Technological evolution has allowed that tasks, usually performed by humans, can now be performed accurately by automated systems, often with superior performance. The healthcare area has been paradigmatic in the automation of processes, as the need to optimize costs, ensuring the provision of quality care, is crucial for the success of organizations. Diabetes, whose prevalence has increased significantly in the last decade, could be a case of application of several technologies that facilitate diagnosis, tracking and monitoring. Such tasks demand a great effort from health systems, requiring the allocation of material, human and financial resources, under penalty of worsening symptoms and emergence of serious complications. In this chapter the authors will present and explore how different technologies can be integrated to provide better healthcare, ensuring quality and safety standards, with reference to the case of diabetes.
TopIntroduction
Diabetes is a metabolic disease characterized by uncontrolled blood glucose regulation mechanisms. This chronic disease occurs when the pancreas does not produce enough insulin or when the body cannot use it effectively. The disease prevalence is on the rise worldwide, affecting 8.8% of the world’s adult population in 2017, with the anticipation of a further increase to 9.9% by 2045 (World Health Organization, 2021). Beyond the personal and social consequences of untreated diseases, from a clinical perspective, the diagnostic, monitoring and treatment of this condition represents a major effort for healthcare systems. Material and human resources must be allocated to ensure an adequate tracking of each case guarantying that symptoms are controlled. A bad prognosis can easily lead to severe health conditions.
Chronic hyper-glycemia associated with uncontrolled diabetes damages various organs and systems, causing chronic diabetic complications, leading to disabilities, poor quality of life, and ultimately death. Diabetic peripheral neuropathy (DFN) is a major complication of diabetes mellitus, being the leading cause of foot ulceration and lower extremity amputations (Walicka et al., 2021). All patients with diabetes must have their feet evaluated, at least, once per year for the presence of the predisposing factors for ulceration and amputation (neuropathy, vascular disease, and deformities) (Boulton et al., 2008).
The Semmes-Weinstein monofilament examination is the recommended procedure for screening plantar sensitivity, as an early biomarker for DFN (Monofilament Testing for Loss of Protective Sensation of Diabetic/Neuropathic Feet for Adults & Children, 2012). The examination, performed by a clinician, consists of touching the plantar surface with a 10gf calibrated monofilament, on specific test locations, and wait for the patient’s sensitivity feedback. This process is widely used but can be improved in several aspects: a) The monofilament quality can vary with usage, environmental temperature and humidity; b) The task can be tedious when large populations must be screened; c) Only a “feel”/”don’t feel” feedback is registered while more variables can be observed and considered (image, force, time-tracking, among others) (Martins & Coelho, 2021).
The present chapter aims to describe how modern technologies, such as computer vision, artificial intelligence, cloud storage and computing, and robotics can be integrated and used to tackle healthcare related challenges, contributing to better and more cost-effective services. To describe the involvement of each technology we base our description on an application case for diabetic foot management.
The proposed chapter will be organized as follows. After presenting the underlying motivations and, in general terms, the consequences and prognoses in an unaccompanied evolution of diabetes, we will present a pipeline proposal for the automated management of the diabetic foot, a potential major complication of this condition. The following section will focus on the topic of computer vision. Here, the general architecture of these systems is exposed, followed by a brief explanatory passage of the various stages involved in them. A set of sensors used for image acquisition and processing are presented, followed by an explanation of useful techniques and tools. The following section will be related to robotics, more particularly in their collaborative versions, where their main characteristics are highlighted. In addition, attention will be given to a set of systems and real applications where the use of robotics shows potential in terms of support for medicine, with the intention of framing its use in the development of the proposed system. In the next section, the role of the cloud as a support technology for the system's operation will be discussed and the main advantages and disadvantages of its use will be presented. Finally, the main conclusions are presented and some guidelines for the future are established.