Anastasios Delopoulos

Aristotelis Ballas, Vasileios Papapanagiotou, Anastasios Delopoulos et al.

Heart murmurs are abnormal sounds present in heartbeats, caused by turbulent blood flow through the heart. The PhysioNet 2022 challenge targets automatic detection of murmur from audio recordings of the heart and automatic detection of normal vs. abnormal clinical outcome. The recordings are captured from multiple locations around the heart. Our participation investigates the effectiveness of selfsupervised learning for murmur detection. We train the layers of a backbone CNN in a self-supervised way with data from both this year's and the 2016 challenge. We use two different augmentations on each training sample, and normalized temperature-scaled cross-entropy loss. We experiment with different augmentations to learn effective phonocardiogram representations. To build the final detectors we train two classification heads, one for each challenge task. We present evaluation results for all combinations of the available augmentations, and for our multipleaugmentation approach. Our team's, Listen2YourHeart, SSL murmur detection classifier received a weighted accuracy score of 0.737 (ranked 13th out of 40 teams) and an outcome identification challenge cost score of 11946 (ranked 7th out of 39 teams) on the hidden test set.

Alexandros Papadopoulos, Anastasios Delopoulos

Data-driven approaches for remote detection of Parkinson's Disease and its motor symptoms have proliferated in recent years, owing to the potential clinical benefits of early diagnosis. The holy grail of such approaches is the free-living scenario, in which data are collected continuously and unobtrusively during every day life. However, obtaining fine-grained ground-truth and remaining unobtrusive is a contradiction and therefore, the problem is usually addressed via multiple-instance learning. Yet for large scale studies, obtaining even the necessary coarse ground-truth is not trivial, as a complete neurological evaluation is required. In contrast, large scale collection of data without any ground-truth is much easier. Nevertheless, utilizing unlabelled data in a multiple-instance setting is not straightforward, as the topic has received very little research attention. Here we try to fill this gap by introducing a new method for combining semi-supervised with multiple-instance learning. Our approach builds on the Virtual Adversarial Training principle, a state-of-the-art approach for regular semi-supervised learning, which we adapt and modify appropriately for the multiple-instance setting. We first establish the validity of the proposed approach through proof-of-concept experiments on synthetic problems generated from two well-known benchmark datasets. We then move on to the actual task of detecting PD tremor from hand acceleration signals collected in-the-wild, but in the presence of additional completely unlabelled data. We show that by leveraging the unlabelled data of 454 subjects we can achieve large performance gains (up to 9% increase in F1-score) in per-subject tremor detection for a cohort of 45 subjects with known tremor ground-truth.

Vasileios Papapanagiotou, Christos Diou, Anastasios Delopoulos

The importance of automated and objective monitoring of dietary behavior is becoming increasingly accepted. The advancements in sensor technology along with recent achievements in machine-learning--based signal-processing algorithms have enabled the development of dietary monitoring solutions that yield highly accurate results. A common bottleneck for developing and training machine learning algorithms is obtaining labeled data for training supervised algorithms, and in particular ground truth annotations. Manual ground truth annotation is laborious, cumbersome, can sometimes introduce errors, and is sometimes impossible in free-living data collection. As a result, there is a need to decrease the labeled data required for training. Additionally, unlabeled data, gathered in-the-wild from existing wearables (such as Bluetooth earbuds) can be used to train and fine-tune eating-detection models. In this work, we focus on training a feature extractor for audio signals captured by an in-ear microphone for the task of eating detection in a self-supervised way. We base our approach on the SimCLR method for image classification, proposed by Chen et al. from the domain of computer vision. Results are promising as our self-supervised method achieves similar results to supervised training alternatives, and its overall effectiveness is comparable to current state-of-the-art methods. Code is available at https://github.com/mug-auth/ssl-chewing .

Christos Siargkas, Vasileios Papapanagiotou, Anastasios Delopoulos

Transportation mode recognition (TMR) is a critical component of human activity recognition (HAR) that focuses on understanding and identifying how people move within transportation systems. It is commonly based on leveraging inertial, location, or both types of signals, captured by modern smartphone devices. Each type has benefits (such as increased effectiveness) and drawbacks (such as increased battery consumption) depending on the transportation mode (TM). Combining the two types is challenging as they exhibit significant differences such as very different sampling rates. This paper focuses on the TMR task and proposes an approach for combining the two types of signals in an effective and robust classifier. Our network includes two sub-networks for processing acceleration and location signals separately, using different window sizes for each signal. The two sub-networks are designed to also embed the two types of signals into the same space so that we can then apply an attention-based multiple-instance learning classifier to recognize TM. We use very low sampling rates for both signal types to reduce battery consumption. We evaluate the proposed methodology on a publicly available dataset and compare against other well known algorithms.

Ioannis Levi, Konstantinos Kyritsis, Vasileios Papapanagiotou et al.

Accurate monitoring of eating behavior is crucial for managing obesity and eating disorders such as bulimia nervosa. At the same time, existing methods rely on multiple and/or specialized sensors, greatly harming adherence and ultimately, the quality and continuity of data. This paper introduces a novel approach for estimating the weight of a bite, from a commercial smartwatch. Our publicly-available dataset contains smartwatch inertial data from ten participants, with manually annotated start and end times of each bite along with their corresponding weights from a smart scale, under semi-controlled conditions. The proposed method combines extracted behavioral features such as the time required to load the utensil with food, with statistical features of inertial signals, that serve as input to a Support Vector Regression model to estimate bite weights. Under a leave-one-subject-out cross-validation scheme, our approach achieves a mean absolute error (MAE) of 3.99 grams per bite. To contextualize this performance, we introduce the improvement metric, that measures the relative MAE difference compared to a baseline model. Our method demonstrates a 17.41% improvement, while the adapted state-of-the art method shows a -28.89% performance against that same baseline. The results presented in this work establish the feasibility of extracting meaningful bite weight estimates from commercial smartwatch inertial sensors alone, laying the groundwork for future accessible, non-invasive dietary monitoring systems.

Ioannis Sarafis, Alexandros Papadopoulos, Anastasios Delopoulos

In this paper, we propose a weakly supervised semantic segmentation approach for food images which takes advantage of the zero-shot capabilities and promptability of the Segment Anything Model (SAM) along with the attention mechanisms of Vision Transformers (ViTs). Specifically, we use class activation maps (CAMs) from ViTs to generate prompts for SAM, resulting in masks suitable for food image segmentation. The ViT model, a Swin Transformer, is trained exclusively using image-level annotations, eliminating the need for pixel-level annotations during training. Additionally, to enhance the quality of the SAM-generated masks, we examine the use of image preprocessing techniques in combination with single-mask and multi-mask SAM generation strategies. The methodology is evaluated on the FoodSeg103 dataset, generating an average of 2.4 masks per image (excluding background), and achieving an mIoU of 0.54 for the multi-mask scenario. We envision the proposed approach as a tool to accelerate food image annotation tasks or as an integrated component in food and nutrition tracking applications.

Athanasios Kirmizis, Konstantinos Kyritsis, Anastasios Delopoulos

The consumption of tobacco has reached global epidemic proportions and is characterized as the leading cause of death and illness. Among the different ways of consuming tobacco (e.g., smokeless, cigars), smoking cigarettes is the most widespread. In this paper, we present a two-step, bottom-up algorithm towards the automatic and objective monitoring of cigarette-based, smoking behavior during the day, using the 3D acceleration and orientation velocity measurements from a commercial smartwatch. In the first step, our algorithm performs the detection of individual smoking gestures (i.e., puffs) using an artificial neural network with both convolutional and recurrent layers. In the second step, we make use of the detected puff density to achieve the temporal localization of smoking sessions that occur throughout the day. In the experimental section we provide extended evaluation regarding each step of the proposed algorithm, using our publicly available, realistic Smoking Event Detection (SED) and Free-living Smoking Event Detection (SED-FL) datasets recorded under semi-controlled and free-living conditions, respectively. In particular, leave-one-subject-out (LOSO) experiments reveal an F1-score of 0.863 for the detection of puffs and an F1-score/Jaccard index equal to 0.878/0.604 towards the temporal localization of smoking sessions during the day. Finally, to gain further insight, we also compare the puff detection part of our algorithm with a similar approach found in the recent literature.

Vasileios Papapanagiotou, Stefanos Ganotakis, Anastasios Delopoulos

While automatic tracking and measuring of our physical activity is a well established domain, not only in research but also in commercial products and every-day life-style, automatic measurement of eating behavior is significantly more limited. Despite the abundance of methods and algorithms that are available in bibliography, commercial solutions are mostly limited to digital logging applications for smart-phones. One factor that limits the adoption of such solutions is that they usually require specialized hardware or sensors. Based on this, we evaluate the potential for estimating the weight of consumed food (per bite) based only on the audio signal that is captured by commercial ear buds (Samsung Galaxy Buds). Specifically, we examine a combination of features (both audio and non-audio features) and trainable estimators (linear regression, support vector regression, and neural-network based estimators) and evaluate on an in-house dataset of 8 participants and 4 food types. Results indicate good potential for this approach: our best results yield mean absolute error of less than 1 g for 3 out of 4 food types when training food-specific models, and 2.1 g when training on all food types together, both of which improve over an existing literature approach.

Alexandros Papadopoulos, Fotis Topouzis, Anastasios Delopoulos

Diabetic Retinopathy (DR) is a leading cause of vision loss globally. Yet despite its prevalence, the majority of affected people lack access to the specialized ophthalmologists and equipment required for assessing their condition. This can lead to delays in the start of treatment, thereby lowering their chances for a successful outcome. Machine learning systems that automatically detect the disease in eye fundus images have been proposed as a means of facilitating access to DR severity estimates for patients in remote regions or even for complementing the human expert's diagnosis. In this paper, we propose a machine learning system for the detection of referable DR in fundus images that is based on the paradigm of multiple-instance learning. By extracting local information from image patches and combining it efficiently through an attention mechanism, our system is able to achieve high classification accuracy. Moreover, it can highlight potential image regions where DR manifests through its characteristic lesions. We evaluate our approach on publicly available retinal image datasets, in which it exhibits near state-of-the-art performance, while also producing interpretable visualizations of its predictions.

Konstantinos Kyritsis, Christos Diou, Anastasios Delopoulos

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 heldout experiments, respectively). Experiments are performed using our publicly available FIC and the newly introduced FreeFIC datasets.

Alexandros Papadopoulos, Konstantinos Kyritsis, Lisa Klingelhoefer et al.

Parkinson's Disease (PD) is a slowly evolving neuro-logical disease that affects about 1% of the population above 60 years old, causing symptoms that are subtle at first, but whose intensity increases as the disease progresses. Automated detection of these symptoms could offer clues as to the early onset of the disease, thus improving the expected clinical outcomes of the patients via appropriately targeted interventions. This potential has led many researchers to develop methods that use widely available sensors to measure and quantify the presence of PD symptoms such as tremor, rigidity and braykinesia. However, most of these approaches operate under controlled settings, such as in lab or at home, thus limiting their applicability under free-living conditions. In this work, we present a method for automatically identifying tremorous episodes related to PD, based on IMU signals captured via a smartphone device. We propose a Multiple-Instance Learning approach, wherein a subject is represented as an unordered bag of accelerometer signal segments and a single, expert-provided, tremor annotation. Our method combines deep feature learning with a learnable pooling stage that is able to identify key instances within the subject bag, while still being trainable end-to-end. We validate our algorithm on a newly introduced dataset of 45 subjects, containing accelerometer signals collected entirely in-the-wild. The good classification performance obtained in the conducted experiments suggests that the proposed method can efficiently navigate the noisy environment of in-the-wild recordings.

Ioannis Sarafis, Christos Diou, Anastasios Delopoulos

Weighted SVM (or fuzzy SVM) is the most widely used SVM variant owning its effectiveness to the use of instance weights. Proper selection of the instance weights can lead to increased generalization performance. In this work, we extend the span error bound theory to weighted SVM and we introduce effective hyperparameter selection methods for the weighted SVM algorithm. The significance of the presented work is that enables the application of span bound and span-rule with weighted SVM. The span bound is an upper bound of the leave-one-out error that can be calculated using a single trained SVM model. This is important since leave-one-out error is an almost unbiased estimator of the test error. Similarly, the span-rule gives the actual value of the leave-one-out error. Thus, one can apply span bound and span-rule as computationally lightweight alternatives of leave-one-out procedure for hyperparameter selection. The main theoretical contributions are: (a) we prove the necessary and sufficient condition for the existence of the span of a support vector in weighted SVM; and (b) we prove the extension of span bound and span-rule to weighted SVM. We experimentally evaluate the span bound and the span-rule for hyperparameter selection and we compare them with other methods that are applicable to weighted SVM: the $K$-fold cross-validation and the $ξ-α$ bound. Experiments on 14 benchmark data sets and data sets with importance scores for the training instances show that: (a) the condition for the existence of span in weighted SVM is satisfied almost always; (b) the span-rule is the most effective method for weighted SVM hyperparameter selection; (c) the span-rule is the best predictor of the test error in the mean square error sense; and (d) the span-rule is efficient and, for certain problems, it can be calculated faster than $K$-fold cross-validation.

Angelos Katharopoulos, Despoina Paschalidou, Christos Diou et al.

This paper introduces a family of local feature aggregation functions and a novel method to estimate their parameters, such that they generate optimal representations for classification (or any task that can be expressed as a cost function minimization problem). To achieve that, we compose the local feature aggregation function with the classifier cost function and we backpropagate the gradient of this cost function in order to update the local feature aggregation function parameters. Experiments on synthetic datasets indicate that our method discovers parameters that model the class-relevant information in addition to the local feature space. Further experiments on a variety of motion and visual descriptors, both on image and video datasets, show that our method outperforms other state-of-the-art local feature aggregation functions, such as Bag of Words, Fisher Vectors and VLAD, by a large margin.