In recent years, outstanding growth of the Internet of Things (IoT) was enabled by the parallel progress of sensing technologies, miniaturization of electronic devices, cost reduction, widespread diffusion of consumer electronics, ubiquitous connectivity, and rise of cloud computing. Every technological advancement ideally has the ultimate goal of bringing benefit to humanity, hence the IoT paradigm unveiled the great potential of connected sensor systems for the improvement of the health conditions of all living being [1].

In this scenario of pervasive electronics, many IoT systems have been developed for monitoring vital signs and have later become commercially available. Examples include systems for the monitoring of the heartbeat [2], blood pressure [3], oxygen saturation [4], glucose level [5], breathing rate [6]. The monitoring of vital signs can be achieved by embedding sensors in wristbands, smartwatches, chest straps, or wearable apparel, and provides immediate feedback on the health condition of people that can be used either for anamnestic reasons or to trigger timely intervention in case of sudden events (e.g., heart attacks, strokes, suffocation) [7]. In the long term, health is strongly affected by environmental conditions, hence systems that monitor the air quality in terms of pollution and concentration of harmful particles also fall within the user-centered IoT measurement systems for human well-being [8]. In a broader context, considering the improvement of people’s general well-being, sensors, and IoT systems can be employed to monitor the perceived ambient conditions and eventually operate to set parameters, such as temperature and humidity, according to personal preferences [9,10,11,12]. Moreover, connected sensor systems also enable health preservation by doing environmental surveillance for safety reasons, for example, by checking constantly for the presence of smoke or toxic substances [13,14] or by performing structural health monitoring [15,16].

Figure 1 recaps the main examples of IoT measurement system applications designed to improve human well-being.

IoT measurement systems find application either in wearables or in smart homes and buildings, which include home and work environments. In wearable systems, sensors are embedded in clothes, fabrics, or accessories that the user carries on-body and are, therefore, suitable for monitoring vital signs and movement of all people. In home and work environments, sensors are hosted in the furniture or assembled in self-standing systems, therefore, they interact with the person either directly (e.g., sensors embedded in chairs) or indirectly (e.g., location systems detecting people’s presence in the room). Systems designed for home environment are generally aimed at the assisted living of elders. Systems designed for work environments target the preservation of the health and safety of workers. For example, in the context of smart homes and assisted living, Stern et al. [17] recently described a system for in-bed posture recognition by convolutional neural networks based on the pressure data acquired from a pressure mat positioned on the mattress aimed at investigating the relation between the sleeping positions and the insurgence of pressure sores in bedridden people. Leone et al. [18] developed a posture recognition machine learning algorithm for the classification of four postures: Standing, sitting, bending, and laying down. Concerning the monitoring of workforce health conditions, Lind et al. [19] overviewed the available systems based on motion capture instruments to collect kinematics data for the prevention of work-related musculoskeletal disorders.

Many jobs of the present time involve the use of personal computers for a long time. Typically, computer-based jobs require people to sit during work hours, enhancing the risks correlated to sedentary lifestyles. Among these, the problems related to a prolonged incorrect sitting posture are largely diffused and represent a health issue that many people around the world must deal with. The negative effects of prolonged bad sitting postures include musculoskeletal disorders involving the hip joints, lumbar area, upper back, shoulder blades, shoulders, and neck [20]. The condition of persistent loading on the intervertebral disks associated with poor postures is also responsible for the insurgence of low back pain [21], which affects people of all ages and is one of the main causes of absence from work [22]. Back pain is especially frequent in people with scoliosis, a pathological curvature of the spine that can cause chronic pain and prolonged asymmetric postures is, therefore, even more relevant for people who suffer from this condition. Although the incidence of scoliosis is not gender-related, it is found that female subjects are more likely to experience a progression of the spine curve [23]. The wide interest in posture-related health issues has originated several proposals for monitoring the workers’ sitting behavior.

In the past, different approaches to monitoring the sitting postures of long-time sitting workers were investigated in the context of the IoT applications for health risks prevention. Originally, several systems were proposed based on video analysis, though they never took place due to privacy issues. Then, textile sensors were largely considered for the development of smart fibers and wearables that could analyze the person’s posture by relying on dedicated posture classification algorithms. More recently, thanks to the widespread diffusion of low-cost and well-performing pressure sensors, the first solutions were replaced by sensor array-based systems.

2. IoT Systems for Real-Time Posture Asymmetry Detection