In recent times, several studies have presented single-modality systems for non-contact biosignal monitoring. While these systems often yield estimations correlating with clinical-grade devices, their practicality is limited due to constraints in real-time processing, scalability, and interoperability. Moreover, these studies have seldom explored the combined use of multiple modalities or the integration of various sensors. Addressing these gaps, we introduce a distributed computing architecture designed to remotely acquire biosignals from both radars and cameras. This architecture is supported by conceptual blocks that distribute tasks across sensing, computing, data management, analysis, communication, and visualization. Emphasizing interoperability, our system leverages RESTful APIs, efficient video streaming, and standardized health-data protocols. Our framework facilitates the integration of additional sensors and improves signal analysis efficiency. While the architecture is conceptual, its feasibility has been evaluated through simulations targeting specific challenges in networked remote photoplethysmography (rPPG) systems. Additionally, we implemented a prototype to demonstrate the architectural principles in action, with modules and blocks operating in independent threads. This prototype specifically involves the analysis of biosignals using mmWave radars and RGB cameras, illustrating the potential for the architecture to be adapted into a fully distributed system for real-time biosignal processing.
Many cloud and shadow detection methods have been proposed already, but improvements can be made on accuracy or automation. In this study, we propose a Fully Convolutional Network model for the detection of clouds and shadows in optical satellite images. The proposed model was trained on 165 Landsat images in Finland, and tested on an independent set of images. The cloud and shadow detection accuracy reached 95%, outperforming both quantitatively and qualitatively a selection of other deep learning architectures.
In this article, we introduce a new approach for estimating the heart rate from noisy photoplethysmography (PPG) signals. We propose the use of two-dimensional representations of signals that are fed into a residual deep neural network that performs the regression task. Our approach leverages transfer learning and pre-trained models to further reduce the prediction error, resulting in state-of-the-art results in a challenging benchmark dataset.
This study investigates the use of explainable artificial intelligence (AI) for predicting sales offer success in the business-to-business (B2B) context of configure-to-order industrial equipment. Using real sales data from over 400 sales cases and extracted from customer relationship management and Configure-Price-Quote systems of an industrial partner, we built a gradient boosted decision tree (XGBoost) model to predict win/loss outcomes of sales cases. Model explainability was enabled through Shapley additive explanations (SHAP) to identify and quantify feature contributions to individual predictions. The predictive performance of the model was compared against assessments made by experienced sales experts using the same cases. The model achieved balanced accuracies of 83.9% on the full dataset and 74.1% on a regional subset, substantially outperforming expert predictions, which reached 63.0 and 54.4%, respectively. To evaluate the practical value of explainable AI, 5 sales experts reviewed SHAP-based explanations through a structured questionnaire using real regional sales cases. The results show partial alignment between model-identified important features and expert reasoning, while also revealing features with strong predictive power that experts regarded as unintuitive or unimportant. Despite these discrepancies, experts reported that the explanations were understandable and actionable in a sales context. In particular, they perceived the system as useful for prioritizing sales opportunities and identifying ways to modify offers to increase success probability. Overall, the findings demonstrate that explainable AI can outperform human experts in sales success prediction while also providing interpretable insights that support decision-making in B2B sales practice.
8 September 2026 in Applied Artificial Intelligence