Some of the core issues also involve the technological aspect of artificial intelligence and its development. However, these issues are generally seen as easier ones to come by than the previous ones as they do not involve changing the attitude of people (Pearl 2018). The key observations of this chapter are presented in table 4.
Table 4. Technological challenges of AI in healthcare and their effects.
Technological challenges Effects
AI is not yet developed enough for exten-sive use in healthcare.
More time is needed for AI and its appli-cations to evolve.
AI is dependent on data and healthcare data is highly regulated.
Data needs to be available and standard-ized for AI to reach its full potential.
Information security of patient data and AI algorithms.
Information security must be properly handled so the algorithms and datasets cannot be breached or altered.
Use of AI in decision making requires significant computing power.
Investments are needed to get the hard-ware capable of supporting the use of AI.
Artificial intelligence of today is still quite basic and far from its full potential (The Med-ical Futurist 2017). The technology is constantly evolving but the current technologMed-ical state of artificial intelligence is one of the limiting factors of its use. Doctors do not see AI applications such as IBM Watson Health ready for proper use yet (KevinMD 2018).
However, artificial intelligence is evolving at a rapid phase (Reichental 2017; The Medi-cal Futurist 2017).
Due to its nature, artificial intelligence is highly dependent on digital data. This is partially problematic as data related to healthcare is highly regulated by legislation which hinders its use in training of artificial intelligence. Currently, there are legal limitations on access to medical data and this restricted access causes a barrier to development of good and reliable health-focused algorithms (PokitDok 2018). To use machine learning in teaching the artificial intelligence, it is necessary to train it with patient health records, healthcare statistics and other personal information related to medicine (Infosys 2018). This causes technical issues as well as ethical ones with the material required. Quality of the data and inconsistencies in its availability restrict the potential of artificial intelligence (Nuffield Council on Bioethics 2018). Using artificial intelligence effectively would also require streamlining and standardizing medical records to a format from which the algorithms can utilize them. (The Medical Futurist 2017; Nuffield Council on Bioethics 2018) Anal-ysis of large and complex data sets also requires significant computing power (Nuffield Council on Bioethics 2018). This means that medical institutions need to make invest-ments to expensive hardware capable of supporting the use artificial intelligence properly.
There is also the problem of data security (Bresnick 2018). What if someone can breach the datasets related to the AI algorithms? Of course, data security is a valid concern even without the use of artificial intelligence, but its widespread use would make this concern even more significant, especially if the algorithm data is stored in a single repository.
Alternating the datasets or even the algorithms via hacking is an issue that should not be understated. New technologies like Blockchain may help with these issues in the future but currently they are not used extensively (Bresnick 2018).
5. DISCUSSION
Artificial intelligence is the technology that people have been talking about for many years and history of artificial intelligence starts from as early as 1940s. Healthcare indus-try is one indusindus-try that utilizes artificial intelligence significantly and we found that arti-ficial intelligence gives the biggest advantages to productivity in healthcare. We defined productivity in healthcare in a way how many patients doctor can treat in her/his working hours.
In this work, we identified three potential artificial intelligence applications to increase doctors’ productivity. These three identified artificial intelligence applications are pattern recognition, voice-to-text transcription and AI-assisted medical diagnosis. Pattern recog-nition can help doctors to examine medical imaging studies and this artificial intelligence tool can give 5-10% more accurate results and much faster examining. Voice-to-text tran-scription enables faster documentation and it can save doctors time up to 17% of their work time.
Artificial intelligence -assisted medical diagnosis helps in more difficult cases and this way it saves doctors time and helps them to give faster diagnosis for patients. These productivity-improving findings are important because they increase doctors’ productiv-ity a lot and that way more people can be treated in less time. With new technologies, there are also lot of thing to be worried about and things that can be challenging for tech-nology to solve. The barriers of artificial intelligence in healthcare industry that we dis-covered in this work are ethical issues, cultural issues and technological challenges. Other potential issues can also be found but these three are overall the most significant ones.
Artificial intelligence can be seen in two ways, it can assist human in healthcare processes or it can replace human in those processes. The biggest barrier in ethical standpoint is that who is responsible if and when artificial intelligence makes a wrong decision. Issue in cultural standpoint is that medical culture values doctors’ intuition so the change re-sistance for utilizing more artificial intelligence in healthcare processes is inevitable. For many people, it is hard to trust something other than doctors when they need medical diagnosis and treatment.
The last main issue for the artificial intelligence in healthcare is the technological aspect.
There are few things to consider from this standpoint. There are inconsistencies in the availability and quality of data restrict the potential of artificial intelligence. It is also needed to discuss that do the medical staff and patients have enough know-how to use artificial intelligence. In technical aspect, it is important to see that this kind of technology needs lot of power from computers because they utilize a lot of data and that can be also the barrier for better use of the technology in this industry.
6. REFERENCES
Accenture (2017)., “Artificial Intelligence: Healthcare’s New Nervous System”, availa-ble: https://www.accenture.com/us-en/insight-artificial-intelligence-healthcare
Anyoha, R. (2017) The History of Artificial Intelligence. Harvard University. Available:
http://sitn.hms.harvard.edu/flash/2017/history-artificial-intelligence/
Bresnick (2018)., “Arguing the Pros and Cons of Artificial Intelligence in Healthcare”, Health IT Analytics, available: https://healthitanalytics.com/news/arguing-the-pros-and-cons-of-artificial-intelligence-in-healthcare
Coeckelbergh, M. (2010) “Health Care, Capabilities, and AI Assistive Technologies”, Ethical Theory and Moral Practice, Volume 13, Issue 2, pp 181–190.
Copeland, B. J. (2019) ”Artificial Intelligence”. Encyclopedia Britannica. Available:
https://www.britannica.com/technology/artificial-intelligence
Foote, K. D. (2016) ”A Brief History of Artificial Intelligence”. Dataversity. Available:
https://www.dataversity.net/brief-history-artificial-intelligence/
Gill-Cox, A., 2018. ”Consumer health applications, machine learning, and systems neu-roscience: The use of artificial intelligence algorithms in clinical medicine and healthcare delivery”. American Journal of Medical Research 5, 46.
Google (2016)., “Deep Learning for Detection of Diabetic Eye Disease”, Google AI Blog, available: https://ai.googleblog.com/2016/11/deep-learning-for-detection-of-dia-betic.html
Google (2017)., “Assisting Pathologists in Detecting Cancer with Deep Learning”, Google AI blog, availa-ble: https://ai.googleblog.com/2017/03/assisting-pathologists-in-detecting.html
Google (2018)., “Improving the Effectiveness of Diabetic Retinopathy Models”, Google AI blog, availa-ble: https://ai.googleblog.com/2018/12/improving-effectiveness-of-dia-betic.html
Hamet, P., Tremblay, J., (2017). ”Artificial intelligence in medicine”. Metabolism 69, S36–S40.
Hand, D.J., (1987). ”Artificial intelligence and medicine: discussion paper”. J R Soc Med 80, 563–565.
IDC (2013)., “The Digital Universe Driving Data Growth in Healthcare”, EMC, availa-ble: https://www.emc.com/analyst-report/digital-universe-healthcare-vertical-report-ar.pdf
Infosys (2018)., “AI for Healthcare: Balancing Efficiency and Ethics”, Infosys, availa-ble: https://www.infosys.com/smart-automation/Documents/ai-healthcare.pdf
Javanmardian & Lingampally (2018)., “Can AI Address Health Care’s Red-Tape Prob-lem?”, Harvard Business Review, available: https://hbr.org/2018/11/can-ai-address-health-cares-red-tape-problem
KevinMD (2018)., “Artificial intelligence in medicine: not ready for prime time”, Kev-inMD, available: https://www.kevinmd.com/blog/2018/08/artificial-intelligence-in-med-icine-not-ready-for-prime-time.html
Lorenzetti (2016)., “Here’s How IBM Watson Health Is Transforming the Health Care Industry”, For-tune, available: http://fortune.com/ibm-watson-health-business-strategy/
Meskó, B., Hetényi, G., Győrffy, Z., (2018). ”Will artificial intelligence solve the human resource crisis in healthcare?” BMC Health Services Research 18.
Noorbakhsh-Sabet, N., Zand, R., Zhang, Y., Abedi, V., (2019). ”Artificial Intelligence Transforms the Future of Healthcare”. The American Journal of Medicine.
Nuffield Council on Bioethics (2018)., “Bioethics Briefing Notes, Artificial intelligence (AI) in healthcare and research”, Nuffield Council on Bioethics, available: http://nuffield-bioethics.org/project/briefing-notes/artificial-intelligence-ai-healthcare-research
Pearl (2018)., “Artificial Intelligence In Healthcare: Separating Reality From Hype”, Forbes, available: https://www.forbes.com/sites/robertpearl/2018/03/13/artificial-intelli-gence-in-healthcare/#4db44ce1d750
Pearson (2017)., “Artificial Intelligence in Radiology: The Game-Changer on Everyone’s Mind”, Radiol-ogy Business, available: https://www.radiologybusiness.com/topics/tech-nology-management/artificial-intelligence-radiology-game-changer-everyones-mind Pennic, F. (2018). “Johns Hopkins Deploys Nuance’s AI-Powered Clinical Documenta-tion to Combat Clinician Burnout”, HIT Consultant, available: https://hitconsult- ant.net/2018/12/21/johns-hopkins-deploys-nuances-ai-powered-clinical-documentation-to-combat-clinician-burnout/#.XJe5eHduKbg
Pirkkalainen, H. (2019)., “Introduction on Information Systems (IS) research lens”, Global Information Systems Management, Course material, Tampere University
Pokitdok (2018)., “The Challenges And Opportunities Of Implementing AI In Healthcare”, Pokitdok, available: https://blog.pokitdok.com/ai-in-healthcare/
Reichental (2017)., “The Evolution Of AI And Robotics Will Be Amazing (And Pain-ful)”, Forbes, available: https://www.forbes.com/sites/forbestechcouncil/2017/12/28/the-evolution-of-ai-and-robotics-will-be-amazing-and-painful/#40480e4d4597
Rick (2011)., “Top 12 reasons why people resist change”, Ameliorate, available:
https://www.torbenrick.eu/blog/change-management/12-reasons-why-people-resist-change/
Rupp (2018), “How AI Can Help Improve Healthcare Efficiency”, Electronic Health Re-porter, available: https://electronichealthreporter.com/how-ai-can-help-improve-healthcare-efficiency/
Schwartz, W.B., Patil, R.S., Szolovits, P., (1987). ”Artificial Intelligence in Medicine”.
The New England Jour-nal of Medicine; Boston 316, 685–688.
Smith, C., McGuire, B., Huang, T. & Yang, G. (2006) The History of Artificial Intelli-gence. University of Washington.
The Medical Futurist (2017)., “Six Challenges To Tackle Before Artificial Intelligence Redesigns Healthcare”, The Medical Futurist, available: https://medicalfuturist.com/six-challenges-to-tackle-before-artificial-intelligence-redesigns-healthcare
Uckun, S., (2018).” AI in Medicine – A Historical Perspective”. RowAnalytics. available:
https://rowanalytics.com/blog-post/ai-in-medicine-a-historical-perspective/
Waugh (2017)., “Which Industries Could Benefit from Virtual Reality?”, The Telegraph, available: https://www.telegraph.co.uk/connect/small-business/tech/pc-world/benefits-of-virtual-reality/
Weintraub (2018)., “Artificial Intelligence Is Infiltrating Medicine -- But Is It Ethical?”, Forbes, available: https://www.forbes.com/sites/arleneweintraub/2018/03/16/artificial-intelligence-is-infiltrating-medicine-but-is-it-ethical/
WHO, (2019). ”WHO/Europe | Health workforce - Data and statistics”. available:
http://www.euro.who.int/en/health-topics/Health-systems/health-workforce/data-and-statistics