semi-free living dataset collected in the context of the SPLENDID project indicate high effectiveness, with an accuracy of 0.980 and F1 score of 0.883.