Internet of Things, Digital Biomarker, and Artificial Intelligence in Spine: Current and Future Perspectives

Neurospine Neurospine 2019;16(4):705-711. https://doi.org/10.14245/ns.1938388.194 Review Article Corresponding Author In Ho Han https://orcid.org/0000-0001-7193-6533 Department of Neurosurgery, School of Medicine, Pusan National University Hospital, 179 Gudeok-ro Seo-gu, Busan 49241, Korea Tel: +82-51-240-7257 Fax: +82-51-244-0282 E-mail : Received: November 17, 2019 Revised: December 3, 2019 Accepted: December 5, 2019 This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. Copyright © 2019 by the Korean Spinal Neurosurgery Society pISSN 2586-6583 eISSN 2586-6591 Internet of Things, Digital Biomarker, and Artificial Intelligence in Spine: Current and Future Perspectives Kyoung Hyup Nam, Dong Hwan Kim, Byung Kwan Choi, In Ho Han Department of Neurosurgery, Medical Research Institute, Pusan National University Hospital, Busan, Korea Recent interest in medical artificial intelligence (AI) has increased with onset of the fourth industrial revolution. Real-time monitoring of patients is an important research area of medical AI. The medical AI is very closely related to the Internet of Things (IoT), a core element of the fourth industrial revolution. Attempts to diagnose and treat patients using IoT have been already applied to patients with chronic disease such as hypertension and arrhythmia. However, in the spine, research on IoT and digital biomarkers are still in the early stages. The digital biomarker obtained by IoT devices is objective and could represent real-time, real-world, and abundant data. Based on its characteristics, IoT and digital biomarkers can also be useful in the spine. Currently, research on real-time monitoring of physical activity or spinal posture is ongoing. Therefore, the authors introduce the basic concepts of IoT and digital biomarkers, their relationship to AI, and recent trends. Current and future perspectives of IoT and digital biomarker in spine are also discussed. In the future, it is expected that IoT, digital biomarkers, and AI will lead to a paradigm shift in the diagnosis and treatment of spinal diseases. Keywords: Artificial intelligence, Internet of Things, Digital biomarker, Wearable device, Spine INTRODUCTION of spinal structures, classification of disc degeneration, measurement of various spinal curve parameters, and diagnosis of fractures and spinal tumors.5,12 Research into AI to predict and prevent disease by monitoring biomedical signals in real time is particularly active in internal medicine.13 In particular, devices to monitor blood glucose or electrocardiogram in real time have been already commercialized.13,14 Despite such high interest in real-time monitoring of patients, AI and real-time monitoring in spine, are still in early stages of research.5 AI related to realtime monitoring of patients is closely related to IoT and digital biomarkers.15 Here, authors introduce the basic concept of IoT, digital biomarker, and their relationships with AI. In addition, current and future perspectives of IoT and digital biomarker in spine are discussed. Recent interest in the fourth industrial revolution, represented by artificial intelligence (AI), big data, the Internet of Things (IoT), and robotics, is higher than ever.1,2 In particular, studies are increasingly exploring the medical application of AI, which is the key element of the fourth industrial revolution.3-5 Medical AI can be used in 3 different ways: First, AI is used to diagnose the disease or predict the treatment outcome with clinical data or genomic data.6-8 Second, AI automatically reads medical images and diagnoses diseases instead of doctors, and third, AI can be used to prevent or predict disease by monitoring biomedical signals in real time.9,10 In the field of spine, AI has been used to predict the outcome after surgery or survival of patients with spinal metastasis.6,11 The focus of AI in medical imaging has been on segmentation  www.e-neurospine.org 705 Nam KH, et al. CONCEPTS AND RELATIONSHIPS OF INTERNET OF THINGS, DIGITAL BIOMARKER, AND ARTIFICIAL INTELLIGENCE The AI and machine learning have been actively studied even in the spine recently, and a few spine surgeons understand the basic concept of AI.5 Moreover, most of spine surgeons are not familiar with IoT and digital biomarker. IoT is the network of physical objects that contain embedded technology to communicate and sense or interact with their internal states or the external environment.16 The “thing” in IoT can be anything in the world, not just mobile or wearable devices. Thus, IoT has recently been referred to as the Internet of Everything.17 Previously, IoT and AI were defined separately. In recent years, IoT has evolved as a technology including AI. IoT is generally composed of devices with embedded sensors, gateway, cloud, analytics, and user interface.18,19 Devices with embedded sensor continuously collect information from the environment and deliver it to the next levels via wireless network. Low-power networks such as Wi-Fi, Bluetooth, and Long Range are commonly used.20 Recently, 5G network has been regarded as an optimized technology of IoT with ultralow delay and ultraconnectivity.21 Gateway plays a key role of receiving information from the sensors of various devices and delivering it to a cloud. Cloud provides a tool to collect, analyze, and store an abundance of data produced from devices in real time.19 In analytics, AI derives meaningful patterns in unstructured data and provides feedback to the user interface. The user interface facilitates visualization of feedback for patients or medical staff. The concept of IoT system is very similar to that of human nervous system. Various sensory organs such as skin, eyes, ears, nose, and organs of equilibrium correspond to digital devices equipped with sensors. The spinal cord and nerves correspond to the wireless network, and the human brain corresponds to cloud and AI. Actuator or feedback service is similar to response of human body such as muscle action to sensory information (Fig. 1). The actuator coupled to the IoT system becomes a moving robot. Robot is a complex of various IoT systems.22 In terms of IoT, the digital biomarker represents digitized data acquired from patients via IoT devices. The biomarker is defined as an objective parameter and indicator of normal biological, and pathological processes, or represents pharmacological response to therapeutic intervention.23 Therefore, the digital biomarker can be defined as a biomarker that is objectively and quantitatively measured using digital devices and be used 706 www.e-neurospine.org Internet of Things and Spine IoT Human Nervous Systme Fig. 1. The concept of Internet of Things (IoT) system is very similar to tha (...truncated)