An Adaptable and Unsupervised TinyML Anomaly Detection System for Extreme Industrial Environments.

Industrial assets often feature multiple sensing devices to keep track of their status by monitoring certain physical parameters. These readings can be analyzed with machine learning (ML) tools to identify potential failures through anomaly detection, allowing operators to take appropriate corrective actions. Typically, these analyses are conducted on servers located in data centers or the cloud. However, this approach increases system complexity and is susceptible to failure in cases where connectivity is unavailable. Furthermore, this communication restriction limits the approach's applicability in extreme industrial environments where operating conditions affect communication and access to the system. This paper proposes and evaluates an end-to-end adaptable and configurable anomaly detection system that uses the Internet of Things (IoT), edge computing, and Tiny-MLOps methodologies in an extreme industrial environment such as submersible pumps. The system runs on an IoT sensing Kit, based on an ESP32 microcontroller and MicroPython firmware, located near the data source. The processing pipeline on the sensing device collects data, trains an anomaly detection model, and alerts an external gateway in the event of an anomaly. The anomaly detection model uses the isolation forest algorithm, which can be trained on the microcontroller in just 1.2 to 6.4 s and detect an anomaly in less than 16 milliseconds with an ensemble of 50 trees and 80 KB of RAM. Additionally, the system employs blockchain technology to provide a transparent and irrefutable repository of anomalies.

Design and Implementation of an Energy-Efficient Weather Station for Wind Data Collection.

Agriculture faces critical challenges caused by changing climatic factors and weather patterns with random distribution. This has increased the need for accurate local weather predictions and weather data collection to support precision agriculture. The demand for uninterrupted weather stations is overwhelming, and the Internet of Things (IoT) has the potential to address this demand. One major challenge of energy constraint in remotely deployed IoT devices can be resolved using weather stations that are energy neutral. This paper focuses on optimizing the energy consumption of a weather station by optimizing the data collected and sent from the sensor deployed in remote locations. An asynchronous optimization algorithm for wind data collection has been successfully developed, using the development lifecyle specifically designed for weather stations and focused on achieving energy neutrality. The developed IoT weather station was deployed in the field, and it has the potential to reduce the power consumption of the weather station by more than 60%.

A Decentralized Peer-to-Peer Remote Health Monitoring System.

Within the Internet of Things (IoT) and blockchain research, there is a growing interest in decentralizing health monitoring systems, to provide improved privacy to patients, without relying on trusted third parties for handling patients' sensitive health data. With public blockchain deployments being severely limited in their scalability, and inherently having latency in transaction processing, there is room for researching and developing new techniques to leverage the security features of blockchains within healthcare applications. This paper presents a solution for patients to share their biomedical data with their doctors without their data being handled by trusted third party entities. The solution is built on the Ethereum blockchain as a medium for negotiating and record-keeping, along with Tor for delivering data from patients to doctors. To highlight the applicability of the solution in various health monitoring scenarios, we have considered three use-cases, namely cardiac monitoring, sleep apnoea testing, and EEG following epileptic seizures. Following the discussion about the use cases, the paper outlines a security analysis performed on the proposed solution, based on multiple attack scenarios. Finally, the paper presents and discusses a performance evaluation in terms of data delivery time in comparison to existing centralized and decentralized solutions.

Therapeutic approach to bronchiolitis: why pediatricians continue to overprescribe drugs?

BACKGROUND: Bronchiolitis guidelines suggest that neither bronchodilators nor corticosteroids, antiviral and antibacterial agents should be routinely used. Although recommendations, many clinicians persistently prescribe drugs for bronchiolitis. AIM OF THE STUDY: To unravel main reasons of pediatricians in prescribing drugs to infants with bronchiolitis, and to possibly correlate therapeutic choices to the severity of clinical presentation. Also possible influence of socially deprived condition on therapeutic choices is analyzed. METHODS: Patients admitted to Pediatric Division of 2 main Hospitals of Naples because of bronchiolitis in winter season 2008-2009 were prospectively analyzed. An RDAI (Respiratory Distress Assessment Instrument) score was assessed at different times from admission. Enrollment criteria were: age 1-12 months; 1st lower respiratory infection with cough and rhinitis with/without fever, wheezing, crackles, tachypnea, use of accessory muscles, and/or nasal flaring, low oxygen saturation, cyanosis. Social deprivation status was assessed by evaluating school graduation level of the origin area of the patients. A specific questionnaire was submitted to clinicians to unravel reasons of their therapeutic behavior. RESULTS: Eighty-four children were enrolled in the study. Mean age was 3.5 months. Forty-four per cent of patients presented with increased respiratory rate, 70.2% with chest retractions, and 7.1% with low SaO2. Mean starting RDAI score was 8. Lung consolidation was found in 3.5% on chest roentgenogram. Data analysis also unraveled that 64.2% matched clinical admission criteria. Social deprivation status analysis revealed that 72.6% of patients were from areas "at social risk". Evaluation of length of stay vs. social deprivation status evidenced no difference between "at social risk" and "not at social risk" patients. Following therapeutic interventions were prescribed: nasal suction (64.2%), oxygen administration (7.1%), antibiotics (50%), corticosteroids (85.7%), bronchodilators (91.6%). Statistically significant association was not found for any used drug with neither RDAI score nor social deprivation status. The reasons of hospital pediatricians to prescribe drugs were mainly the perception of clinical severity of the disease, the clinical findings at chest examination, and the detection of some improvement after drug administration. CONCLUSIONS: We strongly confirm the large use of drugs in bronchiolitis management by hospital pediatricians. Main reason of this wrong practice appears to be the fact that pediatricians recognize bronchiolitis as a severe condition, with consequent anxiety in curing so acutely ill children without drugs, and that sometimes they feel forced to prescribe drugs because of personal reassurance or parental pressure. We also found that social "at risk" condition represents a main reason for hospitalization, not correlated to clinical severity of the disease neither to drug prescription. Eventually, we suggest a "step-by-step" strategy to rich a more evidence based approach to bronchiolitis therapy, by adopting specific and shared resident guidelines.

Appropriateness of hospitalization for CAP-affected pediatric patients: report from a Southern Italy General Hospital.

BACKGROUND: Community-acquired pneumonia (CAP) is a common disease, responsible for significant healthcare expenditures, mostly because of hospitalization. Many practice guidelines on CAP have been developed, including admission criteria, but a few on appropriate hospitalization in children. The aim of this study was to evaluate appropriate hospital admission for CAP in a pediatric population. METHODS: We evaluated appropriate admission to a Pediatric Unit performing a retrospective analysis on CAP admitted pediatric patients from a Southern Italy area. Diagnosis was made based on clinical and radiological signs. Appropriate hospital admission was evaluated following clinical and non-clinical international criteria. Family ability to care children was assessed by evaluating social deprivation status. RESULTS: In 2 winter seasons 120 pediatric patients aged 1-129 months were admitted because of CAP. Median age was 28.7 months. Raised body temperature was scored in 68.3% of patients, cough was present in 100% of cases, and abdominal pain was rarely evidenced. Inflammatory indices (ESR and CRP) were found elevated in 33.3% of cases. Anti-Mycoplasma pneumoniae antibodies were found positive in 20.4%. Trans-cutaneous (TC) SaO2 was found lower than 92% in 14.6%. Dyspnoea was present in 43.3%. Dehydration requiring i.v. fluid supplementation was scored in 13.3%. Evaluation of familial ability to care their children revealed that 76% of families (derived from socially depressed areas) were "at social risk", thus not able to appropriately care their children. Furthermore, analysis of CAP patients revealed that "at social risk" people accessed E.D. and were hospitalized more frequently than "not at risk" patients (odds ratio = 3.59, 95% CI: 1,15 to 11,12; p = 0.01), and that admitted "at social risk" people presented without clinical signs of severity (namely dyspnoea, and/or SaO2

Prognostic stratification of acute pulmonary embolism: focus on clinical aspects, imaging, and biomarkers.

Pulmonary embolism (PE) represents a common disease in emergency medicine and guidelines for diagnosis and treatment have had wide diffusion. However, PE morbidity and mortality remain high, especially when associated to hemodynamic instability or right ventricular dysfunction. Prognostic stratification to identify high risk patients needing to receive more aggressive pharmacological and closer monitoring is of utmost importance. Modern guidelines for management of acute PE are based on risk stratification using either clinical, radiological, or laboratory findings. This article reviews the modern treatment of acute PE, which is customized upon patient prognosis. Accordingly the current risk stratification tools described in the literature such as clinical scores, echocardiography, helical computer tomography, and biomarkers will be reviewed.

The Risk-based Treatment of Acute Pulmonary Embolism.

UNLABELLED: Risk evaluation and prognostic stratification based upon clinical and radiological findings and new cardiac biomarkers, such as natriuretic peptides (NP) and troponins, represent key points in modern management of acute pulmonary embolism (PE). Literature evidence shows that normotensive PE with right heart dysfunction (RHD), defined as submassive PE, has poorer prognosis when compared to normotensive PE without RHD, defined as non-massive PE; thus whether submassive PE should be managed more aggressively and with closer monitoring represents the crucial question about acute PE treatment. Although the answer is yet unclear, the most recent guidelines address to thrombolysis as treatment choice in selected high risk patients with submassive PE. Guidelines also clarify the indications for unfractioned and low molecular weight heparins and fondaparinux. Therefore, in the present article, the authors focus on modern risk-based therapeutic guidelines of acute PE. KEYWORDS: Pulmonary embolism; Treatment; Prognosis; Biomarkers; Chocardiography; Hemodynamic; Guidelines.

Pulmonary embolism in the elderly: a review on clinical, instrumental and laboratory presentation.

OBJECTIVE: Diagnosis of pulmonary embolism (PE) remains difficult and is often missed in the elderly due to nonspecific and atypical presentation. Diagnostic algorithms able to rule out PE and validated in young adult patients may have reduced applicability in elderly patients, which increases the number of diagnostic tools use and costs. The aim of the present study was to analyze the reported clinical presentation of PE in patients aged 65 and more. MATERIALS AND METHODS: Prospective and retrospective English language studies dealing with the clinical, instrumental and laboratory aspects of PE in patients more than 65 and published after January 1987 and indexed in MEDLINE using keywords as pulmonary embolism, elderly, old, venous thromboembolism (VTE) in the title, abstract or text, were reviewed. RESULTS: Dyspnea (range 59%-91.5%), tachypnea (46%-74%), tachycardia (29%-76%), and chest pain (26%-57%) represented the most common clinical symptoms and signs. Bed rest was the most frequent risk factor for VTE (15%-67%); deep vein thrombosis was detected in 15%-50% of cases. Sinus tachycardia, right bundle branch block, and ST-T abnormalities were the most frequent ECG findings. Abnormalities of chest X-ray varied (less than 50% in one-half of the studies and more than 70% in the other one-half). Arterial blood gas analysis revealed severe hypoxemia and mild hypocapnia as the main findings. D-Dimer was higher than cut-off in 100% of patients in 75% of studies. Clinical usefulness of D-Dimer measurement decreases with age, although the strategies based on D-Dimer seem to be cost-effective at least until 80 years. CONCLUSION: Despite limitations due to pooling data of heterogeneous studies, our review could contribute to the knowledge of the presentation of PE in the elderly with its diagnostic difficulties. A diagnostic strategy based on reviewed data is proposed.