key: cord-0045530-7v5t241z authors: Kose, Utku; Deperlioglu, Omer; Alzubi, Jafar; Patrut, Bogdan title: Future of Medical Decision Support Systems date: 2020-06-18 journal: Deep Learning for Medical Decision Support Systems DOI: 10.1007/978-981-15-6325-6_10 sha: 9ed28bdee4c37a1bb67f8ae1c185ab78d3f8764c doc_id: 45530 cord_uid: 7v5t241z After having an introduction to the essential topics, the previous chapters have all provided effective use of deep learning for diagnosis of important diseases, as they are base for the medical decision support systems. There are of course many more research aspects to be discussed but if is also a good approach to focus on some insights regarding future of medical decision support systems. .1 Some of the foremost technologies as well as topics that can be considered as the components for the future scenarios day. As computer as well as communication technologies such as Internet, wireless communication ensure critical roles in storing the information in the context of a digital world, the technological developments caused the IoT to rise as a great solution for an autonomous future with smart devices surrounding us to make everything easier and more practical (Of course there are many issues appearing within use of every technology, but that discussion regarding the IoT is another point of interest, as out of scope of this chapter/book). Briefly, IoT allows communicating among all devices that can take part in a network so data regarding people, environment, the other devices can be used accordingly for getting decision makings and performing some actions such as solving a tasks, adjusting the environmental factors, analyzing something or at least ensuring interaction with the people in order to inform them about the world around them. Here, advantages of IoT systems are indicated in Fig. 10 .2. All the mentioned advantages and the communication-oriented mechanisms of IoT are all because of innovative developments in artificial intelligence and the communication solutions such as wireless sensors, wireless communication standards, and also mobile technologies and communication approaches [5] [6] [7] [8] . Nowadays, it is remarkable that IoT have already been widely used in different areas [9] [10] [11] [12] [13] . As that technology is more employed within a specific field, it is also re-called with new names, which are appropriate to the scope of the related field. Internet of Health Things (IoHT) is among them. IoHT is briefly a type of IoT that is applied for medical applications [14, 15] . Because the future will be probably with full of autonomous devices, use of IoT as well as IoHT will be probably a common thing because the field of medical will be always at the first places to benefit from innovative technologies. In accordance to that, the future of medical decision support systems will include intense use of IoHT systems. In detail, possible scenarios will be like as follows: • After getting up in the morning smart mirrors and cameras in our houses will support us to be ready to the day and they will also track for any mood changes or possible illness. • Toilets will be devices analyzing urine and stool to make diagnosis of diseases and/or early diagnosis of experiencing bad-way life standards. • All our medical data will be kept in secure over blockchain running encrypted cloud so that all smart devices will be decision making about our health state. • Our cars will be tracking our health state and possible tiredness symptoms. • While working at office environment, smart devices will be tracking our performance as well as mental and health state against any disease or lowering in our well-being. • Surgeries will be made by decision-making smart robotic systems accurately and in a faster manner. • All treatments will be tracked by smart devices around us so that we will be faster recovering. • Thanks to early actions by smart devices, people will not easily be infected or at least be preventing themselves from possible diseases. • Smart devices will support us to have healthy food and track our medical data for a healthy aging in time. Since it is still possible to imagine more and more about the possible future scenarios, IoHT here ensure a critical role for supporting us for having good health and well-being generally. As IoHT systems of the future will be associated with deep learning (any maybe more advanced forms with new names), the analyzing, diagnosing, and treatment processes will be even faster and more effective according to today. Covering all the explanations so far, Fig. 10 .3 provides essential benefits to be provided by the future IoHT systems. As associated with the IoHT perspective, the future of medical decision support systems will be also with wearable technologies. Currently, there are many different types of wearable technologies ready to be used ( Fig. 10 .4 [14] ). Wearable technologies can effectively track our data and ensure smart features to support us for an easier life and even decision making generally. Considering the medical, wearable technologies will be probably common components of IoHT upper-systems and will be essential tools for understanding more about us and people for ensuring general well-being. When a discussion on artificial intelligence and the future is made, the robotics technologies are certain topic that is widely explained. As appropriate to that, the future of the medical decision support systems will be intensively associated with more use of robotics. Even nowadays, there are many examples of robotics usage in different fields and it is already a constant component of the future [16] [17] [18] . In the context of medical problems and decision support tasks, the following scenarios can be thought accordingly: [14] • In the future, there will be kiosk-like robotic systems serving public spaces for helping people for simple medical diagnosis and treatments. • As surgeries are already supported by hard and soft robotics [19, 20] , the future will be including wider use of robotics during surgery. Such robotic systems will not only be performing-supporting the surgery but will also help doctors in decision making processes. • Artificial intelligence is already used in physical rehabilitation applications [21] [22] [23] . In the future, there will be more rehabilitation robots and at homes, people (especially older people) will be supported by personal assistant robots with medical knowledge. • Many of simple medical tests (i.e. taking blood sample) at hospitals will be made by service robotics. • There will be advanced diagnosis robotics-based rooms for performing general check-up processes. • Medical tasks that are generally dangerous to be done by humans will be done by robots. • There will be more use of soft robotics because of their advantages according to hard robotics. • There will be more robotic expert systems performing answer-question related interactions with people, in order to get ideas about health state, possible disorders/diseases, and collection pre-information for supporting doctors in decision making. • Robotic systems will be probably even in smaller and as integrated to wearable technologies for better tracking purposes. As it was mentioned and emphasized in the previous chapters, information discovery is among critical solution ways of the artificial intelligence. Thanks to deep learning and big data use, information discovery has gained more momentum in time. Information discovery can be used to derive new combination of information, wider knowledge and new patterns resulting to the exact mechanism-role of the discovery [24, 25] . Diseases has always been problem for the humankind. Currently, the humankind is even experiencing a pandemic caused by the coronavirus COVID-19 and because of that almost all research works of medical has directed to finding effective treatment of COVID-19. That situation and past experiences with different diseases, which were strong then and weak (or eliminated) nowadays indicate the importance of using information discovery for the drug discovery done with the employment of artificial intelligence in the field of medical. Moving from that, the future will intensively include running drug discovery (discovering medicine-vaccine as well as treatment strategies) by employing effective algorithms-techniques and running them in the context of advanced decision support systems. For even nowadays, there are many examples of drug discovery studies done with deep learning [26] [27] [28] [29] [30] [31] so that future diseases, viral infections, disorders and new type of micro-organisms will be often subject to drug discovery studies. Here drug discovery can be an effective step of a whole medical health management system, including use of intelligent systems for analyzes, diagnosis, drug discovery and then treatment processes in the context of a decision support flow, as illustrated in Fig. 10 .5. Use of automated medical diagnosis as the component of medical decision support systems has already been a good weapon of the humankind for all kinds of diseases. As indicated under the previous paragraph, the future of medical will be still including dealing with infections, micro-organisms and maybe today's vital disease: cancer will be still analyzed, diagnosed and then treated with the use of intelligent systems running medical support systems. Currently, there are already research works done for effective diagnosis of different cancer types [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] . With especially good collaboration with image processing, deep learning architectures have been effectively using against cancer diagnosis (Fig. 10.6 ). On the other hand, there is also rare diseases including i.e. viral, bacterial infections (when they are not pandemic yet), specific allergies, and genetical disorders [43] [44] [45] . As some of technological tools and changes in life standards are also affecting directly or indirectly human metabolism and causing rare diseases, there will be still need for medical decision support systems to deal with rare diseases, too. Since it is also possible to ensure a balanced, healthy life thanks to i.e. IoHT systems as explained before, there is still possibility of rare diseases or cancers as there is the chaos in the universe and the nature. However, future intelligent systems will be key factor to deal with such issues, as a remarkable insight for the future. As indicated before, the humankind is currently (at the time of writing that section of this chapter; March 2020) dealing with the coronavirus type: COVID-19 and it became a pandemic in a short time. Because of the COVID-19, governments around the world has applied remarkable policies including breaks at works, schools, universities with remote-online working conditions at homes, quarantines for preventing people from getting COVID-19, which causes deaths in remarkably short times. As general, there is a great emergency state, which is something like the life around the world stopped. That situation has shown researchers to effective use of technology for early diagnosis of such viral infections before they become pandemics or at least running effective treatments and discoveries (i.e. vaccine) rapidly for eliminating that devastating disease. Nowadays, there are already research works focusing on use of artificial intelligence/deep learning for diagnosis of COVID-19 and deriving alternative treatments for it [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] . Since that problem has taught many things to the humankind, it can be clearly expressed that the future of medical decision support systems will include pandemics control approaches with deeper analyze and tracking of the data around the world. At this point, a possible system scenario can be expressed briefly as follows (Fig. 10.7) : • The key point in the system is using as much as remote communication possibilities in order to keep near contact of humans against any viral infection appearance. In order to ensure that the IoHT system will be common components of the COVID-19 and pandemic control system. • IoHT based devices will be connected world-wide over cloud systems and the data security will be ensured with effective technologies such as blockchain [57, 58] or tangle [59, 60] . World-wide communication of IoHT based devices will be organized with efficient, optimized approaches and models based on globally accepted standards. • IoHT devices will include especially following groups of devices: (1) patient tracking and treatment devices at hospitals, (2) public health support devices spread around cities, roads, restaurants, schools, and similar public places, (3) personal health tracking devices at homes (these may be revised according to needs). Roles of these devices will include especially followings: -(1) Patient tracking and treatment devices will be used at hospitals for as far as possible interaction with the patients and tracking states of the patients with viral-bacterial diseases remotely. That will also allow doctors, medical staff to track everything in even remote mode and using i.e. e-signatures to give orders or perform any tasks that can be done remotely. Eventually, near interactions of people will be as low as possible except from cases such as surgeries, emergency actions-diagnosis. -(2) Public health support devices will support people to keep themselves clean and in safe against diseases. Also, these devices will give people advices, announce emergency states, and ensuring trainings for keeping awareness at a desired level. These devices will also in contact with cleaning services and any other sustainability-green environment technologies in the context of IoT/IoHT. -(3) Personal health tracking devices will be responsible to check all people's health state in the same home environment and direct them or ensure communication with hospitals/doctors remotely in case of any suggestions, regular checks, or emergency state predictions. • As the whole three groups of devices will be in connect with each other, it will be possible to control a virus infection case, by i.e. keeping infected people at home and acting to treat them in accordance to the quarantine rules, perform remote checks-controls at hospitals, improving level of tasks to be done by IoHT devices in public spaces, and many more tasks to do collaboratively by IoHT devices running over artificial intelligence, sensor technologies, mobile communication channels, secure algorithms, and additional technologies to achieve a global, accurate smart system. • Public spaces will be supported with intense use of sensors, cameras, and drones for achieving better IoT-oriented communication as well as tracking actions for people-public health. • There will be world-wide software environment allowing public tracking of worldwide viral infections, and also detailed tracking for authorized users such as doctors, policy makers, governments, as having changing authorization levels. • There will be world-wide up-to-date agreements such as medical data regarding viral-bacterial diseases/infections will be shared instantly around the globe, by securing patient personal data, and the countries will be responsible to share their data of such diseases/infections with an upper-commission or i.e. World Health Organization. • There will be of course specially designed diagnosis, treatment technologies and separate smart hospitals, quarantine houses-hotels in case of any pandemic. • There will be artificial intelligence/deep learning based accurate-fast diagnosis systems, intelligent vaccine development kits, and may be robotics-based services for remote treatment of patients with virus. Analyzes, predictions, control of smart-intelligent systems will be all supported with Data Science and artificial intelligence solutions. • Thanks to the synergy ensured among IoHT devices world-wide, it will be possible for every people to track spreading of any pandemic instantly, like it is an online streaming video or instantly changing data such as economic time series. As that can be done with mobile applications, such applications will also ensure effective communications, announcements world-wide and country-specific. • All the explained working mechanism of the scenario may be revised-updated with addition of alternative technologies, and more use of detailed tasks in control of pandemics such as COVID-19. As it can be understood, use of intelligent systems as well as data controltracking solutions (use of Data Science generally) will give important advantages to the humankind for fighting against pandemics like COVID-19. It is clear that the more a medical decision support system is structured over combinations of solution techniques, the more it could be robust and sustainable for vital medical tasks like predicting, tracking, controlling, treating pandemics. The future of medical decision support systems is all wide open for further research and innovative developments. As a result of improvements and rapid developments in different technologies, the outcomes have always been effective on practical technological solutions for daily-life. After the twenty-first century, that state has been experiencing widely in all different fields. As a critical field, medical will always keep its top place for the newest solutions. Here, solutions by intelligent systems will be key triggering factor for both solving medical problems as well as taking it steps away for better well-being of the humankind. As also a final touch to this book, this chapter discussed some specific subjects, which the authors think will be important for the future of medical decision support systems. The explanations including IoT, IoHT, wearable technologies, robotics, drug discovery, diagnosis of cancers as well as rare diseases are important topics for further research. Of course, as the current threat for the existence of the humankind, COVID-19 and pandemics have been also discussed in the sense of their control and treatment, as the final, vital topic. As concluding comments for the book, the authors have provided a current view on use of deep learning for rising medical decision support systems. Employment of different deep learning architectures, using them for especially diagnosis solutions so forming essentials of medical decision support systems are rapidly improving research topics so that there will be always need for such reference book for better understanding the latest research and having future ideas. The authors would like to thank all readers for reading that book and desire to see same interests for the future works by them. In order to have some more future insights from very recent works regarding future of artificial intelligence as well as its role in medical, readers can read [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] . For learning more about smart medical applications (with especially support by IoT, mobile communication, and some other technologies) as well as some future perspectives, the readers are referred to [77] [78] [79] [80] [81] . Future internet: the internet of things The internet of things: a survey Designing the Internet of Things Internet of things: challenges and opportunities Learning Internet of Things Internet of things in industries: a survey A survey on energy conserving mechanisms for the internet of things: wireless networking aspects Internet-of-things-based smart environments: state of the art, taxonomy, and open research challenges Internet of things: vision, application areas and research challenges The internet of things in agriculture for sustainable rural development Exploring the internet of "educational things" (IoET) in rural underprivileged areas Internet of things technologies in manufacturing: application areas, challenges and outlook Smart electricity distribution in residential areas: Internet of Things (IoT) based advanced metering infrastructure and cloud analytics Enabling technologies for the internet of health things Design research opportunities in the internet of health things: a review of reviews Springer Handbook of Robotics Robotics and Automation Handbook The future digital work force: robotic process automation (RPA) 30 years of robotic surgery A hybrid soft robotic surgical gripper system for delicate nerve manipulation in digital nerve repair surgery Physical fatigue detection through EMG wearables and subjective user reports: a machine learning approach towards adaptive rehabilitation Control strategies and artificial intelligence in rehabilitation robotics Interactive hybrid approach to combine machine and human intelligence for personalized rehabilitation assessment Amalthaea information discovery and filtering using a multiagent evolving ecosystem Probabilistic author-topic models for information discovery From machine learning to deep learning: progress in machine intelligence for rational drug discovery Comparison of deep learning with multiple machine learning methods and metrics using diverse drug discovery data sets Deep learning for drug design: an artificial intelligence paradigm for drug discovery in the big data era Deep Learning for the Life Sciences: Applying Deep Learning to Genomics, Microscopy, Drug Discovery, and More Deep learning-driven research for drug discovery: tackling malaria Big data and artificial intelligence modeling for drug discovery Using deep learning to enhance cancer diagnosis and classification Computer aided lung cancer diagnosis with deep learning algorithms Prostate cancer diagnosis using deep learning with 3D multiparametric MRI Enhancing deep convolutional neural network scheme for breast cancer diagnosis with unlabeled data Breast cancer multi-classification from histopathological images with structured deep learning model Realtime colorectal cancer diagnosis using PR-OCT with deep learning Skin cancer diagnosis based on optimized convolutional neural network Deep learning-based histopathological image analysis for automated detection and staging of melanoma Application of deep learning to the diagnosis of cervical lymph node metastasis from thyroid cancer with CT: external validation and clinical utility for resident training Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study Using smartphone and deep learning technology to help diagnose skin cancer Rare and Uncommon Diseases Rare Diseases and Orphan Products: Accelerating Research and Development Preparing for the future of rare diseases Serial quantitative chest CT assessment of COVID-19: deep-learning approach Covid-19: automatic detection from X-Ray images utilizing transfer learning with convolutional neural networks Lung Infection quantification of COVID-19 in CT images with deep learning Artificial intelligence distinguishes COVID-19 from community acquired pneumonia on chest CT A deep learning algorithm using CT images to screen for Corona Virus Disease COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest radiography images Multiple-input deep convolutional neural network model for COVID-19 forecasting in China Deep learning system to screen coronavirus disease 2019 pneumonia Automatic detection of coronavirus disease (COVID-19) using X-ray images and deep convolutional neural networks COVID-19 coronavirus vaccine design using reverse vaccinology and machine learning Potential COVID-2019 3C-like protease inhibitors designed using generative deep learning approaches Blockchain challenges and opportunities: a survey Enabling distributed intelligence in the internet of things using the IOTA tangle architecture IOTA-next generation block chain An intelligent future for medical imaging: a market outlook on artificial intelligence for medical imaging Artificial intelligence and the future art of medicine Artificial intelligence: the future of urinary stone management? What the near future of artificial intelligence could be Artificial Intelligence in medicine: today and tomorrow Future entrepreneurship in digital health Proactive preventive and evidence-based artificial intelligene models: future healthcare Artificial intelligence and robotic surgery: current perspective and future directions Artificial intelligence in medical devices and clinical decision support systems Artificial intelligence in dentistry: current applications and future perspectives Wearables, artificial intelligence, and the future of healthcare, in AI and Big Data's Potential for Disruptive Innovation The role of artificial intelligence in digital health Artificial intelligence and the future of psychiatry: insights from a global physician survey Robotics, artificial intelligence, and the evolving nature of work The ethical adoption of artificial intelligence in radiology The ethics of medical ai and the physician-patient relationship 5G-based smart healthcare network: architecture, taxonomy, challenges and future research directions Healthfog: an ensemble deep learning based smart healthcare system for automatic diagnosis of heart diseases in integrated IoT and fog computing environments Healthcare electronics-A step closer to future smart cities Healthcare systems: an overview of the most important aspects of current and future m-health applications Wearables, artificial intelligence, and the future of healthcare, in AI and Big Data's Potential for Disruptive Innovation