The impact of the number of high temporal resolution water meters on the determinism of water consumption in a district metered area.

Developments in data mining techniques have significantly influenced the progress of Intelligent Water Systems (IWSs). Learning about the hydraulic conditions enables the development of increasingly reliable predictive models of water consumption. The non-stationary, non-linear, and inherent stochasticity of water consumption data at the level of a single water meter means that the characteristics of its determinism remain impossible to observe and their burden of randomness creates interpretive difficulties. A deterministic model of water consumption was developed based on data from high temporal resolution water meters. Seven machine learning algorithms were used and compared to build predictive models. In addition, an attempt was made to estimate how many water meters data are needed for the model to bear the hallmarks of determinism. The most accurate model was obtained using Support Vector Regression (8.9%) and the determinism of the model was achieved using time series from eleven water meters of multi-family buildings.

Entropy Measurements for Leukocytes' Surrounding Informativeness Evaluation for Acute Lymphoblastic Leukemia Classification.

The study of leukemia classification using deep learning techniques has been conducted by multiple research teams worldwide. Although deep convolutional neural networks achieved high quality of sick vs. healthy patient discrimination, their inherent lack of human interpretability of the decision-making process hinders the adoption of deep learning techniques in medicine. Research involving deep learning proved that distinguishing between healthy and sick patients using microscopic images of lymphocytes is possible. However, it could not provide information on the intermediate steps in the diagnosis process. As a result, despite numerous examinations, it is still unclear whether the lymphocyte is the only object in the microscopic picture containing leukemia-related information or if the leukocyte's surroundings also contain the desired information. In this work, entropy measures and machine learning models were applied to study the informativeness of both whole images and lymphocytes' surroundings alone for Leukemia classification. This work aims to provide human-interpretable features marking the probability of sickness occurrence. The research stated that the hue distribution of images with lymphocytes obfuscated alone is informative enough to facilitate 93.0% accuracy in healthy vs. sick classification. The research was conducted on the ALL-IDB2 dataset.

Study of the Few-Shot Learning for ECG Classification Based on the PTB-XL Dataset.

The electrocardiogram (ECG) is considered a fundamental of cardiology. The ECG consists of P, QRS, and T waves. Information provided from the signal based on the intervals and amplitudes of these waves is associated with various heart diseases. The first step in isolating the features of an ECG begins with the accurate detection of the R-peaks in the QRS complex. The database was based on the PTB-XL database, and the signals from Lead I-XII were analyzed. This research focuses on determining the Few-Shot Learning (FSL) applicability for ECG signal proximity-based classification. The study was conducted by training Deep Convolutional Neural Networks to recognize 2, 5, and 20 different heart disease classes. The results of the FSL network were compared with the evaluation score of the neural network performing softmax-based classification. The neural network proposed for this task interprets a set of QRS complexes extracted from ECG signals. The FSL network proved to have higher accuracy in classifying healthy/sick patients ranging from 93.2% to 89.2% than the softmax-based classification network, which achieved 90.5-89.2% accuracy. The proposed network also achieved better results in classifying five different disease classes than softmax-based counterparts with an accuracy of 80.2-77.9% as opposed to 77.1% to 75.1%. In addition, the method of R-peaks labeling and QRS complexes extraction has been implemented. This procedure converts a 12-lead signal into a set of R waves by using the detection algorithms and the k-mean algorithm.

IoT Application of Transfer Learning in Hybrid Artificial Intelligence Systems for Acute Lymphoblastic Leukemia Classification.

Acute lymphoblastic leukemia is the most common cancer in children, and its diagnosis mainly includes microscopic blood tests of the bone marrow. Therefore, there is a need for a correct classification of white blood cells. The approach developed in this article is based on an optimized and small IoT-friendly neural network architecture. The application of learning transfer in hybrid artificial intelligence systems is offered. The hybrid system consisted of a MobileNet v2 encoder pre-trained on the ImageNet dataset and machine learning algorithms performing the role of the head. These were the XGBoost, Random Forest, and Decision Tree algorithms. In this work, the average accuracy was over 90%, reaching 97.4%. This work proves that using hybrid artificial intelligence systems for tasks with a low computational complexity of the processing units demonstrates a high classification accuracy. The methods used in this study, confirmed by the promising results, can be an effective tool in diagnosing other blood diseases, facilitating the work of a network of medical institutions to carry out the correct treatment schedule.

Deep Learning Techniques in the Classification of ECG Signals Using R-Peak Detection Based on the PTB-XL Dataset.

Deep Neural Networks (DNNs) are state-of-the-art machine learning algorithms, the application of which in electrocardiographic signals is gaining importance. So far, limited studies or optimizations using DNN can be found using ECG databases. To explore and achieve effective ECG recognition, this paper presents a convolutional neural network to perform the encoding of a single QRS complex with the addition of entropy-based features. This study aims to determine what combination of signal information provides the best result for classification purposes. The analyzed information included the raw ECG signal, entropy-based features computed from raw ECG signals, extracted QRS complexes, and entropy-based features computed from extracted QRS complexes. The tests were based on the classification of 2, 5, and 20 classes of heart diseases. The research was carried out on the data contained in a PTB-XL database. An innovative method of extracting QRS complexes based on the aggregation of results from established algorithms for multi-lead signals using the k-mean method, at the same time, was presented. The obtained results prove that adding entropy-based features and extracted QRS complexes to the raw signal is beneficial. Raw signals with entropy-based features but without extracted QRS complexes performed much worse.

ECG Signal Classification Using Deep Learning Techniques Based on the PTB-XL Dataset.

The analysis and processing of ECG signals are a key approach in the diagnosis of cardiovascular diseases. The main field of work in this area is classification, which is increasingly supported by machine learning-based algorithms. In this work, a deep neural network was developed for the automatic classification of primary ECG signals. The research was carried out on the data contained in a PTB-XL database. Three neural network architectures were proposed: the first based on the convolutional network, the second on SincNet, and the third on the convolutional network, but with additional entropy-based features. The dataset was divided into training, validation, and test sets in proportions of 70%, 15%, and 15%, respectively. The studies were conducted for 2, 5, and 20 classes of disease entities. The convolutional network with entropy features obtained the best classification result. The convolutional network without entropy-based features obtained a slightly less successful result, but had the highest computational efficiency, due to the significantly lower number of neurons.

Changes of Conformation in Albumin with Temperature by Molecular Dynamics Simulations.

This work presents the analysis of the conformation of albumin in the temperature range of 300 K - 312 K , i.e., in the physiological range. Using molecular dynamics simulations, we calculate values of the backbone and dihedral angles for this molecule. We analyze the global dynamic properties of albumin treated as a chain. In this range of temperature, we study parameters of the molecule and the conformational entropy derived from two angles that reflect global dynamics in the conformational space. A thorough rationalization, based on the scaling theory, for the subdiffusion Flory-De Gennes type exponent of 0 . 4 unfolds in conjunction with picking up the most appreciable fluctuations of the corresponding statistical-test parameter. These fluctuations coincide adequately with entropy fluctuations, namely the oscillations out of thermodynamic equilibrium. Using Fisher's test, we investigate the conformational entropy over time and suggest its oscillatory properties in the corresponding time domain. Using the Kruscal-Wallis test, we also analyze differences between the root mean square displacement of a molecule at various temperatures. Here we show that its values in the range of 306 K - 309 K are different than in another temperature. Using the Kullback-Leibler theory, we investigate differences between the distribution of the root mean square displacement for each temperature and time window.

Cross-Entropy as a Metric for the Robustness of Drone Swarms.

Due to their growing number and increasing autonomy, drones and drone swarms are equipped with sophisticated algorithms that help them achieve mission objectives. Such algorithms vary in their quality such that their comparison requires a metric that would allow for their correct assessment. The novelty of this paper lies in analysing, defining and applying the construct of cross-entropy, known from thermodynamics and information theory, to swarms. It can be used as a synthetic measure of the robustness of algorithms that can control swarms in the case of obstacles and unforeseen problems. Based on this, robustness may be an important aspect of the overall quality. This paper presents the necessary formalisation and applies it to a few examples, based on generalised unexpected behaviour and the results of collision avoidance algorithms used to react to obstacles.

Hyaluronan-Chondroitin Sulfate Anomalous Crosslinking Due to Temperature Changes.

Glycosaminoglycans are a wide class of biopolymers showing great lubricating properties due to their structure and high affinity to water. Two of them, hyaluronic acid and chondroitin sulfate, play an important role in articular cartilage lubrication. In this work, we present results of the all-atom molecular dynamics simulations of both molecules placed in water-based solution. To mimic changes of the physiological conditions, especially temperature, of the synovial fluid in joints under successive load (e.g., walking, jogging, jumping), simulations have been performed at different physiological temperatures in the range of 300 to 320 Kelvin (normal intra-articular temperature is 305 K). The stability of the biopolymeric network at equilibrium (isothermal and isobaric) conditions has been studied. To understand the process of physical crosslinking, the dynamics of intra- and intermolecular hydrogen bonds forming and breaking have been studied. The results show that following addition of chondroitin sulfate, hyaluronan creates more intermolecular hydrogen bonds than when in homogeneous solution. The presence of chondroitin in a hyaluronan network is beneficial as it may increase its stability. Presented data show hyaluronic acid and chondroitin sulfate as viscosity modifiers related to their crosslinking properties in different physicochemical conditions.

Anomalous Behavior of Hyaluronan Crosslinking Due to the Presence of Excess Phospholipids in the Articular Cartilage System of Osteoarthritis.

Lubrication of articular cartilage is a complex multiscale phenomenon in synovial joint organ systems. In these systems, synovial fluid properties result from synergistic interactions between a variety of molecular constituent. Two molecular classes in particular are of importance in understanding lubrication mechanisms: hyaluronic acid and phospholipids. The purpose of this study is to evaluate interactions between hyaluronic acid and phospholipids at various functionality levels during normal and pathological synovial fluid conditions. Molecular dynamic simulations of hyaluronic acid and phospholipids complexes were performed with the concentration of hyaluronic acid set at a constant value for two organizational forms, extended (normal) and coiled (pathologic). The results demonstrated that phospholipids affect the crosslinking mechanisms of hyaluronic acid significantly and the influence is higher during pathological conditions. During normal conditions, hyaluronic acid and phospholipid interactions seem to have no competing mechanism to that of the interaction between hyaluronic acid to hyaluronic acid. On the other hand, the structures formed under pathologic conditions were highly affected by phospholipid concentration.

Molecular Dynamic Analysis of Hyaluronic Acid and Phospholipid Interaction in Tribological Surgical Adjuvant Design for Osteoarthritis.

Tribological surgical adjuvants constitute a therapeutic discipline made possible by surgical advances in the treatment of damaged articular cartilage beyond palliative care. The purpose of this study is to analyze interactions between hyaluronic acid and phospholipid molecules, and the formation of geometric forms, that play a role in the facilitated lubrication of synovial joint organ systems. The analysis includes an evaluation of the pathologic state to detail conditions that may be encountered by adjuvants during surgical convalescence. The synovial fluid changes in pH, hyaluronic acid polydispersity, and phospholipid concentration associated with osteoarthritis are presented as features that influence the lubricating properties of adjuvant candidates. Molecular dynamic simulation studies are presented, and the Rouse model is deployed, to rationalize low molecular weight hyaluronic acid behavior in an osteoarthritic environment of increased pH and phospholipid concentration. The results indicate that the hyaluronic acid radius of gyration time evolution is both pH- and phospholipid concentration-dependent. Specifically, dipalmitoylphosphatidylcholine induces hydrophobic interactions in the system, causing low molecular weight hyaluronic acid to shrink and at high concentration be absorbed into phospholipid vesicles. Low molecular weight hyaluronic acid appears to be insufficient for use as a tribological surgical adjuvant because an increased pH and phospholipid concentration induces decreased crosslinking that prevents the formation of supramolecular lubricating forms. Dipalmitoylphosphatidylcholine remains an adjuvant candidate for certain clinical situations. The need to reconcile osteoarthritic phenotypes is a prerequisite that should serve as a framework for future adjuvant design and subsequent tribological testing.