Nerea Arandia, Conceptualization , Investigation , Writing – original draft , 1, * Jose Ignacio Garate, Methodology , Writing – review & editing , Supervision , 2 and Jon Mabe, Methodology , Writing – review & editing , Supervision 1
1 TEKNIKER, Basque Research and Technology Alliance (BRTA), 20600 Eibar, Spain
Find articles by Nerea Arandia2 Department of Electronics Technology, University of the Basque Country (UPV/EHU), 48080 Bilbao, Spain
Find articles by Jose Ignacio Garate1 TEKNIKER, Basque Research and Technology Alliance (BRTA), 20600 Eibar, Spain
Find articles by Jon Mabe Christian Baumgartner, Academic Editor and Annie Lanzolla, Academic Editor 1 TEKNIKER, Basque Research and Technology Alliance (BRTA), 20600 Eibar, Spain2 Department of Electronics Technology, University of the Basque Country (UPV/EHU), 48080 Bilbao, Spain
* Correspondence: se.rekinket@aidnara.aeren Received 2022 Nov 2; Accepted 2022 Dec 14. Copyright © 2022 by the authors.Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
The evolution of technology enables the design of smarter medical devices. Embedded Sensor Systems play an important role, both in monitoring and diagnostic devices for healthcare. The design and development of Embedded Sensor Systems for medical devices are subjected to standards and regulations that will depend on the intended use of the device as well as the used technology. This article summarizes the challenges to be faced when designing Embedded Sensor Systems for the medical sector. With this aim, it presents the innovation context of the sector, the stages of new medical device development, the technological components that make up an Embedded Sensor System and the regulatory framework that applies to it. Finally, this article highlights the need to define new medical product design and development methodologies that help companies to successfully introduce new technologies in medical devices.
Keywords: embedded systems, medical devices, new product development, medical device regulationEmbedded Sensor Systems have become the key element of the advances in medical devices; the high versatility they offer enables the development of new diagnostic and advanced monitoring devices for patients in both home and hospital environments.
The medical device industry is regulated by different national notified or regulatory bodies. Two of the world’s main regulatory bodies are the European Commission Directorate and US Food & Drugs Administration (FDA). Figure 1 shows the most important regulatory authorities around the world.
Regulatory authorities around the world.
In the European Union, medical devices are regulated by harmonised health regulations. Any manufacturer who wants to put a device on the European market must go to a notified body to have its device assessed [1]. If it is considered approved, a certificate of conformity with the CE mark is emitted, which allows it to be sold in all the countries of the European Union [2].
According to the European Medical Device Regulation (MDR), the following are considered medical devices:
“Any instrument, apparatus, appliance, software, implant, reagent, material or other article intended by the manufacturer to be used, alone or in combination, for human beings for one or more of the following specific medical purposes: Diagnosis, prevention, monitoring, prediction, prognosis, treatment or alleviation of disease, diagnosis, monitoring, treatment, alleviation of, or compensation for, an injury or disability, investigation, replacement or modification of the anatomy or of a physiological or pathological process or state, providing information by means of in vitro examination of specimens derived from the human body, including organ, blood and tissue donations and which does not achieve its principal intended action by pharmacological, immunological or metabolic means, in or on the human body, but which may be assisted in its function by such means [. ]”
In contrast, the U.S. Food and Drug Administration (FDA) considers a medical device:
“An instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article, including a component part, or accessory which is: (A) recognized in the official National Formulary, or the United States Pharmacopoeia, or any supplement to them, (B) intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals, or (C) intended to affect the structure or any function of the body of man or other animals, and which does not achieve its primary intended purposes through chemical action within or on the body of man or other animals and which does not achieve its primary intended purposes through chemical action within or on the body of man or other animals and which is not dependent upon being metabolized for the achievement of its primary intended purposes”
In both cases, we can see that to discern whether a product is a medical device or not, it is necessary to define its intended use [5]. For example, smartwatches, which have become so common recently, depending on their intended use, can be considered medical devices. A smartwatch that measures the pulse or performs an electrocardiogram is considered a medical device only if the readings are used to perform medical actions such as diagnosing a disease or establishing a treatment [6]. On the other hand, with the same technology, this smartwatch is not considered a medical device if it is only used for informative purposes. For example, when a user is interested in measuring the pulse rate after a workout. This is the category for devices such as the Apple Watch [7].
The design and development of these systems require considering very strict regulations from the conception phase of the idea. In [8], a study is presented detailing the challenges for the development of medical devices according to FDA regulations. Currently, regardless of country or region, the design and development of medical devices are regulated by several standards that guarantee the quality of the devices and minimise the exposition risk to healthcare professionals and patients [9,10].
The regulations to be applied, as well as their interpretation, depends on the intended use of the medical device and the technologies used. For example, there are purely software devices [11], ones that include software and electronics, or purely mechanical devices. For all these cases, there are common and specific regulations as well [12].
Innovation in the medical sector is often driven by start-ups, which often have great ideas but lack experience in the development of medical devices in accordance with medical regulations. Inexperience in the sector, together with difficulty in identifying these regulations and translating them into technical requirements, results in the development of new innovative medical devices that are not always successful. Article [13] highlights the difficulty for start-ups to cope with the new regulation that applies to the development of medical devices. In addition, the author of [14] states that there is a risk of slowing down innovation because the new MDR requires more costly and high-quality testing. It also requires more technical documentation to comply with the new regulation.
Therefore, the goal of this article is to review the major challenges and requirements in the development of Embedded Sensor Systems for medical devices. To this end, Section 2 presents the context of this article, the innovation in Health Technology. In Section 3, the possibilities offered by embedded systems in the healthcare sector are introduced. Then, in Section 4, the design and development process of a medical device is outlined. In Section 5, the technological blocks that compose an Embedded Medical Sensor System are described and in Section 6, the regulations that apply to each of the technological blocks are presented. Finally, Section 7 states the conclusions of this work and some insights about future lines of work.
There are several articles where technical and regulatory solutions are partially discussed, for example, in [15]. The author reviews the main regulatory challenges that a wearable sensor faces in order to be considered a medical device. Although the main regulations are discussed, the author does not go into detail on the main requirements of all of them. Likewise, in [16], different challenges associated with injectable hydrogels are presented. Emphasis is placed on the technical challenges, but no specific regulatory requirements are discussed. In [17], the author presents different technologies for measuring parameters related to sleep-disordered breathing. However, this article does not analyse the technical solutions from a regulatory point of view. Rather, it focuses on a technical comparison of measurement solutions. Similarly, in [18], the existing sensor technologies to be integrated into wearable solutions are reviewed.
This article, in contrast, aims to delve deeper into the technical and regulatory aspects that embedded sensor solutions must comply with to be compliant with medical product requirements. To this end, this article focuses on identifying those key points that designers of medical devices with Embedded Sensor Systems must consider. It also seeks to identify challenges and requirements that are common to embed medical devices and not specific to a particular solution or application.
The medical device industry has constantly been evolving over the last few years. On the one hand, new healthcare challenges are emerging, such as COVID-19 or the problem of an elderly society [19]. On the other hand, the rapid evolution of technology is making it possible to improve current medical devices and solutions.
COVID-19, a disease caused by the new coronavirus known as SARS-CoV-2, emerged on 31 December 2019 in Wuhan (China) [20]. In spite of becoming a global pandemic, COVID-19 also introduced technological innovation into the healthcare sector. Tele-medicine has come to stay; before COVID-19, it was usual to go to the doctor face-to-face [21]. However, due to the collapse of health systems, mobile applications and information systems for patient care have been developed around the world [22]. Likewise, disinfecting robots [23], devices for monitoring temperature in public spaces [24] or low-cost oximeters for home use [25] are clear examples of the technological evolution that this pandemic has brought about.
The problem of an ageing population is another challenge to be faced. The increase in life expectancy and the considerable decrease in the birth rate make it essential to take measures to help manage and optimise patient care. The WHO (World Health Organisation) estimates that between 2015 and 2050, the world’s population over the age of 60 will increase from 12% to 22% [26]. In this context, technological developments oriented towards patient monitoring in both home and hospital environments are of special relevance.
Not only the emergence of new challenges in the sector has brought new technological innovations. The development of technology in aspects such as the Internet of Things (IoT) and Artificial Intelligence (AI) also makes it possible to develop a multitude of innovative Medical Technologies [27]. As a result, new solutions and devices have appeared in the healthcare sector that allow (i) the prevention of diseases or damages [28,29], (ii) the diagnosis of diseases or special conditions [30,31], (iii) the monitoring of the patient’s condition [32,33], (iv) helping treat and overcome diseases [34,35] and (v) caring for and facilitating the process of patient recovery [36,37].
Some of the advances and developments that are revolutionising the medical sector include:
Optimisation of data management: an increasing number of mobile or portable devices are being used to monitor all kinds of patient parameters, temperature, oxygen, pulse rate, etc. [38]. Storing, organising and analysing all this data is not an easy task. However, thanks to Big Data and AI, it is possible to manage huge volumes of data in an efficient way [39,40]. The wearable system proposed in [41] is able to diagnose diabetes using machine learning and big data. In [42], a big data system is developed to support the rehabilitation of strokes and lung diseases. The heterogeneity of data capture systems leads to the development of architectures to support such different solutions. In [43], a semantic big data architecture to address the heterogeneity of data between different wearable platforms is presented.
Artificial vision systems: these systems are a great enabler for the development of new systems in the field of healthcare. Complex systems, such as endoscopes, radiology, ophthalmology, surgery, etc., use this technology. Moreover, embedded vision, which is based on the integration of adapted camera modules that are directly incorporated into medical devices, enables intelligent image processing in a variety of portable applications. One of them are eye-tracking systems, which can be used for diagnostics or patient care. In [44], a vision algorithm is presented to detect eye movement for the identification of ocular pathologies, such as strabismus. In [45], an algorithm is presented that, used together with the Irisbond eye-tracker [46], is able to assess mathematics in children with cerebral palsy. Eye-tracking systems can even be useful for healthcare professionals. In [47], the possible use of eye-detection systems to assist neonatal resuscitation processes is presented. In [48], a pilot study is presented for the same purpose.
Early diagnosis of diseases: in a few years, artificial intelligence will make it possible to diagnose diseases such as lung cancer [49]. The analysis of thousands of digital scans will identify early stages of cancer that would not have been possible with traditional technology [50]. In [51], different AI algorithms used for the diagnosis and treatment of prostate cancer are reviewed. Similar analyses are also supported by [52] for colorectal cancer detection and [53] for breast cancer detection.
Patient monitoring at home and in the hospital: vital signs monitoring allows the patient’s progress to be evaluated and ensures early detection of undesirable effects. Advances in embedded electronics with integrated sensorisation allow reliable measurements of temperature, oxygen, pulse or blood pressure using comfortable, self-powered devices. More and more work is being performed on the development of wearable solutions, such as smartwatches, that allow continuous patient monitoring in a non-invasive way [54]. Article [55] presents a system capable of measuring heart rate, SpO2 and respiratory rate. It is a low-cost system, which makes it interesting for deployment in low-resource settings. There are also solutions in the literature that are capable of detecting falls [56,57]. These systems are especially interesting for elderly or very fragile patients [58].
In this context, the Research and Development (R&D) strategies of leading medical companies indicate the need to evolve and develop the current technology. For this purpose, many of them have alliances with universities and research centres in which they invest a large percentage of their annual revenues. Roche, the world’s largest biotech company, with revenues of about 65 billion Swiss francs in 2021 [59], invests around 9 billion Swiss francs in R&D every year [60], one of the highest innovation spending figures across all sectors. Medtronic, a leading manufacturer of medical technologies whose portfolio includes infusion pumps, medical devices and advanced electrical instrumentation for surgery, with a revenue of about USD 30 billion [61], invests more than USD 2.5 billion in R&D each year [62]. Other leading companies in the sector, such as Siemens [63] or Abbott [64], also invest around 10% of their turnover in R&D [65].
Medtronic launched the first patient procedure with the Hugo RAS robotic-assisted surgery system in 2021. This platform includes AI technology that records and processes images from the operating room [66]. Roche, through its collaboration with Microsoft since 2017, is transforming in vitro diagnostics with solutions based on the Microsoft Azure IoT Platform. In this way, Roche achieves intelligent and remote management of its in vitro devices [67]. Based on the success of this collaboration, in December 2021, Roche and Microsoft have signed a new agreement to integrate AI and cloud technology into their devices [68].
The evolution of technology, together with the multitude of programs promoted by the administrations to improve the healthcare of citizens, is generating a massive wave of start-up companies that aim to design, develop and put new medical devices on the market. Moreover, according to a report published by the Spanish Association of Business Angels Networks (AEBAN), 40% of the start-ups that have been created in Spain during the last few years belong to the medical sector [69]. It also explains that the most attractive sectors for Business Angels are mobility, health and energy.
Among the Spanish start-ups, Koa Health, Inbrain Neuroelectronics and MedLumics stand out as the start-ups with the most funding in 2021, with the three companies together totalling more than 60 million euros [70]. Koa Health works on a wide range of mental health solutions, from digital wellness to digital therapy, with the tools it offers aiming to improve the mental well-being of users anytime and anywhere [71]. Inbrain Neuroelectronics is developing a minimally invasive neural interface that can detect and modify specific biomarkers using AI and Big Data to help improve personalised neurological therapies [72]. MedLumics specialises in the development of cardiac optical imaging ablation devices for atrial treatment [73].
Embedded systems are electronic devices that are specifically designed to perform certain functions. They provide high levels of system integration for the development of manufacturing processes and the use of goods and services.
Embedded systems are usually composed of hardware, firmware and software. The hardware consists of the physical, electronic components that are required to fulfil the functionality of the embedded system. The main element is usually a processing unit (microprocessor, microcontroller or Digital Signal Processor) that controls the integrated circuits, such as memories, analogue-digital converters, power supplies or battery controllers. The software is a set of instructions or programs that are programmed in the processing unit to respond to specific use cases or functions of the system. The firmware is the set of instructions implemented at the processing unit to control the electronic circuitry. Firmware is considered the link between hardware and software.
Embedded systems have been transforming the healthcare industry over the last few years. An increasing number of smart devices are enabling continuous monitoring of vital signs, glucose, etc. Ref. [74] reviews different embedded solutions for monitoring vital signs. It presents several solutions based on smartwatches or even sensors integrated into textiles or lenses. These smaller and connected devices are making it easier to capture and transmit this information to healthcare centres. Article [75] presents how smart embedded systems offer secure, low-cost communication interfaces for healthcare services. Once a significant amount of data is available, this information can be post-processed using AI diagnostic algorithms to improve the results of the diagnosis. In [76], there is evidence that AI can improve the diagnosis of rare diseases. For example, pacemakers made by embedded systems are a significant breakthrough for patients with heart disease. These devices can monitor heartbeat and react to cardiac malfunctions [77]. They also register all the data so that doctors can adjust the patient’s therapy in a more efficient way.
Sensors for healthcare monitoring are usually devoted to measuring vital signs. Currently, four basic parameters are defined as vital signs [78]: blood pressure, heart rate, respiratory rate and body temperature.
According to WHO, blood pressure is defined as “the force exerted by circulating blood against the walls of the body’s arteries“ [79]. Among the solutions for blood pressure measurement, oscillometric systems that are able to analyse the vibration of the arterial wall based on the signal transduction method. In [80,81], different wearable designs based on capacitive sensors are presented. There are also auscultatory systems based on microphones that can interpret sounds during the measurement process, but due to their measurement principle, these should be used in low-noise environments [82]. There are also other methods that allow the estimation of blood pressure, an example of which is the one presented in [83], which is able to estimate blood pressure using a photoplethusmogram or the one presented in [84], which estimates blood pressure without contact using video analysis.
Heart rate, heart beeps per minute, it is commonly measured by electrocardiographs that measure the potential generated by the electrical signals that control the expansion and contraction of the heart. In [85], the author presents the design of a portable electrocardiograph. It can also be measured by optical systems that determine the heart rate by emitting a beam of light into the subchoroidal vessels and measuring the reflected light in a photo-sensor. There are many wearable developments based on the optical system as its measurement principle does not require the use of electrodes, making it suitable for these systems [86,87]. In [88], both measurement methods are compared. This study concludes that although the traditional electrocardiograph-based measurement is the most reliable, with the optical system, it is possible to measure heart rate variability with high accuracy. There are also less precise measurement systems, such as those based on videos. In [89], the author presents a system based on facial images. The author in [90] develops a system based on an infrared CMOS camera to measure heart rate.
Respiratory rate is the number of breaths per minute [91]. This parameter can be measured by an impedance spirometer that measures the variation in body resistance during breathing [92]. It is common to measure the respiratory rate through acoustic systems placed on the neck [93]. However, one of the most accurate systems is based on capnography. It measures the concentration of carbon dioxide in the patient’s airway to determine the respiratory rate [94]. Nevertheless, it is a contact-based system that cannot always be used. There are non-contact systems, such as the one presented in [95], which use a Doppler sensor placed on the ceiling of an intensive care unit. In [96], a system capable of measuring the heart rate by using imaging systems is presented.
Finally, body temperature can be measured using different methods. On the one hand, it is possible to carry out such measurements with contact thermometers. These include temperature sensors combined with predictive algorithms for fast measurement or higher sensitivity. The system presented in [97] is based on continuous temperature measurement in the ear channel and combines the reading with statistical learning algorithms for higher accuracy. In X, the author presents a smart pillow that is able to estimate body temperature using machine-learning algorithms [98]. It is also possible to measure body temperature without contact; there are several developments based on infrared measurement [99,100].
Embedded systems are not only present in monitoring medical devices; in recent years, they have been extended to all health technology categories [101]. They can be found in diagnostic (blood glucose monitors, blood INR monitors, defibrillators and digital thermometers), prognostic (PET, digital X-ray and MRI), patient management (self-test devices for remote patient monitoring) and telemedicine applications [102].
The use of embedded systems in the medical sector has become very interesting, as it provides many important advantages. These systems are considered highly customizable and controllable, as the design of both the software and the hardware is usually tailor-made for each application. Furthermore, complete design of the hardware, firmware and software allows the developer to control the system at all times. The author in [103] presents different design techniques for lightweight, re-configurable medical systems based on embedded systems. Regarding its cost, these are low-cost systems with a dedicated design that makes these systems cost-effective; this feature has opened the door to disposable or widely adopted electronic devices, such as wearable electronics. The author of [104] highlights the use of embedded systems in countries where funds are tight. Finally, due to being highly optimised systems, response times can be minimal, ensuring real-time execution. This feature is key in the medical sector as it minimises the sanitary reaction time or even the time required to dose the treatment. In [105], the importance of real-time systems in insulin pump devices is evidenced.
On the other hand, the aforementioned systems also have disadvantages or drawbacks that need to be carefully monitored when used in the health sector. These systems, especially those with accessible communication ports, can present vulnerabilities as they are susceptible to being hacked [106]. In an industrial or consumer PC-based system, it is possible to install anti-virus applications or firewalls that act as a barrier against attacks. Embedded systems are not immune to such attacks, and firewalls and anti-virus applications cannot usually be integrated into the processing unit. Therefore, it is required for the firmware/software developer and the hardware designer to implement advanced security mechanisms [107]. As it is mentioned in [108], it is considered good practice to implement secure update mechanisms, secure key storage elements, data encryption, etc.
In addition, there are many free hardware and software packages available on the web that enable almost any technician with some knowledge of electronics to implement solutions based on embedded systems. Although this may seem like an advantage, it has become a huge problem for the medical sector. These free hardware and software are not usually designed to satisfy the requirements of the healthcare sector and, therefore, do not comply with the required regulations to commercialise these solutions [109]. Usually, start-ups, due to the lack of specific technical knowledge and unfamiliarity with the medical sector requirements, develop their products based on open-source hardware and software platforms. Eventually, when they attempt to market them, they realise that a complete redesign of the developed device is required.
The development of a new medical device is a complex and resource-intensive process. The complexity of medical device development lies in compliance with the associated regulations. A weak identification of regulations and requirements or a technically optimal design without considering regulatory requirements can result in an unsuccessful new medical device development. Figure 2 presents the phases of a medical product design and development strategy. These stages are detailed in the following [110]:
Feasibility: this phase identifies the market needs, clinical and regulatory aspects of the project development and economic impact.
Design and Development: all the functional requirements of the development are identified in this phase, and the project plan is established. After this definition, the first prototype is developed. It is common to have design iterations during this process.
Design verification: development is verified and, therefore, a functional prototype is obtained; in this stage, it is ensured that the components fulfil the established requirements and safety standards.
Certification and Qualification: in this step, it is necessary to guarantee the compliance of the product with the requirements established by the accredited certification and standardisation organisms. This phase includes (i) clinical trials: depending on the category of the equipment, it is necessary to carry out clinical trials to guarantee compliance with the requirements, (ii) electrical safety tests, electromagnetic compatibility tests, etc., and (iii) achieve the product’s commercialisation acceptance.
Industrialisation: during this phase, the product is transferred to production; for a successful industrialisation, this step must include quality assurance control mechanisms.
Post-market surveillance: once the product is on the market, it is necessary to monitor it in order to identify problems in the field, to know the satisfaction with the use of the device and to publish software updates in response to identified vulnerabilities.
