Chronic wounds are a syndrome that affects around 4% of the world population due to several pathologies. The COV-19 pandemic has enforced the need of developing new techniques and technologies that can help clinicians to monitor the affected patients easily and reliably. In this prospective observational study a new device, the Wound Viewer, that works through a memristor-based Discrete-Time Cellular Neural Network (DT-CNN) has been developed and tested through a clinical trial of 150 patients. The WV has been developed to serve as the state-of-art tool, capable to return the actual clinical information that is most needed by the caregivers: through the WBP scale, it classifies four classes of wounds by the type of tissue: A-only granular tissue; B-50% slough; D-necrosis. This work aims to describe in depth the technology and the computational techniques that have been implemented, and to demonstrate reliability in automatically identifying, classifying through internationally accepted clinical scales and measuring such wounds, that peaked to over a 90% of accuracy.

Background: Chronic diabetic foot ulcers are a global health challenge, affecting approximately 18.6 million individuals each year. The timely and accurate prediction of wound healing paths is crucial for improving treatment outcomes and reducing complications. Methods: In this study, we apply predictive modeling to the case study of diabetic foot ulcers, analyzing and comparing multiple models based on Deep Neural Networks (DNNs) and Machine Learning (ML) algorithms to enhance wound prognosis and clinical decision making. Our approach leverages a dataset of 1766 diabetic foot wounds, each monitored for at least three visits, incorporating key clinical wound features such as WBP scores, wound area, depth, and tissue status. Results: Among the 12 models evaluated, the highest accuracy (80%) was achieved using a three-layer LSTM recurrent DNN trained on wound instances with four visits. The model performance was assessed through AUC (0.85), recall (0.80), precision (0.79), and F1-score (0.80). Our findings indicate that the wound depth and area at the first visit followed by the wound area and granulated tissue percentage at the second visit are the most influential factors in predicting the wound status. Conclusions: As future developments, we started building a weakly supervised semantic segmentation model that classifies wound tissues into necrosis, slough, and granulation, using tissue color proportions to further improve model performance. This research underscores the potential of predictive modeling in chronic wound management, specifically in the case of diabetic foot ulcers, offering a tool that can be seamlessly integrated into routine clinical practice.

The abilities of quantitative description of noise are restricted due to its origin, and only statistical and spectral analysis methods can be applied, while an exact time evolution cannot be defined or predicted. This emphasizes the challenges faced in many applications, including communication systems, where noise can play, on the one hand, a vital role in impacting the signal-to-noise ratio, but possesses, on the other hand, unique properties such as an infinite entropy (infinite information capacity), an exponentially decaying correlation function, and so on. Despite the deterministic nature of chaotic systems, the predictability of chaotic signals is limited for a short time window, putting them close to random noise. In this article, we propose and experimentally verify an approach to achieve Gaussian-distributed chaotic signals by processing the outputs of chaotic systems. The mathematical criterion on which the main idea of this study is based on is the central limit theorem, which states that the sum of a large number of independent random variables with similar variances approaches a Gaussian distribution. This study involves more than 40 mostly three-dimensional continuous-time chaotic systems (Chua’s, Lorenz’s, Sprott’s, memristor-based, etc.), whose output signals are analyzed according to criteria that encompass the probability density functions of the chaotic signal itself, its envelope, and its phase and statistical and entropy-based metrics such as skewness, kurtosis, and entropy power. We found that two chaotic signals of Chua’s and Lorenz’s systems exhibited superior performance across the chosen metrics. Furthermore, our focus extended to determining the minimum number of independent chaotic signals necessary to yield a Gaussian-distributed combined signal. Thus, a statistical-characteristic-based algorithm, which includes a series of tests, was developed for a Gaussian-like signal assessment. Following the algorithm, the analytic and experimental results indicate that the sum of at least three non-Gaussian chaotic signals closely approximates a Gaussian distribution. This allows for the generation of reproducible Gaussian-distributed deterministic chaos by modeling simple chaotic systems.

The abilities of quantitative description of noise are restricted due to its origin and only the statistical and spectral analysis methods can be applied while an exact time evolution cannot be defined or predicted. It emphasizes the challenges faced in many applications including communication systems where noise can play, on one hand, a vital role impacting signal-to-noise ratio but possessing, on the other hand, unique properties such as infinite entropy (infinite information capacity), exponentially decaying correlation function and so on. Despite the deterministic nature of chaotic systems, the predictability of chaotic signals is limited for a short time window putting them close to random noise. In this article, we propose and experimentally verify an approach to achieve Gaussian-distributed chaotic signals by processing the outputs of chaotic systems. The mathematical criteria, on which the main idea of the study is based on, is the Central Limit Theorem which states that the sum of a large number of independent random variables with similar variances approaches a Gaussian distribution. The study involves more than 40 mostly three-dimensional continuous-time chaotic systems (Chua’s, Lorenz’s, Sprott’s, memristor-based, etc.) whose output signals are analyzed according to criteria that encompass the probability density functions of a chaotic signal itself and its envelope and phase, statistical and entropy-based metrics such as skewness, kurtosis, and entropy power. We found that two chaotic signals of Chua’s and Lorenz’s systems exhibited superior performance across the chosen metrics. Furthermore, our focus extended to determining the minimum number of independent chaotic signals necessary to yield a Gaussian-distributed combined signal. The analytic and experimental results indicate that the sum of at least three non-Gaussian chaotic signals closely approximates Gaussian distribution. It allows generation of reproducible Gaussian-distributed deterministic chaos by modeling simple chaotic systems.

Chaos-based wireless communication systems can enhance the physical-layer security of IoT devices, but their reliability depends on stable chaotic behavior under real conditions. We investigate a modified Colpitts oscillator with a tunable base bias voltage, introduced as an independent control parameter to flexibly adjust nonlinear regimes. Using numerical studies, SPICE simulations, and hardware experiments, we show that simplified numerical models predict only a DC offset shift, whereas realistic implementations reveal qualitative changes in the dynamics, highlighting the need for experimental validation. We further demonstrate hybrid synchronization between the analog oscillator and an FPGA-based digital model. Despite model simplifications and non-idealities, synchronization is successfully achieved using the Pecora–Carroll method, showing that preserving the core dynamic structure is more critical than exact waveform replication. These results clarify the constraints of idealized models for predicting dynamical patterns while confirming the robustness of hybrid synchronization for secure, resource-constrained communication systems.

Chaos-based wireless communication systems offer promising opportunities for enhancing the physical-layer security of IoT devices. However, their successful integration depends on the ability to maintain robust chaotic behavior under environmental fluctuations, component tolerances, and parameter variations. This study investigates the nonlinear dynamics and hybrid synchronization of a modified Colpitts chaotic oscillator with a tunable base bias voltage. A comprehensive analysis is conducted through numerical simulations, SPICE-level modelling, and experimental validation using an analog prototype. While the simplified mathematical model shows only a DC offset shift with the added bias voltage, the SPICE simulations and hardware measurements reveal qualitative changes in the oscillator's dynamics. These findings underscore the importance of experimental verification, as idealized models often fail to capture critical behaviors in nonlinear systems. Hybrid synchronization is demonstrated between the analog oscillator and its FPGA-based digital counterpart, implemented using fixed-point arithmetic and the Euler–Cromer method. Despite model simplifications, effective synchronization is achieved via the Pecora–Carroll technique, highlighting that preserving the oscillator’s core dynamic structure is more critical than exact waveform replication. This result supports the practical use of chaos-based synchronization in resource-constrained secure communication systems.