Methods

Results

Discussion

References

Abstract

Background The bioreactance technique is a relatively new, totally non-invasive technique used to measure cardiac output that is easy to use. Non Invasive Cardiac Output Monitor (NICOM) is one such system (Cheetah Medical Inc). Although approved by FDA for measurement of stroke volume, there is a paucity of literature validating this technology in decompensated heart failure and cardiogenic shock (CS). Methods and Results Fifty patients admitted to our cardiac intensive care unit (CICU) for CS and Swan Ganz catheter guided therapy were prospectively enrolled in the study after informed consent. Simultaneous measurements of cardiac output (CO) were obtained using NICOM, indirect Fick (IF) and bolus thermodilution (TD). Intraclass correlation coefficient (ICC) was used to assess the precision of NICOM for CO using the three repeated measurements of CO over the pooled data. The agreement of the NICOM device in the defined clinical population compared to IF and TD, was evaluated by comparing the Pearson Correlation Coefficient, the Bland-Altman plot, and Lin’s Concordance Correlation Coefficient. ICC for cardiac output measured by NICOM showed excellent repeatability (ICC = 0.93, 95% CI = 0.92 – 0.94, n = 262) in the pooled data. Pearson Correlation Coefficient for cardiac output measured by NICOM was poor when compared to IF (n = 263, r = 0.132, p = 0.033) and TD (n = 258, r = 0.275, p < 0.001). Conclusions NICOM technology is not a reliable method of measuring CO in patients with decompensated heart failure and CS.

Introduction

Invasive nature of pulmonary artery catheter (PAC) is often cited as an added risk to the patients with unclear mortality benefit among critically ill patients. [1,2] Eisenberg et al. and Mimoz et al. demonstrated that physicians could predict cardiac output with only 50% accuracy in the absence of direct hemodynamic data [3, 4], indicating the need for hemodynamic monitoring in critically ill patients. Current guidelines support the use of invasive hemodynamic monitoring in selected patients in whom a clinical evaluation does not provide sufficient data to determine optimal medical therapy. [5] For these reasons, measurement of cardiac output (CO) using non-invasive or minimally invasive devices has gained popularity. The bioreactance technique is a relatively new, non-invasive technique used to measure CO. Non Invasive Cardiac Output Monitor (NICOM) is one such system (Cheetah Medical Inc).

1- Introduction

2- Methodology

3- Online cardiac monitoring

4- Discussion and reference model

5- Open issues, learned lessons and future trends

References

Abstract

The Internet of Health Things promotes personalized and higher standards of care. Its application is diverse and attracts the attention of a substantial section of the scientific community. This approach has also been applied by people looking to enhance quality of life by using this technology. In this paper, we perform a survey that aims to present and analyze the advances of the latest studies based on medical care and assisted environment. We focus on articles for online monitoring, detection, and support of the diagnosis of cardiovascular diseases. Our research covers published manuscripts in scientific journals and recognized conferences since the year 2015. Also, we present a reference model based on the evaluation of the resources used from the selected studies. Finally, our proposal aims to help future enthusiasts to discover and enumerate the required factors for the development of a prototype for online heart monitoring purposes.

Introduction

Recently, cardiovascular diseases (CVD) have become the leading cause of death in the world, accounting for about 17.7 million, or 31%, of all deaths in 2017 [1]. CVDs are associated with disorders of the heart and blood vessels which might often lead to cardiac arrhythmia, stroke, hypertension, and heart failure. CVDs are classified as chronic non-communicable diseases (NCD) which also include cancer, diabetes, and chronic respiratory diseases, for example [2]. NCDs are associated with a slow, long-term, or even lifelong advancement, and are usually silent or symptomatic, thereby affecting the quality of life [3]. Several factors contribute to the development of CVDs, but poor lifestyle management such as inadequate diet, sedentary lifestyle, stress, and legal and illicit drug use are determining factors for the progression of these pathologies [4]. Therefore, proper care can significantly reduce the likelihood of heart problems as well as routine examinations and follow-ups by a cardiologist as treatment expenditures become excessive when cardiovascular risk factors are poorly controlled.

1. Introduction

2. Basic Situation of External Cooling Water of the Converter Valve

3. Feasibility Analysis

4. The Structure of Remote Monitoring System

5.Conclusion

References

A sewage treatment remote monitoring system is designed in this paper due to long physical distance, complex process of zero discharge treatment techonology that requires monitoring. The paper firstly introduces the external cold water discharge of the valve cooling system of the converter station, and proposes the treatment measures for zero discharge of sewage in the case of phosphate, COD and salt in the discharged wastewater. The remote monitoring system is designed according to the process routes of the zero-emission scheme. The system adopts a hierarchical network control structure and mainly includes locale execution layer, harmonization layer, and remote monitoring layer. The mechanisms and functions of each layer are introduced in detail. The remote monitoring system is effective in monitoring the water treatment process and is helpful to achieve zero discharge of external cooling water in converter valve.

Introduction

The discharge of cold water outside the converter station accounts for more than 90% of the total discharge of the converter station. Therefore, how to efficiently deal with the discharge of cold water outside the converter valve is of great significance. On the basis of summarizing relevant literature and data as well as water quality testing data of cold water outside each station, this paper puts forward the technical scheme of zero discharge of cold water outside the station 1 . Zero discharge is an ideal closed water system, without external drainage. The water in the system is constantly recycled or reused after treatment, so as to save water resources and protect the environment 2 , which requires real time monitoring of the water treatment situation in different processes. A remote monitoring system is designed according to the process routes of the zero-emission scheme. The system adopts a hierarchical network control structure and mainly includes locale execution layer, harmonization layer, and remote monitoring layer 3 . The mechanisms and functions of each layer are introduced in detail. The remote monitoring system is effective in monitoring the water treatment process and is helpful to achieve zero discharge of external cooling water in converter valve.

Convolutional neural networks (CNNs) have been recently applied to tackle a variety of computer vision problems. However, because of its high computational cost, careful considerations are required to design cost-effective CNNs. In this paper, we propose a CNN inspired by MobileNet for fire detection in surveillance systems. In the proposed network, color features emphasized by the channel multiplier are extracted through depthwise separable convolution, and squeeze and excitation modules further increase the representation of the channel-wise convolution. Custom Swish is used as an activation function to limit exceedingly high weights from the effects of the channel multiplier. Our proposed network achieves 95.44% accuracy for fire detection, which is higher than those achieved other existing networks. Furthermore, the number of parameters used is 38.50% fewer than that of MobileNetV2, the smallest among other networks. We believe that using the proposed CNN, CNN-based surveillance systems could be implemented in lightweight devices without using expensive dedicated processors.

Introduction

Fires can occur anywhere, at any time, and if they are not detected early, they can cause severe damages to property and people. Surveillance systems consisting of multiple CCTVs can be very useful in detecting fires because they are designed to monitor the surroundings 24 hours a day. Furthermore, they can be very useful to monitor fires in a wide range of areas, including inaccessible areas. Consequently, there has been a huge demand for intelligent video-based fire monitoring systems that can alert people to respond quickly by processing and analyzing video streams in real-time. A video-based fire detection system can inform an operator by analyzing videos from CCTVs without using heat, smoke, or flame sensors. Owing to the significant development of video analysis, video signals from CCTVs can be automatically analyzed and can provide alarms to surveillance personnel to enable quick response. Traditional vision-based fire detection methods use handcrafted features, such as color, motion, and texture. Prior studies [1]–[4] detected fire by making full use of color features because fire is generally brighter and has higher contrast than other objects. Ko et al. [1] detected specific fire regions from their color and, then employed a model using wavelet coefficients to detect fire with a support vector machine (SVM) classifier.

Person re-identification (Re-ID), which is for matching pedestrians across disjoint camera views in surveillance, has made great progress in supervised learning. However, requirement of a large number of labelled identities leads to high cost for large-scale Re-ID systems. Consequently, it is significant to study learning Re-ID with unlabelled data and limited labelled data, that is, semi-supervised person re-identification. When labelled data is limited, the learned model tends to overfit the data and cannot generalize well. Moreover, the scene variations between cameras lead to domain shift in the feature space, which makes mining auxiliary supervision information from unlabelled data more difficult. To address these problems, we propose a Distilled Camera-Aware Self Training framework for semi-supervised person re-identification. To alleviate the overfitting problem for learning from limited labelled data, we propose a Multi-Teacher Selective Similarity Distillation Loss to selectively aggregate the knowledge of multiple weak teacher models trained with different subsets and distill a stronger student model. Then, we exploit the unlabelled data by learning pseudo labels by clustering based on the student model for self training. To alleviate the effect of scene variations between cameras, we propose a Camera-Aware Hierarchical Clustering (CAHC) algorithm to perform intra-camera clustering and cross-camera clustering hierarchically. Experiments show that our method outperformed the state-of-the-art semi-supervised person re-identification methods.

Introduction

Person re-identification (Re-ID) has received much attention in recent years due to its significance in video surveillance applications. When abundant labelled data is given, many works [1]–[7] have made great progress in supervised learning. However, labelling cost should be considered in largescale Re-ID system that consists of many cameras. To reduce labelling cost, studying semi-supervised learning to exploit unlabelled data and limited labelled data is a practical solution. Unsupervised person re-identification [8]–[15] has been studied to learn representation from unlabelled data, but how to effectively learn from limited labelled data is not considered in these methods. So far, semi-supervised person re-identification [16]–[20] is still under-explored. For semi-supervised Re-ID, exploiting unlabelled data and limited labelled data brings about some challenges. First, insufficient training data leads to overfitting for model learning and thus degrades generalization performance. Second, scene variations between cameras, such as illumination, background and viewpoint, cause domain shift in the feature space and create difficulty for mining auxiliary supervision information in unlabelled data to assist model training. The effect of scene variations is discussed in Section III-B later. To address the challenges for semi-supervised Re-ID, we propose a Distilled Camera-Aware Self Training framework, as shown in Figure 1.

Numerous methods have been developed in the context of expert and intelligent systems for structural health monitoring (SHM) with wireless sensor networks (WSNs). However, these techniques have been proven to be efficient when dealing with continuous signals, and the applicability of such expert systems with discrete noisy signals has not yet been explored. This study presents an intelligent data mining methodology as part of an expert system developed for SHM with noisy and delayed signals, which are generated by a through-substrate self-powered sensor network. The noted sensor network has been demonstrated as an effective means for minimizing energy consumption in WSNs for SHM. Experimental vibration tests were conducted on a cantilever plate to evaluate the developed expert system for SHM. The proposed data mining method is based on the integration of pattern recognition, an innovative probabilistic approach, and machine learning. The novelty of the proposed system for SHM with data interpretation methodology lies in the integration of the noted intelligent techniques on discrete, binary, noisy, and delayed patterns of signals collected from self-powered sensing technology in the application to a practical engineering problem, i.e., data-driven energy-efficient SHM. Results confirm that the proposed data mining method employing a probabilistic approach can be effectively used to reconstruct delayed and missing signals, thereby addressing the important issue of energy availability for intelligent SHM systems being used for damage identification in civil and aerospace structures. The applicability and effectiveness of the expert system with the data mining approach in detecting damage with noisy signals was demonstrated for plate-like structures with an accuracy of 97%. The present study successfully contributes to advance data mining and signal processing techniques in the SHM domain, indicating a practical application of expert and intelligent systems applied to damage detection in SHM platforms. Findings from this research pave a way for development of the data analysis techniques that can be employed for interpreting noisy and incomplete signals collected from various expert systems such as those being used in intelligent infrastructure monitoring systems and smart cities.

Introduction

Energy-efficient wireless sensor networks (WSNs) for structural health monitoring (SHM) have emerged due to progress in self-powered sensors and low-power data communication protocols. One such network is the through-substrate ultrasonic selfpowered sensor network (Das et al., 2017), which employs ultra-sonic pulses to communicate binary signals (from self-powered sensors) through the material substrate. However, the noted network creates time-delay on the generated signals due to the power budgeting required for sensing and data transmission. This study presents an expert system with a data mining approach for SHM in order to deal with such discrete noisy and delayed signals. Signal time-delay estimation/reconstruction is a problem that has attracted considerable attention in the SHM community. Numerous techniques have been developed for signal delay estimation for SHM and damage identification (Giurgiutiu, 2005; Giurgiutiu & Cuc, 2005; Nichols, 2003; Sun, Chaudhry, Rogers, Majmundar, & Liang, 1995; Yan, Royer, & Rose, 2010). Ultrasonic techniques (e.g., lamb wave methods) have been used for time-delay estimation for SHM (C. H. Dib & Udpa, 2016; Park, Farrar, di Scalea, & Coccia, 2006; Petculescu, Krishnaswamy, & Achenbach, 2007; Wang, Rose, & Chang, 2004). Aranguren, Monje, Cokonaj, Barrera, and Ruiz (2013) proposed a piezoelectric-based SHM system for damage detection using lamb waves and delayed signals. Kudela, Radzienski, Ostachowicz, and Yang (2018) developed an SHM approach based on lamb waves to increase damage imaging resolution, where the effectiveness of the approach for damage detection with time-delayed signals was demonstrated on a structural plate.

INTRODUCTION

There is multiplicity as well as ambiguity in the identification of Intangible Assets. Grojer and Johanson (1999) and Guthrie & Petty (2000) termed these as “Soft” assets and “Weightless Wealth”. Fincham and Roslender (2003) and Lev (2001) called these as “Intellectual capital” and “Knowledge assets”. Andersen (1992) describes intangible assets as non-physical in nature, capable of producing future economic net benefits. Edvinsson & Malone (1997) calls these as “hidden capabilities” of an organization. Brooking (1997) describes them as the “combination of market based intangible assets, intellectual property, human-centered and infrastructure that enables the company to function”. Strassmann (1999) defines intangible assets as the difference between the market value of a company and the book value of its tangible assets. By whatever name the Intangible Assets are called; these assets for sure have gained high prominence in the operations of companies especially in the contemporary decade. Investment in Intangible Assets helps the companies to be innovatory (OECD, 2008) and hence vie their competitors (Canibano et al., 1999; Boujelben and Fedhila, 2011). These are important components of a firm’s strategic planning and operations (Wheelen and Hunger, 2011). These assets enhance company’s market value (Mishra and Jhunjhunwala, 2009). These provide a firm with improved customer attainment and preservation (OECD, 2008). These help to build customer loyalty and strengthen the brand image of a company (OECD, 2008). Specifically mentioning the relevance of some of these assets, Goodwill is given due consideration at the time of mergers and acquisitions (PWC, 2014). A higher value is paid by a company with the expectation to take advantage of the existing technology, knowledge and name of an established reputed firm (Canibano et al. 1999; Gu and Lev, 2001; PWC, 2014). Investment in R&D helps in producing novel products (Pradhan, 2003). Patents, copyrights and trademarks lead to a superior performance by extending monopoly to the firms (Greenhalgh and Rogers, 2012). Authors like Hirschey and Weygandt (1985); Chauvin and Hirschey (1993); Bosworth and Rogers (1998); Andras and Srinivasan (2003); Lantz and Sahut (2005); Nagaoka (2006); Heiens et al. (2007); Ehie and Olibe (2010); Boujelben and Fedhila (2011) have shown that R&D intensity is positively associated with the firm performance. Similarly, Advertising and Customer Relationship Management (CRM), another category of Intangibles, are documented as generator of high profits for the companies. Advertised products are easily recallable and identifiable which minimizes the chances of confusion and boost sales (Sahay and Pillai, 2009). Studies as Hirschey and Weygandt (1985); Chauvin and Hirschey (1993); Sahay and Pillai (2009); Kundu et al. (2010); Boujelben and Fedhila (2011) found a positive relation of advertising intensity with firm performance. Last but not the least; human asset these days is regarded as the most vital animate Intangible Asset. It forms intellectual capital of companies that helps in earning high profits through their skills and abilities (Arrighetti et al., 2014). Findings of Pew Tan et al (2007); Kamath (2008); Wang (2008); Ghosh and Mondal (2009); Clarke et al (2011); Wang (2011); Pal and Soriya (2012); Mondal and Ghosh (2012) confirm that Intellectual Capital enhances the firm performance.

1- Introduction

2- Related work

3- Proposed cognitive IoT-cloud framework and seizure detection technique

4- Results

5- Conclusion

References

Abstract

We propose a cognitive Internet of Things (IoT)–cloud-based smart healthcare framework, which communicates with smart devices, sensors, and other stakeholders in the healthcare environment; makes an intelligent decision based on a patient’s state; and provides timely, low-cost, and accessible healthcare services. As a case study, an EEG seizure detection method using deep learning is also proposed to access the feasibility of the cognitive IoT–cloud smart healthcare framework. In the proposed method, we use smart EEG sensors (apart from general healthcare smart sensors) to record and transmit EEG signals from epileptic patients. Thereafter, the cognitive framework makes a real-time decision on future activities and whether to send the data to the deep learning module. The proposed system uses the patient’s movements, gestures, and facial expressions to determine the patient’s state. Signal processing and seizure detection take place in the cloud, while signals are classified as seizure or nonseizure with a probability score. The results are transmitted to medical practitioners or other stakeholders who can monitor the patients and, in critical cases, make the appropriate decisions to help the patient. Experimental results show that the proposed model achieves an accuracy and sensitivity of 99.2 and 93.5%, respectively.

Introduction

The Internet of Things (IoT), which can be considered an interconnected network of intelligent sensor devices, often has limited storage and low processing power capability. IoT, together with cloud computing, which has a large storage and sufficient processing power capability, has made essential services, such as smart healthcare [1, 2], possible in a smart city environment. However, monitoring and communicating remotely with patients are necessary in such environments. In addition, the need to provide low-cost, high-quality, and patient-centric smart healthcare to patients has emerged. The advancements in the field of IoT [3] and cloud technologies [4] has resulted in a tremendous demand for realtime, intelligent, and remote healthcare services under the paradigm of smart cities. Furthermore, the integration of IoT and cloud technologies has provided a seamless and ubiquitous framework for smart healthcare monitoring. At present, residents in smart cities have access to smart sensor devices and advanced mobile technologies. In an environment such as that of a smart city, finding specialized doctors, healthcare centers, and hospitals nearby is difficult. The movement of patients in critical conditions is also quite difficult. Hence, we need to create a smart healthcare monitoring framework by integrating the resources available at our disposal to improve the quality and accessibility of healthcare services. In such a smart healthcare monitoring framework, we can transmit and process medical-related multimedia signals from smart sensors and mobile devices to provide timely assistance and quality healthcare services to patients. However, such healthcare data and signals are often naturally large and challenging to handle because of their complexity. The healthcare industry has emerged as one of the major industries with tremendous demands. Apart from providing patients with critical and crucial services, this industry is also generating large revenues for the government and the private sector.

Internet of Things (IOT) is advanced area which includes embedded technology, wireless sensor network, artificial intelligence and machine learning etc. to intercommunicate things. IOT improves user convenience and also enhances data collection. It optimizes various technologies in single area called IOT. IOT has applications across different fields like industries, engineering and medical. To improve existing medical systems IOT is applicable. In medical IOT, Wireless Body Area Network (WBAN) is key concept which is network of autonomous medical sensors which are deployed inside or outside of the patient body for health monitoring. In WBAN, deployed wearable sensors senses health conditions like pulse rate, heart beats, blood pressure, temperature etc. using IOT exploited devices. Data from multiple sensor nodes is collected to sink called data fusion and it is transmitted to analyser. In wireless sensor networks, data is passes through communication from one end to another end. Information can be travel hop by hop or multihop manner in the sensor network [5]. But for reliable data delivery over the network, faster data transmission is necessary during communication. So, different types of routing protocols are used by researchers. Routing protocols are helps to search minimum distance path between sources and sink node [6] [7]. Energy efficiency is important factor in wireless sensor networks. Heavily congested network will be consuming more energy in the network. So, different MAC and routing protocols are used for to increase the energy efficiency of the node [8]. Some hybrid protocols also designed for reduce power consumption by the nodes. Clustering technique is a better solution for achieve energy efficiency over the network communication [9] [10]. Energy parameter effected due to the collision of packets during packets travel from one node another node. In case of large application, multiple nodes can send packet at a same time to the receiver end [11] [12]. So collision event will be generating near the receiver end. Hybrid MAC protocols are designed and developed for solve congestion and collision problem in wireless sensor networks [13][14]. Routing as well as MAC protocols are necessary for reliable data delivery in the network.[15] In case of intermediate node failure, routing protocols should be select alternative path for packets delivery over the network [16] [17].

Diabetes Mellitus is a chronic disease that affects about 347 million people (about 5 percent of the world’s population) out of which more than three-quarters live in developing countries. People suffering from diabetes need to be monitored, since their blood sugar ought to be regulated. However, there are several remote diabetes patients monitoring systems but mostly they cannot make decision about the patients; this is very essential if we look at the challenges faced in developing countries like lack of qualified medical personnel and health care facilities among others [1]. On the other hand, to develop any system, application, or software the following system development stages need to be followed, these include identifying the requirements/specifications, design, development, testing and finally implementation [2, 3]. In the course of the design, the architecture of the system need to be defined first then the system will be designed based on that architecture [4, 5]. The architecture is the abstract representation of the subsystems, and the flow of information within subsystems that make up the whole system [5, 6]. In this work, we are proposing an architecture of a system that would help doctors to make decision on their patients; these types of systems are commonly referred to as Decision Support Systems. Decision Support System (DSS) is an information system that supports decision making activities in an organization, quite often DSSs are interactive [7]. A DSS that focus on clinical applications is called Clinical Decision Support System (CDSS). CDSS as described by Jao and Hier [8], is a system that uses case-based reasoning to help clinicians access disease status, make diagnosis, select appropriate therapy or make other clinical decisions. The use of words like support, help, etc. in defining DSS is very important, especially to clarify the fact that DSS are not replacing the experts or decision makers [9]. Generally, DSS application consists of five basic components but the content and structure may vary, these are: database, model base, knowledge base, user interface and users [10]. The database stores the data that is used for the decision making, it also allows the user to access and manipulate the data. Model base allows the analysis of the problem using models like financial, statistical, etc. Knowledge base is like a repository where knowledge about the problem is kept; usually it represents the expert in a particular field. User interface deals with communication between the user and the DSS. Users are the individuals that use the DSS; in other word are the decision makers [11].

I. INTRODUCTION

Over the last few years, investment on contemporary medical equipments has increased rapidly. These equipments gradually play a very important role in helping doctors and clinician’s works. Patient monitoring system is one of essential devices in hospitals. They measure vital biological signals such as ECG, RESP, SpO2, Blood pressure NIBP and temperature TEMP… almost hospitals from urban to rural sector has been equipped patient monitoring system from several to tens system based on hospital scale [1]. However, all these devices are working separately which monitor and measure individual parameters from single patient. From studies in several clinics and opinions from healthcare authorities, we propose and design a WiFi central patient monitoring system. Data from individual monitor has been retrieved and send to wireless transmitter. In centre wireless receiver, data from different monitoring systems has been archiving and then display parameters from different monitoring systems simultaneously. In the centre monitor, doctors can read and review data from many patients which facilitate monitoring and therapy process. Besides, doctors can respond to urgent cases rapidly, effectively and precisely. With this model, clinics can easily employ and implement without affect to current workflow of hospitals and with low investment cost.

Information technologies have played key roles in the medical field. New needs are regularly emerging, and computer technology is in the Frontline of the answers to those demands. Classical computation paradigms fall short when trying to model an environment with a large number of users and complex processes and interactions. The aim of this paper is to propose a conceptual model for the development of a Medical Information System (MIS) operating in a distributed environment, and based on Multi-Agent Systems (MAS) and allows intelligent patient monitoring. To demonstrate the feasibility, we developed a prototype under JADE (Java Agent DEvelopment Framework).

In recent years, there has been an ever-growing need for a sustainable health system which manages not only acute care (in hospital wards or emergency departments), but also the care of outpatients, especially those with chronic conditions. Worldwide spending on chronic conditions has increased to a point that immediate action is now required. A chronic condition is defined as a health condition that can be managed but often cannot be cured; common examples include heart disease, stroke, cancer, diabetes and arthritis [1]. The number of people with multiple chronic conditions is likely to continue to increase with the ageing of the population. Major contributing factors are unhealthy lifestyles, along with the impact of the economic downturn on mental and physical health [1]. Wearable patient monitoring (WPM) systems are emerging as an effective tool for the prevention, early detection and management of chronic conditions. Recent estimates for annual expenditure on healthcare in the US are between $US 210.9 billion and $US 306 billion [2]. In the UK, nearly 29% of the total population now lives with a chronic medical condition, and as much as 80% of the healthcare budget is spent on the management of chronic diseases [1, 3]. As wearable sensors, smart textiles and body-worn garments become smaller, cheaper and more consumer-accessible, it is expected that they will be used more extensively across a wide variety of contexts. The expansion of wearable systems for health data collection offers the potential for user engagement and self-management of chronic diseases [4]. The rapid increase in the global adoption of WPM systems has occurred over the last couple of decades. Some of the early foundation research includes; (1) the wearable healthcare system (WEALTHY) project, which investigated the use of fabric integrated sensors to continuously monitor the vital signs of highrisk patients, including those undergoing rehabilitation [5]; (2) an investigation of a custom-developed ubiquitous healthcare (u-health-care) system consisting of custom 802.15.4-capable nodes interfaced with electrocardiogram (ECG) and blood pressure sensors, as well as a basic cell phone device for data display and signal feature extraction [6]; (3) the Human++ project in the Netherlands which developed a body area network consisting of three sensor nodes and a base station [7]; (4) the CodeBlue project developed by researchers at Harvard University [8]; (5) research from the Media Laboratory of the Massachusetts Institute of Technology (MIT) which involved designing LiveNet [9]; (6) the development of the SmartVest, a wearable physiological monitoring system that consists of a vest and a variety of sensors integrated into the garment’s fabric to collect several biosignals [10]; (7–9) and three European IST FP6 programs (the MERMOTH, MyHeart and HeartCycle projects), which are recent examples of WPM systems [11–14]. The aim of this review is to investigate how current technological barriers and challenges limiting the global clinical adoption of bed-side patient monitoring have been reduced by the use of WPM systems. In addition, this review will highlight opportunities and recommend the best possible approach for the sustainable adoption of WPM in acute care, as well as in community care settings. This continues on from previous literature reviews conducted on WPM systems [6, 15–18].