Power Efficient Machine Learning Models Deployment on Edge IoT Devices.

Computing has undergone a significant transformation over the past two decades, shifting from a machine-based approach to a human-centric, virtually invisible service known as ubiquitous or pervasive computing. This change has been achieved by incorporating small embedded devices into a larger computational system, connected through networking and referred to as edge devices. When these devices are also connected to the Internet, they are generally named Internet-of-Thing (IoT) devices. Developing Machine Learning (ML) algorithms on these types of devices allows them to provide Artificial Intelligence (AI) inference functions such as computer vision, pattern recognition, etc. However, this capability is severely limited by the device's resource scarcity. Embedded devices have limited computational and power resources available while they must maintain a high degree of autonomy. While there are several published studies that address the computational weakness of these small systems-mostly through optimization and compression of neural networks- they often neglect the power consumption and efficiency implications of these techniques. This study presents power efficiency experimental results from the application of well-known and proven optimization methods using a set of well-known ML models. The results are presented in a meaningful manner considering the "real world" functionality of devices and the provided results are compared with the basic "idle" power consumption of each of the selected systems. Two different systems with completely different architectures and capabilities were used providing us with results that led to interesting conclusions related to the power efficiency of each architecture.

Venous Thromboembolic Disease in Chronic Inflammatory Lung Diseases: Knowns and Unknowns.

Persistent inflammation within the respiratory tract underlies the pathogenesis of numerous chronic pulmonary diseases. There is evidence supporting that chronic lung diseases are associated with a higher risk of venous thromboembolism (VTE). However, the relationship between lung diseases and/or lung function with VTE is unclear. Understanding the role of chronic lung inflammation as a predisposing factor for VTE may help determine the optimal management and aid in the development of future preventative strategies. We aimed to provide an overview of the relationship between the most common chronic inflammatory lung diseases and VTE. Asthma, chronic obstructive pulmonary disease, interstitial lung diseases, or tuberculosis increase the VTE risk, especially pulmonary embolism (PE), compared to the general population. However, high suspicion is needed to diagnose a thrombotic event early as the clinical presentation inevitably overlaps with respiratory disorders. PE risk increases with disease severity and exacerbations. Hence, hospitalized patients should be considered for thromboprophylaxis administration. Conversely, all VTE patients should be asked for lung comorbidities before determining anticoagulant therapy duration, as those patients are at increased risk of recurrent PE episodes rather than DVT. Further research is needed to understand the underlying pathophysiology of in-situ thrombosis in those patients.

Inappropriate Use of Public Hospitals Emergency Departments in Greece: Magnitude and Associated Factors.

The aim of this study was to evaluate the appropriateness of use of the Emergency Departments (EDs) and to identify the reasons for inappropriate use. A study with 805 patients visiting the EDs of four large-scale public hospitals in Athens was conducted using the Hospital Urgencies Appropriateness Protocol (HUAP). 38.1% of the visits (n=307) were estimated as inappropriate, due to several reasons such as increased confidence in hospital rather than primary care services/patients' expectation for improved care in EDs (46.6%), convenience/proximity to patient's residence (44.6%) etc. Ageing, Greek nationality and insurance coverage were related with the appropriate use of EDs (p<0.001, p=0.04 and p=0.005, respectively). The identified distortions must be tackled so as to mitigate ED crowding, waste of resources and increase quality and responsiveness of care.

Ultrastructural changes of the palatal mucosa following application of 4-nitroquinoline-l-oxide (4NQO) in rats subjected to major salivary gland excision.

OBJECTIVE: It has been suggested that saliva exerts a protective role against the carcinogenic effect of various substances in the oral cavity. The objective of this study was to examine the ultrastructural changes of the palatal mucosa caused by the application of 4-nitroquinoline-l-oxide (4NQO) in the presence or absence of saliva. STUDY DESIGN: Wistar-Furth rats subjected and not subjected to total bilateral excision of the major salivary glands were either painted with an aqueous solution of 4NQO or with propylene glycol only (controls). Two animals of each group were humanely killed periodically. The areas of the palatal lesions were immediately sliced and processed for TEM examination. RESULTS: Ultrastructurally, the progressive changes to squamous cell carcinoma were observed in the animals painted with 4NQO. In the desalivated animals group, the ultrastructural alterations appeared earlier than in the group with salivary glands. CONCLUSIONS: Saliva appeared to delay but not hinder tumor induction by 4NQO.