Smart Buildings: Water Leakage Detection Using TinyML.

The escalating global water usage and the increasing strain on major cities due to water shortages highlights the critical need for efficient water management practices. In water-stressed regions worldwide, significant water wastage is primarily attributed to leakages, inefficient use, and aging infrastructure. Undetected water leakages in buildings' pipelines contribute to the water waste problem. To address this issue, an effective water leak detection method is required. In this paper, we explore the application of edge computing in smart buildings to enhance water management. By integrating sensors and embedded Machine Learning models, known as TinyML, smart water management systems can collect real-time data, analyze it, and make accurate decisions for efficient water utilization. The transition to TinyML enables faster and more cost-effective local decision-making, reducing the dependence on centralized entities. In this work, we propose a solution that can be adapted for effective leakage detection in real-world scenarios with minimum human intervention using TinyML. We follow an approach that is similar to a typical machine learning lifecycle in production, spanning stages including data collection, training, hyperparameter tuning, offline evaluation and model optimization for on-device resource efficiency before deployment. In this work, we considered an existing water leakage acoustic dataset for polyvinyl chloride pipelines. To prepare the acoustic data for analysis, we performed preprocessing to transform it into scalograms. We devised a water leak detection method by applying transfer learning to five distinct Convolutional Neural Network (CNN) variants, which are namely EfficientNet, ResNet, AlexNet, MobileNet V1, and MobileNet V2. The CNN models were found to be able to detect leakages where a maximum testing accuracy, recall, precision, and F1 score of 97.45%, 98.57%, 96.70%, and 97.63%, respectively, were observed using the EfficientNet model. To enable seamless deployment on the Arduino Nano 33 BLE edge device, the EfficientNet model is compressed using quantization resulting in a low inference time of 1932 ms, a peak RAM usage of 255.3 kilobytes, and a flash usage requirement of merely 48.7 kilobytes.

On-Device IoT-Based Predictive Maintenance Analytics Model: Comparing TinyLSTM and TinyModel from Edge Impulse.

A precise prediction of the health status of industrial equipment is of significant importance to determine its reliability and lifespan. This prediction provides users information that is useful in determining when to service, repair, or replace the unhealthy equipment's components. In the last decades, many works have been conducted on data-driven prognostic models to estimate the asset's remaining useful life. These models require updates on the novel happenings from regular diagnostics, otherwise, failure may happen before the estimated time due to different facts that may oblige rapid maintenance actions, including unexpected replacement. Adding to offline prognostic models, the continuous monitoring and prediction of remaining useful life can prevent failures, increase the useful lifespan through on-time maintenance actions, and reduce the unnecessary preventive maintenance and associated costs. This paper presents the ability of the two real-time tiny predictive analytics models: tiny long short-term memory (TinyLSTM) and sequential dense neural network (DNN). The model (TinyModel) from Edge Impulse is used to predict the remaining useful life of the equipment by considering the status of its different components. The equipment degradation insights were assessed through the real-time data gathered from operating equipment. To label our dataset, fuzzy logic based on the maintainer's expertise is used to generate maintenance priorities, which are later used to compute the actual remaining useful life. The predictive analytic models were developed and performed well, with an evaluation loss of 0.01 and 0.11, respectively, for the LSTM and model from Edge Impulse. Both models were converted into TinyModels for on-device deployment. Unseen data were used to simulate the deployment of both TinyModels. Conferring to the evaluation and deployment results, both TinyLSTM and TinyModel from Edge Impulse are powerful in real-time predictive maintenance, but the model from Edge Impulse is much easier in terms of development, conversion to Tiny version, and deployment.

Mobility-Aware Proactive Edge Caching Optimization Scheme in Information-Centric IoV Networks.

Edge caching is a promising approach to alleviate the burden on the backhaul of network links. It has a significant role in the Internet of Vehicle (IoV) networks performance by providing cached data at the edge and reduce the burden of the core network caused by the number of participating vehicles and data volume. However, due to the limited computing and storage capabilities of edge devices, it is hard to guarantee that all contents are cached and every requirement of the device are satisfied for all users. In this paper, we design an Information-Centric Network (ICN) with mobility-aware proactive caching scheme to provide delay-sensitive services on IoV networks. The real-time status and interaction of vehicles with other vehicles and Roadside Units (RSU) is modeled using a Markov process. Mobility aware proactive edge caching decision that maximize network performance while minimizing transmission delay is applied. Our numerical simulation results show that the proposed scheme outperforms related caching schemes in terms of latency by 20-25% in terms of latency and by 15-23% in cache hits.

A Data-Driven Predictive Machine Learning Model for Efficiently Storing Temperature-Sensitive Medical Products, Such as Vaccines: Case Study: Pharmacies in Rwanda.

Temperature control is the key element during medicine storage. Pharmacies sell some medical products which are kept in fridges. The opening and closing of the fridge while taking some medicine makes the outside hot air enter the fridge, which will increase the inner fridge temperature. When the frequency of opening and closing of the fridge is increased, the temperature may go beyond the allowed storage temperature range. In this paper, we are proposing a model with the help of machine learning that will be used in multiple chambers fridges to keep indicating the time remaining for the inner temperature to go beyond the allowed range, and if the time is short, the system will propose to the pharmacist not to open that particular room and proposes a room that has enough time slots (time to reach the upper limit temperature). By using training data got from a thermoelectric cooler-based fridge, we constructed a multiple linear regression model that can predict the required time for a given room to reach the cut-off temperature in case that room is opened. The built model was evaluated using the coefficient of determination R 2 and is found to be 77%, and then it can be used to develop a multiple room smart fridge for efficiently storing highly sensitive medical products.

Rainfall-Induced Landslide Prediction Using Machine Learning Models: The Case of Ngororero District, Rwanda.

Landslides fall under natural, unpredictable and most distractive disasters. Hence, early warning systems of such disasters can alert people and save lives. Some of the recent early warning models make use of Internet of Things to monitor the environmental parameters to predict the disasters. Some other models use machine learning techniques (MLT) to analyse rainfall data along with some internal parameters to predict these hazards. The prediction capability of the existing models and systems are limited in terms of their accuracy. In this research paper, two prediction modelling approaches, namely random forest (RF) and logistic regression (LR), are proposed. These approaches use rainfall datasets as well as various other internal and external parameters for landslide prediction and hence improve the accuracy. Moreover, the prediction performance of these approaches is further improved using antecedent cumulative rainfall data. These models are evaluated using the receiver operating characteristics, area under the curve (ROC-AUC) and false negative rate (FNR) to measure the landslide cases that were not reported. When antecedent rainfall data is included in the prediction, both models (RF and LR) performed better with an AUC of 0.995 and 0.997, respectively. The results proved that there is a good correlation between antecedent precipitation and landslide occurrence rather than between one-day rainfall and landslide occurrence. In terms of incorrect predictions, RF and LR improved FNR to 10.58% and 5.77% respectively. It is also noted that among the various internal factors used for prediction, slope angle has the highest impact than other factors. Comparing both the models, LR model's performance is better in terms of FNR and it could be preferred for landslide prediction and early warning. LR model's incorrect prediction rate FNR = 9.61% without including antecedent precipitation data and 3.84% including antecedent precipitation data.

Characterization and Design Improvement of a Thickness-Shear Lead Zirconate Titanate Transducer for Low Frequency Ultrasonic Guided Wave Applications.

Thickness-shear transducers for guided wave testing have been used in industry for over two decades and much research has been conducted to improve the resolution and sensitivity. Due to a geometric feature of the current state-of-the art transducer, there is an out-of-plane component in the propagation direction of the fundamental shear horizontal mode which complicates the signal interpretation. In such case, complex signal processing techniques need to be used for mode discrimination to assess the structural health with higher precision. Therefore, it is important to revise the transducer design to eliminate the out-of-plane components in the propagation direction of fundamental shear horizontal mode. This will enhance the mode purity of fundamental shear horizontal mode for its application in guided wave inspection. A numerical investigation has been conducted on a 3 mm thick 2 m circular steel plate to understand the behaviour and the characteristics of the state-of-the-art thickness-shear transducer. Based on the results, it is noted that the redesigning the electrode arrangement will suppress the out-of-plane components on the propagation direction of the fundamental shear horizontal mode. With the aid of this information current state-of-the-art transducers were redesigned and tested in laboratory conditions using the 3D Laser Doppler Vibrometer. This information will aid future transducer designers improve the resolution of thickness-shear transducers for guided wave applications and reduce the weight and cost of transducer array by eliminating the need of additional transducers to suppress spurious modes.

Clinically apparent long-term electric disturbances in the acute and very long-term of patent foramen ovale device-based closure.

BACKGROUND/PURPOSE: Incidence of electrical disturbances in patients submitted to transcatheter patent foramen ovale (PFO) closure has not been fully clarified in a large population. The aim of the study is to assess the incidence of atrial fibrillation, supraventricular tachi-arrhythmias, and atrio-ventricular block in the acute and very long-term follow-up. METHODS/MATERIALS: We reviewed the medical and instrumental data of 1000 consecutive patients (mean age 47.3±17.1years) prospectively enrolled in two centers over a 13-year period (February 1999 to February 2012) for right-to-left (R-to-L) shunt ICE-aided catheter-based closure using different devices. RESULTS: Successful transcatheter PFO closure was achieved in 99.8% of the patients. Implanted devices were: Amplatzer PFO Occluder in 463 patients (46.3%), Amplatzer ASD Cribriform Occluder in 420 patients (42.0%), Premere Occlusion System in 95 patients (9.5%), and Biostar Occluder in 22 patients (2.2%). Postprocedural electrical complications occurred in 5.9% of patients. The only independent predictors of electrophysiological complications were female gender (OR 2.3, 0.5-5.1 [95% CI], p<0.001) and device disk >30mm (OR 5.0, 1.2-7.2 [95% CI], p<0.001). On a mean follow-up of 12 .3±0.6years (minimum 4- maximum 17years), electrical complications occurred in 1.4% of patients including one only case of complete AVB and 5 cases of permanent AF. The only independent predictors were female gender (OR 2.3, 0.5-5.1 [95% CI], p<0.001) and device disk >30mm (OR 5.0, 1.2-7.2 [95% CI], p<0.001). CONCLUSION: Device-based closure of PFO using different devices, appeared very safe from an electrophysiological point of view with low incidence of electrical disturbances even in the very long-term follow-up.

Long-Term Outcomes and Complications of Intracardiac Echocardiography-Assisted Patent Foramen Ovale Closure in 1,000 Consecutive Patients.

BACKGROUNDS: Long-term fate of patients submitted to patent foramen ovale (PFO) closure is still unclear. The aim of the study was to evaluate the incidence of atrial fibrillation (AF), aortic or atrial free wall erosion, device thrombosis (DT), new onset or worsening of mitral valve regurgitation (MVR), and recurrent cerebral ischemic events in the long-term follow up after intracardiac echocardiography (ICE)-aided PFO closure in a large population. METHODS: We reviewed the medical and instrumental data of 1,000 consecutive patients (mean age 47.3 ± 17.1 years, females) prospectively enrolled in 2 centres over a 13 years period (February 1999-February 2012) for R-to-L shunt ICE-aided catheter-based closure using different devices. RESULTS: Immediate success was 99.8%. Implanted devices were Amplatzer PFO Occluder in 463 patients (46.3%), Amplatzer ASD Cribriform Occluder in 420 patients (42.0%), Premere Occlusion System in 95 patients (9.5%), and Biostar Occluder in 22 patients (2.2%). On a mean follow-up of 12.3 ± 0.6 years (minimum 4- maximum 17 years), permanent AF occurred in 0.5%, DT was apparent in 0.5%, new onset or worsening of MVR was observed in 0.2%, whereas recurrent cerebral ischemic events were 0.8%. Occlusion rate was 93.8%. No aortic or atrial free wall erosion has been observed. CONCLUSION: ICE-aided closure of PFO using different devices, appeared very safe and effective on very long-term follow up with low incidence of erosion, DT, recurrent ischemic events, MVR new onset or worsening, and permanent AF.

On the Design of Smart Parking Networks in the Smart Cities: An Optimal Sensor Placement Model.

Smart parking is a typical IoT application that can benefit from advances in sensor, actuator and RFID technologies to provide many services to its users and parking owners of a smart city. This paper considers a smart parking infrastructure where sensors are laid down on the parking spots to detect car presence and RFID readers are embedded into parking gates to identify cars and help in the billing of the smart parking. Both types of devices are endowed with wired and wireless communication capabilities for reporting to a gateway where the situation recognition is performed. The sensor devices are tasked to play one of the three roles: (1) slave sensor nodes located on the parking spot to detect car presence/absence; (2) master nodes located at one of the edges of a parking lot to detect presence and collect the sensor readings from the slave nodes; and (3) repeater sensor nodes, also called "anchor" nodes, located strategically at specific locations in the parking lot to increase the coverage and connectivity of the wireless sensor network. While slave and master nodes are placed based on geographic constraints, the optimal placement of the relay/anchor sensor nodes in smart parking is an important parameter upon which the cost and efficiency of the parking system depends. We formulate the optimal placement of sensors in smart parking as an integer linear programming multi-objective problem optimizing the sensor network engineering efficiency in terms of coverage and lifetime maximization, as well as its economic gain in terms of the number of sensors deployed for a specific coverage and lifetime. We propose an exact solution to the node placement problem using single-step and two-step solutions implemented in the Mosel language based on the Xpress-MPsuite of libraries. Experimental results reveal the relative efficiency of the single-step compared to the two-step model on different performance parameters. These results are consolidated by simulation results, which reveal that our solution outperforms a random placement in terms of both energy consumption, delay and throughput achieved by a smart parking network.

Ubiquitous Sensor Networking for Development (USN4D): an application to pollution monitoring.

This paper presents a new Ubiquitous Sensor Network (USN) Architecture to be used in developing countries and reveals its usefulness by highlighting some of its key features. In complement to a previous ITU proposal, our architecture referred to as "Ubiquitous Sensor Network for Development (USN4D)" integrates in its layers features such as opportunistic data dissemination, long distance deployment and localisation of information to meet the requirements of the developing world. Besides describing some of the most important requirements for the sensor equipment to be used in a USN4D setting, we present the main features and experiments conducted using the "WaspNet" as one of the wireless sensor deployment platforms that meets these requirements. Furthermore, building upon "WaspNet" platform, we present an application to Air pollution Monitoring in the city of Cape Town, in South Africa as one of the first steps towards building community wireless sensor networks (CSN) in the developing world using off-the-shelf sensor equipment.

On the relevance of using Bayesian belief networks in wireless sensor networks situation recognition.

Achieving situation recognition in ubiquitous sensor networks (USNs) is an important issue that has been poorly addressed by both the research and practitioner communities. This paper describes some steps taken to address this issue by effecting USN middleware intelligence using an emerging situation awareness (ESA) technology. We propose a situation recognition framework where temporal probabilistic reasoning is used to derive and emerge situation awareness in ubiquitous sensor networks. Using data collected from an outdoor environment monitoring in the city of Cape Town, we illustrate the use of the ESA technology in terms of sensor system operating conditions and environmental situation recognition.

Use of the Amplatzer ASD Occluder for closing a persistent left vertical vein.

We report the case of a very large anomalous connection of the veins draining the upper lobe of the left lung to both the left-sided vertical vein and the left atrium, associated with mild rheumatic mitral valve stenosis, in which the atrial septum was intact and the remaining venous system, including the coronary sinus, was otherwise normal (a variant of Lutembacher's syndrome). In order to abolish the left-to-right shunting, a transcatheter approach to close this venous structure was successfully attempted using an Amplatzer ASD Occluder device. The technical aspects and the alternative options of performing a procedure with a device for a purpose outside the scope of its approved label are discussed.

On the relevance of using open wireless sensor networks in environment monitoring.

This paper revisits the problem of the readiness for field deployments of wireless sensor networks by assessing the relevance of using Open Hardware and Software motes for environment monitoring. We propose a new prototype wireless sensor network that fine-tunes SquidBee motes to improve the life-time and sensing performance of an environment monitoring system that measures temperature, humidity and luminosity. Building upon two outdoor sensing scenarios, we evaluate the performance of the newly proposed energy-aware prototype solution in terms of link quality when expressed by the Received Signal Strength, Packet Loss and the battery lifetime. The experimental results reveal the relevance of using the Open Hardware and Software motes when setting up outdoor wireless sensor networks.

Strut fracture: a further concern with drug-eluting stents.

A potential cause of drug-eluting stent restenosis is strut fracture. To date, few cases have been reported in the available literature and a best treatment method has not been established. We describe two cases of sirolimus-eluting stent fracture combined with focal restenosis and also review those previously reported.

Intracardiac echocardiography: gross anatomy and magnetic resonance correlations and validations.

PURPOSE: The feasibility and safety of intracardiac echocardiography (ICE) in humans, using low frequency transducers, and its excellent tissue contrast capabilities that enhances the differentiation of intracardiac structures have been previously demonstrated. However, correlations among ICE imaging and anatomic sections or magnetic resonance (MR) scan planes have never been described before. This study was designed to correlate a simplified ICE approach with the anatomy of the right atrium and great vessels obtained by serial post-mortem sections and cardiac MR images. METHODS: A stepwise approach to ICE, which is based on our experience on over 300 consecutive patients with interatrial communications, has been correlated with anatomic sections from pressure-perfused-fixed hearts and spin echo cardiac MR imaging. A 9F-9 MHz mechanical device was used to record four transverse and one longitudinal sections for an extensive evaluation of the intracardiac architecture. RESULTS: ICE transverse and longitudinal views allowed the detection of all the required information. Moreover, the anatomic sections and cardiac MR scan planes allowed validation of all the structures imaged by ICE. The potential clinical applications of the ICE technique are further discussed. CONCLUSION: This paper demonstrates the basic accuracy of this new imaging modality. Therefore, mechanical ICE might be considered an active investment in cardiac catheterization laboratories, specifically in the percutaneous interventional procedure setting, discovering a new route (let's see and treat) to sophisticated interventions. In this respect, there can be little doubt that the knowledge of cross-sectional mechanical ICE imaging provides the basis for understanding the heart anatomy.

Catheter closure of perforated secundum atrial septal defect under intracardiac echocardiographic guidance using a single amplatzer device: feasibility of a new method.

OBJECTIVES: Little is known about the feasibility and effectiveness of perforated secundum atrial septal defects (ASDs) transcatheter closure. This study tested whether this type of fenestration can be percutaneously closed using a single Amplatzer PFO or Cribriform Occluder device in a patch-like fashion under intracardiac echocardiographic (ICE) guidance and monitoring. METHODS: Twenty-four adult patients with perforated ASDs underwent transcatheter single Amplatzer PFO or Cribriform Occluder device closure by using ICE as the primary imaging tool, both for crossing the flap valve and monitoring each stage of the procedure. The entire atrial septum and fossa ovalis end diastolic dimensions on 2 orthogonal ICE planes, as well as the distance between the eccentric guidewire passage and the more adjacent rim of the fossa ovalis on the longitudinal ICE plane, have been measured. RESULTS: On the basis of the ICE fossa ovalis size and its geometric evaluation, in order to completely cover the fossa ovalis with a single device, 2 Amplatzer PFO Occluders 25 mm, 9 Amplatzer PFO Occluders 35 mm, and 13 Amplatzer Cribriform Occluders (four 25 mm and nine 35 mm) have been implanted successfully. During follow-up (31.4, 7.2 months), complete closure by contrast echo color Doppler occurred in 16/24 (67%) patients after 24 hours, 19/24 (79%) after 1 mouth, 20/24 (83%) after 3 months, 22/24 (92%) after 1 year, and 23/24 (96%) after 2 years. CONCLUSIONS: Transcatheter closure of perforated ASDs using a single Amplatzer Occluder device under ICE guidance appears to be a suitable procedure, providing a low rate of residual shunting.

Acute diastolic dysfunction due to pneumomediastinum following positive end-expiratory pressure--a case report.

A 78-year-old woman was admitted for pulmonary embolism requiring orotracheal intubation and positive end-expiratory pressure. The pulmonary angiography confirmed a massive pulmonary embolism as suggested by echocardiography. Heparin and recombinant tissue plasminogen activator were successfully administrated; nevertheless, cardiogenic shock developed. A diastolic morphology of the right-left cardiac pressures, despite a normalization of lung vasculature, was discovered by a repeated cardiac catheterization and pulmonary angiography. No tamponade was detected by echocardiography. Computed tomography demonstrated a large pneumomediastinum caused by positive end-expiration pressure, as a cause of the acute diastolic dysfunction. The patient died of a cardiac arrest after an unsuccessful drainage attempt.

Pseudo-lesion of internal mammary artery graft and left anterior descending artery during percutaneous transluminal angioplasty--a case report.

New lesions appearing during coronary angioplasty may be due to spasms, dissection, and thrombosis. Straightening of the tortuous vessels by guidewire may produce transient angiographic pseudo-lesions, which mimic severe artery damage. An additional case is reported, in which simultaneous artifactual lesions involved the internal mammary artery and the left anterior descending coronary artery, mimicking thrombosis and dissection. Recognition of this entity is essential to avoid unnecessary interventions and potentially harmful complications.

Transcatheter atrial septal defect closure assisted by intracardiac echocardiography: 3-year follow-up.

Intracardiac echocardiographic (ICE) evaluation of secundum atrial septal defect (ASDs) shows several advantages over transesophageal echocardiography (TEE). The aim of our study is to describe the 3-year results of transcatheter ASDs closure using the Amplatzer septal occluder (ASO) and ICE as the sole imaging tool both to select the device size and to monitor the procedure. Under local anesthesia, 135 consecutive eligible patients with ASDs (male/female = 45/90; mean age 42.2 +/- 19.2 years; BSA 1.71 +/- 0.7 m(2); mean pulmonary arterial pressure 30.4 +/- 11.5 mmHg; mean Qp/Qs ratio 2.1 +/- 0.6) underwent transcatheter closure using ASO and ICE. A 9F-9MHz mechanical transducer was used. Two orthogonal views on the transverse aortic valve and on the longitudinal four-chamber planes were obtained for quantitative ICE measurements, from which the diameters of the implanting ASO waist were to be derived. In all cases, we obtained a proper evaluation of ICE ASDs dimensions, leading to an optimal device size selection (mean size 25.0 +/- 6.7 mm). Moreover, the ICE allowed us to monitor device deployment and to verify the effectiveness of the ASDs stented by ASO. There were no complications related to the procedure or to the use of ICE. During a mean follow-up period of 21.5 +/- 12 months, the cumulative complete occlusion rates were 97.7, 97.0, 97.1, and 98% at 24 hours, 3 month, and 1 and 3 years, respectively. ICE is an effective and safe alternative to TEE and balloon-sizing maneuver during ASDs transcatheter closure procedures, allowing us to avoid the need of general anesthesia and leading to a similarly high percentage of occlusion rates with respect to the conventional method.