Artificial intelligence (AI) has significantly improved drug discovery, particularly in drug-target interaction (DTI) prediction, by enabling faster and more accurate identification of therapeutic compounds. In this work, we propose a structure-aware deep learning framework that leverages graphbased representations for both drugs and proteins. Drugs are modeled using a Topology Adaptive Graph Convolutional Network (TAGCN), while proteins are represented as residue-level graphs processed through a dual Weighted Graph Convolutional Network (WGCN) incorporating distance and angle features. After multi-layer representation learning, a dual cross-attention module aligns relevant substructures between drugs and proteins to predict interactions more effectively. Evaluated on the updated Yamanishi benchmark dataset, our method outperforms several strong baselines, achieving an F1-score of 89% and a ROC-AUC of 95%. These results highlight the importance of integrating detailed structural information and cross-modal attention for advancing AI-driven DTI prediction.

The main goal of this research is to develop a a machine learning based method in order to detect cough from acceleration signals. In this study, two different methods are proposed: a conventional one that uses XGBoost as a classifier and a deep learning which uses CNN-1D as an architecture. We found that these models were able to distinguish between acceleration signals caused by coughing and acceleration signals caused by other activities such as clearing throat, talking, laughing and movements in different directions with high accuracy. This study affirms that cough monitoring based on accelerometer measurements generated by the Hexoskin device is possible, making it a new user-independent tool for cough detection.

The detection of epileptic seizures in EEG signals is a challenging task because it requires careful review of multi-channel EEG recordings over a lengthy time interval. In general, EEG-based seizure detection is strongly dependent on the ability to select descriptive features that are also stable in the sense that they are not sensitive to changes in the training data. This study proposes and investigates a patient-independent seizure detection model that uses stable EEG-based features obtained by comparing multiple feature selection methods. The schemes considered can be divided into five categories, often referred to as similarity, information theoretic, sparse learning, statistical, and graph centrality feature selection methods. The stability of the features these schemes produce was evaluated by random forest classification, using the intersection, frequency, correlation, and similarity measures. In experiments with Temple University Hospital’s Seizure database of EEG records of 341 patients, unsupervised graph centrality feature selection was the most effective method, with a correct classification rate of 91.5%.

Ballistocardiogram signals describe the mechanical activity of the heart. It can be measured by an intelligent mattress in a totally unobtrusive way during periods of rest in bed or sitting on a chair. The BCG signals are highly vulnerable to artefacts such as noise and movement making useful information like respiratory activities difficult to extract. The purpose of this study is to investigate a classification method to distinguish between seven types of respiratory activities such as normal breathing, cough and hold breath. We propose a feature selection method based on a spectral analysis namely spectral flatness measure (SFM) and spectral centroid (SC). The classification is carried out using the nearest neighbor classifier. The proposed method is able to discriminate between the seven classes with the accuracy of 94% which shows its usefulness in context of Telemedicine.

We discuss in this paper a study of a smart and unobtrusive mattress in a clinical setting on a population with cardiorespiratory problems. Up to recently, the vast majority of studies with unobtrusive sensors are done with healthy populations. The unobtrusive monitoring of the Respiratory Rate (RR) is essential for proposing better diagnoses. Thus, new industrial and research activity on smart mattresses is targeting respiratory rate in an Internet-of-Things (IoT) context. In our work, we are interested in the performances of a microbend fiber optic sensor (FOS) mattress on 81 subjects admitted in the Cardiac Intensive Care Unit (CICU) by estimating the RR from their ballistocardiograms (BCG). Our study proposes a new RR estimator, based on harmonic plus noise models (HNM) and compares it with known estimators such as MODWT and CLIE. The goal is to examine, using a more representative and bigger dataset, the performances of these methods and of the smart mattress in general. Results of applying these three estimators on the BCG show that MODWT is more accurate with an average mean absolute error (MAE) of $$1.97\,\pm \,2.12\;\text {BPM}$$ . However, the HNM estimator has space for improvements with estimation errors of $$2.91 \pm 4.07\;\text {BPM}$$ . The smart mattress works well within a standard RR range of 10–20 breaths-per-minute (BPM) but gets less accurate with a bigger range of estimation. These results highlight the need to test these sensors in much more realistic contexts .

Personalization of serious games is an important factor for motivating and engaging players. It requires the identification of players' profiles through the analysis of large volume of data including game data. This research study aims at identifying relevant data from an online serious game and the appropriate data mining methods for deduction of players' profiles. Multiple linear regression is applied to analyze the influence of player's characteristics on his performance. Moreover, clustering technique is used, in particular K-means, to extract players' clusters and to identify their common characteristics. The regression models showed that the number of access to the game, completed quests and advantages used contribute significantly to the scores and the gaming duration, while the clustering revealed three forms of players' participation: beginner, intermediate and advanced; who interact with the game according to their experiences.

Surgical instrumentation for the Adolescent idiopathic scoliosis (AIS) is a complex procedure involving many difficult decisions. Selection of the appropriate fusion level remains one of the most challenging decisions in scoliosis surgery. Currently, the Lenke classification model is generally followed in surgical planning. The purpose of our study is to investigate a computer aided method for Lenke classification and scoliosis fusion level selection. The method uses a self organizing neural network trained on a large database of surgically treated AIS cases. The neural network produces two maps, one of Lenke classes and the other of fusion levels. These two maps show that the Lenke classes are associated with the the proper fusion level categories everywhere in the map except at the Lenke class transitions. The topological ordering of the Cobb angles in the neural network justifies determining a patient scoliotic treatment instrumentation using directly the fusion level map rather than via the Lenke classification.

This study investigates Bayes classification of online Arabic characters represented by histograms of tangent differences and Gibbs modeling of the class-conditional probability density functions. The parameters of these Gibbs density functions are estimated following the Zhu, Wu, and Mumford constrained maximum entropy formalism, originally introduced for image and shape synthesis. We investigate two partition function estimation methods: one uses the training sample and the other draws from a reference distribution. The efficiency of the corresponding Bayes decision methods, and of a combination of these, is shown in experiments using a database of 9504 freely written samples by 22 scriptors. Comparisons to the nearest neighbor rule method and Kohonen neural network methods are provided.

The paper investigates a new representation of shape and its use in handwritten on-line character recognition. This representation is based on the empirical distribution of features such as tangents, and tangent differences at distant points along the character signal. Recognition is carried out by a Kohonen associative memory (also called Kohonen self organizing feature map), trained using this representation, and the Hellinger distance, which measures the distance between distributions. We report on extensive experiments that show the pertinence of the representation and the superior performance of the scheme.

The purpose of this study is to investigate a method for high performance on-line Arabic characters recognition. This method is based on the use of Kohonen maps and their corresponding confusion matrices which serve to prune them of error-causing nodes, and to combine them consequently. We use two Kohonen maps obtained using two distinct character representations, namely, Fourier descriptors and tangents extracted along the characters online signals. The two Kohonen maps are then combined using a majority vote decision rule that, for each character, favors the most reliable map. This combination, without adding in any significant way to the process complexity, affords a much better recognition rate.