A Potential solution for solving forensic is the use of blockchain in software-defined networking (SDN). The blockchain is a distributed peer-to-peer network that can be utilized on SDN-based Internet of Things (IoT) environments for security provisioning. Hence, to meet some challenges in digital forensics such as data integrity, evidence deletion or alteration, blockchain is used. However, some problems such as poor attack detection and slow processing existed in previous works. To address these issues, an efficient forensics architecture is proposed in SDN-IoT that establishes the Chain of Custody (CoC) in blockchain technology. The proposed SDN-based IoT architecture is initiated with flow table rules on switches for the three different traffics Voice over Internet Protocol (VoIP), File Transfer Protocol (FTP), and Hyper Text Transfer Protocol (HTTP). In this work, overloaded switches migrate the packets to nearby switches to balance the packet flow. The packets disobeying flow rules will be discarded by switches. The blockchain-based distributed controller in this forensic architecture is designed to use the Linear Homomorphic Signature (LHS) algorithm for validating users. Each controller is fed with a classifier that uses the Neuro Multi-fuzzy to classify malicious packets based on packet features. The logs of events are used and stored on the blockchain in the proposed SDN-IoT architecture. We evaluated the performance of our forensic architecture and compared it to the existing model using various performance measures. Our evaluation results demonstrate performance improvement by reducing delay, response time and processing time, increasing throughput, accuracy, and security parameters.
Worldwide, cardiovascular diseases are some of the primary causes of death; yet the early detection and diagnosis of such diseases have the potential to save many lives. Technological means of detection are becoming increasingly essential and numerous techniques have been created for this purpose, such as forecasting. Of these techniques, the time series forecasting technique seeks to predict future events. The long-term time series forecasting of physiological data could assist medical professionals in predicting and treating patients based on very early diagnosis. This article presents a model that utilizes a deep learning technique to predict long-term ECG signals. The forecasting model can learn signals’ nonlinearity, nonstationarity, and complexity based on a long short-term memory architecture. However, this is not a trivial task as the correct forecasting of a signal that closely resembles the original complex signal’s structure and behavior while minimizing any differences in amplitude continues to pose challenges. To achieve this goal, we used a dataset available on the Physio net database, called MIT-BIH, with 48 ECG recordings of 30 min each. The developed model starts with pre-processing to reduce interference in the original signals, then applies a deep learning algorithm, based on a long short-term memory (LTSM) neural network with two hidden layers. Next, we applied the root mean square error (RMSE) and mean absolute error (MAE) metrics to evaluate the performance of the model and obtained an average RMSE of and an average MAE of across all simulations. The results indicate that the proposed LSTM model is a promising technique for ECG forecasting, considering the trends of the changes in the original data series, most notably in R-peak amplitude. Given the model’s accuracy and the features of the physiological signals, the system could be used to improve existing predictive healthcare systems for cardiovascular monitoring.
17 March 2025 in Artificial Intelligence in Medicine
Parkinson's disease is a neurodegenerative disease. It is often diagnosed at an advanced stage, which can influence the control over the illness. Therefore, the possibility of diagnosing Parkinson's disease at an earlier stage, and possibly prognosticate it, could be an advantage. Given this, a literature review that covers current studies in the field is relevant. The aim of this study is to present a systematic literature review in which the models used for the diagnosis and prognosis of Parkinson's disease through voice and speech assessment are elucidated. Three databases were consulted to obtain the studies between 2019 and 2023: SienceDirect, IEEE Xplore and ACM Library . One hundred and six studies were considered eligible, considering the definition of inclusion and exclusion criteria. The vast majority of these studies (94.34%) focus on diagnosing the disease, while the remainder (11.32%) focus on prognosis. Voice analysis for the diagnosis and prognosis of Parkinson's disease using machine learning techniques can be achieved, with very satisfactory performance results, like is demonstrated in this systematic literature review.
10 July 2023 in IEEE Transactions on Geoscience and Remote Sensing
Optimized Deep Learning (DL)-based workflows can improve the efficiency and accuracy of earthquake detection and location processes. This paper introduces a six-step automated event detection, phase association, and earthquake location workflow, which integrates the state-of-the-art Pair-Input DL model and waveform Migration Location methods (IPIML). Applying IPIML on an 18-months dataset of Ghana Digital Seismic Network (GHSDN) recorded from 2012-2014, a catalog with 461 events is automatically obtained. Compared to other DL catalogs obtained using EQTransformer (EQT) and Siamese EQTransformer (S-EQT), the seismic event clusters in the IPIML catalog focus more on tectonically active regions or known seismogenic source areas and show a consistent depth distribution. The compiled catalog is 6.3 times larger than the reported catalog obtained by applying EQT with the default settings, indicating the importance of optimization and hyper-parameter tuning when applying DL models. As a result, a previously unknown seismogenic fault with a clear spatial trend has been identified using the new IPIML catalog, which provides more insights into the fault activities and seismic hazards in the region. The IPIML codes and data sets are available at the GitHub repository https://github.com/SigProSeismology/IPIML.git, contributing to the geoscience community.
Epilepsy is a neurological disease that causes uncontrollable seizures that can lead to severe or even lethal damage to the patient. This paper proposes an approach to predict epileptic seizures using peripheral electroencephalogram (EEG) channels from the CHB-MIT dataset. We created a machine learning algorithm to classify between interictal and preictal stages of seizures. The main goal is to assess the possibility of predicting these events using only peripheral channels and to present results for different configurations, such as the number of channels and their combinations. The preliminary performance of the algorithms is promising, with results similar to those in the literature that rely on channel reduction.