(1) Background: Diabetes mellitus (DM) is a chronic, metabolic disease characterized by elevated levels of blood glucose. Recently, some studies approached the diabetes care domain through the analysis of the modifications of cardiovascular system parameters. In fact, cardiovascular diseases are the first leading cause of death in diabetic subjects. Thanks to their cost effectiveness and their ease of use, electrocardiographic (ECG) and photoplethysmographic (PPG) signals have recently been used in diabetes detection, blood glucose estimation and diabetes-related complication detection. This review’s aim is to provide a detailed overview of all the published methods, from the traditional (non machine learning) to the deep learning approaches, to detect and manage diabetes using PPG and ECG signals. This review will allow researchers to compare and understand the differences, in terms of results, amount of data and complexity that each type of approach provides and requires. (2) Method: We performed a systematic review based on articles that focus on the use of ECG and PPG signals in diabetes care. The search was focused on keywords related to the topic, such as “Diabetes”, “ECG”, “PPG”, “Machine Learning”, etc. This was performed using databases, such as PubMed, Google Scholar, Semantic Scholar and IEEE Xplore. This review’s aim is to provide a detailed overview of all the published methods, from the traditional (non machine learning) to the deep learning approaches, to detect and manage diabetes using PPG and ECG signals. This review will allow researchers to compare and understand the differences, in terms of results, amount of data and complexity that each type of approach provides and requires. (3) Results: A total of 78 studies were included. The majority of the selected studies focused on blood glucose estimation (41) and diabetes detection (31). Only 7 studies focused on diabetes complications detection. We present these studies by approach: traditional, machine learning and deep learning approaches. (4) Conclusions: ECG and PPG analysis in diabetes care showed to be very promising. Clinical validation and data processing standardization need to be improved in order to employ these techniques in a clinical environment.

Vascular ageing (vascular ageing) is the deterioration of arterial structure and function which occurs naturally with age, and which can be accelerated with disease. Measurements of vascular ageing are emerging as markers of cardiovascular risk, with potential applications in disease diagnosis and prognosis, and for guiding treatments. However, vascular ageing is not yet routinely assessed in clinical practice. A key step towards this is the development of technologies to assess vascular ageing. In this Roadmap, experts discuss several aspects of this process, including: measurement technologies; the development pipeline; clinical applications; and future research directions. The Roadmap summarises the state of the art, outlines the major challenges to overcome, and identifies potential future research directions to address these challenges.

Photoplethysmography (PPG) is a non-invasive and cost-efficient optical technique used to assess blood volume variations in the microcirculation. PPG technology is widely used in a variety of wearable sensors to investigate the cardiovascular system. Recent studies have demonstrated the utility of PPG analysis for carrying out large-scale screening to prevent and detect diabetes. However, most of these studies require feature extraction and/or several pre-processing steps. Over the past few years, the advent of deep learning has significantly impacted the analysis of biomedical signals. Despite their success in other fields, however, very few studies have focused on the application of deep learning to raw PPG signals for detecting diabetes. Existing studies have proposed large models trained on large amounts of data. In this paper, we present a Light CNN-based model for screening the presence of type 2 diabetes using a single raw pulse extracted from photoplethysmographic signals. In addition to the baseline architecture, we evaluate different model architectures that take as input age and biological sex or PPG handcrafted features. Furthermore, we apply transfer learning to all the tested architectures to evaluate the effectiveness of harnessing pre-trained models in detecting diabetes. We tested a model pre-trained on a general PPG shape dataset and another model pre-trained on a dataset containing hypertension PPG signals. Our model scored an AUC of 75.5 when trained with raw PPG waves, age, and biological sex without applying transfer learning, which is competitive with current state of the art.

Photopletysmography (PPG) is a non-invasive and well known technology that enables the recording of the digital volume pulse (DVP). Although PPG is largely employed in research, several aspects remain unknown. One of these is represented by the lack of information about how many waveform classes best express the variability in shape. In the literature, it is common to classify DVPs into four classes based on the dicrotic notch position. However, when working with real data, labelling waveforms with one of these four classes is no longer straightforward and may be challenging. The correct identification of the DVP shape could enhance the precision and the reliability of the extracted bio markers. In this work we proposed unsupervised machine learning and deep learning approaches to overcome the data labelling limitations. Concretely we performed a K-medoids based clustering that takes as input 1) DVP handcrafted features, 2) similarity matrix computed with the Derivative Dynamic Time Warping and 3) DVP features extracted from a CNN AutoEncoder. All the cited methods have been tested first by imposing four medoids representative of the Dawber classes, and after by automatically searching four clusters. We then searched the optimal number of clusters for each method using silhouette score, the prediction strength and inertia. To validate the proposed approaches we analyse the dissimilarities in the clinical data related to obtained clusters.

Notre société connaît un changement transformateur dans la répartition des âges. Alors que l’espérance de vie continue d’augmenter et que les ressources médicales peinent à accueillir tout le monde, la télémédecine et les solutions de santé électronique devront jouer un rôle central dans les années à venir. Les maladies cardiovasculaires englobent un éventail de troubles affectant le cœur et les vaisseaux sanguins, notamment les crises cardiaques, les accidents vasculaires cérébraux et les maladies artérielles périphériques. En 2017, les maladies cardiovasculaires ont causé le plus grand nombre de décès dans le monde. Reconnaissant le besoin de solutions de santé à distance, les chercheurs et les industries se concentrent de plus en plus sur le développement de systèmes facilitant la surveillance à distance des patients. Parmi toutes les technologies disponibles, la photopléthysmographie suscite un vif intérêt au sein de la communauté de la recherche et de l’industrie. Son coût abordable et sa facilité d’utilisation permettent d’intégrer facilement cette technologie dans plusieurs dispositifs différents facilitant la collecte de données. Malgré les résultats prometteurs, la PPG présente certaines limitations et son analyse reste complexe. Le signal PPG est facilement corrompu par des artefacts de bruit. La lumière extérieure, les mouvements ou le bruit électrique peuvent affecter la forme de l’onde du pouls et rendre l’analyse qui en découle peu fiable. De plus, la forme d’onde photopléthysmographique dépend de la longueur d’onde de la lumière émise, du type de peau, du volume sanguin, de la phase du cycle cardiaque, du vieillissement et de la présence de pathologies. Nos contributions visent à relever certains de ces défis. Tout d’abord, nous proposons un modèle d’apprentissage en profondeur non basé sur des caractéristiques capable de classer les formes de pouls numériques physiologiques parmi les formes bruitées. La détection et l’exclusion ultérieure des formes bruitées de l’analyse améliorent la fiabilité de cette dernière. Nous examinons ensuite la possibilité de déterminer des modèles d’ondes de pouls qui pourraient expliquer la variété de formes d’onde de pouls trouvées dans les données du monde réel grâce à l’apprentissage non supervisé. Nous validons cette approche en analysant des données cliniques connexes, telles que l’âge, la vélocité de l’onde de pouls et la pression artérielle. Enfin, nous présentons une approche d’apprentissage en profondeur pour réaliser une évaluation du risque de diabète en analysant les modifications de la forme des ondes de pouls entre les sujets diabétiques et non diabétiques. Nous étudions également l’influence de la forme de l’onde de pouls et de la vélocité de l’onde de pouls en présence de la rétinopathie diabétique, une atteinte visuelle liée au diabète. Pour chacune de nos études, nous avons mené des expériences approfondies avec des résultats prometteurs.