Introduction
Modern clinics around the world are equipped with high-quality devices to help patients with diseases of varying severity. Some of the most complex instruments are in critical care departments. This equipment will be attached to the most critical patients who require constant supervision and assistance. Among the high-end equipment in intensive care wards, there is equipment based on artificial intelligence. It helps in the treatment and monitoring of patients much faster and better than medical electronics of the past generation. Despite the difficulties associated with its ethical and psychological issues, artificial intelligence supports medical personnel (both doctors and nurses). The COVID-19 pandemic has demonstrated the capabilities of artificial intelligence in medicine and critical care, allowing the development of telemedicine. Artificial Intelligence in critical care is helping to care for patients faster, supervise more patients, calculate the exact dosage for patients, and collect faster and more detailed patient data.
Quantitative methods were used to cover many sources that meet the latest medical and technical standards. The goal was to analyze them, compare them, and identify similarities. In turn, the authorities are replete with surveys, questionnaires, detailed tables, and numerical data based on actual patient cohorts. All these data have been translated into an explanatory plane and interpreted. Medical workers and doctors obtained consistent, clear conclusions; in some materials, scientists predicted plans for further research.
Results and Discussion
Qualification and Using
Artificial intelligence in medicine and critical care is a very active topic filled with new data from year to year. A variety of closures can be drawn from these data obtained in different patient cohorts, in other age groups, and in different clinical settings. Controlling artificial intelligence devices can seriously harm the fast and accurate well-oiled work of doctors and staff (nurses) who usually help doctors. It follows that “productive interactions between physicians and data scientists are needed to enable clinically meaningful automated and predictive data analysis” (Krittanawong et al., 2017, p. 2660). Doctors have no special education to work with big data, information technology, and artificial intelligence.
An order from the clinic’s management to work only using the most complex equipment will cause stress to medical workers. The work with such a technique must be debugged with the help of engineers, “the complexity of algorithms means that a profound and detailed knowledge is needed to really understand them” (Martin et al., 2021, p. 39). Perhaps this means additional qualifications for a doctor or nurse. Maybe this means that the high-tech sphere will soon become very close to medicine, so much so that each clinic will have a staff of engineers and programmers.
Perhaps the nurses mentioned above might guess those companion robots (or other information systems) can replace them. The problem of loss of confidence and protection in the face of highly advanced technologies was mentioned in the research by Sanchez-Pinto et al. (2020): “Another common concern among clinicians is the perceived loss of autonomy in the face of increasingly more sophisticated computational systems” (p. 1244). However, it should not be forgotten that the doctor is a social profession that combines science. It is often challenging to connect the communicative aspect of medical work and the scientific part. Working with people when it comes to communicating and expressing feelings can be devastating.
Emotions, Communication with Patients
Communication and support are essential for the medical staff since patients constantly need psychological help. Often, relatives, children, and friends of the affected people also need psychological help. They want to hear a qualitative analysis of the condition of their relative who is broken by illness and expect that the doctor or nurse will certainly show sympathy. Robots with a developed social sense and communication mechanisms are among the most burning topics in modern society; this has been going on for a long time. Robert (2019) talks about such robots and summarizes the division of duties between the robot and the nurse: “As robots learn to perform nursing functions, such as ambulation support, vital signs measurement, medication administration, and infectious disease protocols, the role of nurses in care delivery will change” (p. 35). Nurses have a lot of work on computers or papers while doctors are engaged in the direct treatment of patients.
Robots or artificial intelligence would allow nurses to join the treatment of patients. Soon, a robot will not be invented that can provide the same emotional support to a person that another person, a medical professional, will be capable of. It suggests that robots would do well at this stage of technology development by filling in tables, graphs, and charting. Robots could draw user-friendly diagrams that include details of a patient’s treatment while nurses were direct treatment and assistance. “In addition, increased AI usage in medicine not only reduces manual labor and frees up the primary care physician’s time but also increases productivity, precision, and efficacy” (Amisha et el. 2019, p. 2330). Many investigations show that the use of artificial intelligence, including in critical care, will free up doctors’ time and spend it more efficiently.
Decision-Making
It was most often mentioned that artificial intelligence and information technologies allow doctors to work much faster and process information faster. Especially useful in a large patient population setting where the clinic cannot physically cope with the flow of people. But artificial intelligence in such a stressful framework must allow doctors to make the right decisions quickly. Rahmatizadeh et al. (2020) explained this mechanism well, mentioning COVID-19 as a factor of chaos in the clinic: “The efforts of the healthcare system to defeat COVID-19 could be supported by an AI-based decision-making system which would double them up and help manage these patients much more efficiently, especially those in COVID-19” (p. 20). This research, in general, is based on the assertion that the work of doctors with artificial intelligence can be high-speed and of high quality. It can help not only in a pandemic or epidemic but also in emergencies or in situations where a large number of critical patients need help.
With artificial intelligence, doctors would not waste time doubting. Yuan et al. (2020) had a similar opinion after the research: “AI [Artificial Intelligence] can also facilitate clinical decision making” (p. 4). Better decision-making will allow for better delegation of physician and nursing work and increase the number of patients eventually receiving care. “The ability to make decisions under conditions of uncertainty and in compliance with the rules of medical ethics is considered a crucial competency of intensivists” (Beil, 2019, p. 10). Making quick decisions is vital from an ethical point of view and a practical point of view. Investigations can assume that artificial intelligence technologies will also be assistants in the field of experimental medicine.
Experimental medicine is based on the careful and careful behavior of the doctor because any data must be entered with precision. Since experimental medicine is still an area of human care, it is necessary to monitor whether patients have worsened. They are not an experiment on rats, but the fate of seriously ill people. The developed technologies help to record the slightest changes in the patient’s body. Usually, such changes elude the doctor’s eye because he is also an ordinary person with his skill level and training experience.
The COVID-19 pandemic has shown the medical world that it is crucial to create high-quality and quick predictions about how the disease will develop. In addition, forecasts about treatment are needed, but this is true for any disease. In critical care, this is especially important since the patient, while in the critical care unit, is on the verge of ordinary life and death. A good prognosis is essential for both doctors and relatives.
Prediction-Making
A lot of research has been aimed at finding out how artificial intelligence copes with the skill of prediction making. One such study was conducted by Zhang, Ho, and Hong in 2019 and showed that rapid machine analysis of data enables “develop and validate a better-performing predictive model” (p. 9). This ability of artificial intelligence and information technology systems takes a connection with decision-making skills. Artificial intelligence, judging by the research, is faster at making decisions and faster at making a patient’s prognosis and treatment (Meyer et al., 2018). This is because these technologies can easily and quickly combine massive databases, work with streams of information and extract the most important from them.
However, one should not assume that artificial intelligence can only be helpful in chaotic circumstances. In critical care, artificial intelligence can be used every day “to improve health care for patients in its regulatory framework” (Shah et al., 2019, p. 2). As stated many times, critical care works similarly with different patients and in various clinical settings. The goals of urgent care and its desired results do not change, with which patients the doctors would not work.
Calculating of Doses
The exact calculation of the dosage of drugs is fundamental in critical care since the patient cannot sign that the dosage was not enough or too much. As a rule, doctors find out about the dosage error after the patient’s death. Komorowski et al.’s (2018) investigations show that using the example of vasopressors; it was possible with the help of artificial intelligence to correctly calculate the doses (pp. 1719-1720). The patients, divided into cohorts, received doses of vasopressor, which were calculated by the artificial intelligence and the doctor, an ordinary person. Artificial intelligence was less likely to make mistakes when calculating amounts; therefore, patients in critical condition did not die; patients in stable condition did not worsen their health. When conducting further research with drugs, it is crucial to perform laboratory tests (Labovitz et al., 2017). It is vital that the analysis of artificial intelligence in medicine does not stop but is combined with others.
Ethic
The ethical issue can be a big problem for the use of artificial intelligence throughout medicine. This is evidenced by research by Nguyen, Ngo, and van Sonnenberg (2020):
- Physicians can explain their thought processes to identify potential areas for mistakes, but AI’s thinking is hidden from human view, although there is current research aimed at graphically clarifying AI’s decision matrix. Many AI models are trained on data that do not include relevant social factors, such as ethnicity or socioeconomic status; this absence potentially can lead to neglect of minorities. (p. 6)
- As the researchers can see, artificial intelligence is not yet decoded enough to make its languages understandable to all patients. You need to understand that people receive medical care regardless of their economic status and education. Failure to understand the language of artificial intelligence and machines creates a lot of mistrust, which can undermine the overall medical reputation.
In addition, the thinking of the machine is very rigid, unable to adapt to the rapidly changing situation in critical care. Futoma et al. (2020) described artificial intelligence as something “nor are they like trained clinicians, gracefully adapting to new circumstances.” (p. 491). The doctor, trusting in his experience, gains natural flexibility over time in difficult situations, for example, intubation. Often, patients trust experienced doctors rather than young educated graduates of elite medical academies. Medicine is an applied science, so the issue of experience and the ability to adapt to unexpected situations is very acute.
Conclusion
Despite the controversial ethical issues (trust in machine technology, the experience of doctors, and a willingness to quickly change circumstances), artificial intelligence and information technologies in critical care allow doctors to accurately calculate the dosage of drugs, collect large amounts of information about patients, predict treatment, and make quick decisions. In the situation of the COVID-19 pandemic, artificial intelligence has shown that it can work in critical care with large patient flows without confusing information and making decisions quickly. However, the widespread use of artificial intelligence and its technologies means additional qualifications for doctors and nurses.
Recommendations
One of the main recommendations may be to invite specialists in engineering and artificial intelligence to clinics equipped with complex machines. It is also necessary to constantly adjust the base of higher medical education, which can now focus on teaching young students the basics of working with complex artificial intelligence mechanisms. Among the population and patients, it is necessary to increase the level of trust and tolerance to the agencies of artificial intelligence, since the results to which machines can come cannot always be empathically presented and explained to seriously ill patients in intensive care, as well as to their relatives and friends.
Summary
Many modern doctors are faced with the problem of inappropriate qualifications to work with complex artificial intelligence technologies. Some are psychologically stressed and anxious that companion robots targeting the communicative side of human life will replace them in the workplace. This problem is often associated with nurses and their core functions. Information technology and artificial intelligence in intensive care can make the work of doctors and nurses easier, as they free up time to work with patients by taking on paperwork. Artificial intelligence in complex situations in intensive care can take on some of the burden or burden of decision-making since it has the power to cover large amounts of data. Based on this data, decisions will be made by the mechanisms.
Artificial intelligence shows promising results in predicting the course of the disease and the effect of treatment. It has helped in critical care during COVID-19 and in the field of experimental medicine. Artificial intelligence mechanisms demonstrate a high level of accuracy in dosage calculations. Patients in intensive care in the artificial intelligence group received a more accurately calculated drug. Their health subsequently improved, and, in general, there were fewer deaths. The issue of ethics is still relevant in the field of using artificial intelligence in critical care. Artificial intelligence data cannot be decrypted for all patients, which contributes to a low level of trust. In addition, artificial intelligence cannot always take into account the socioeconomic factors of patients and present information empathically.
References
Amisha, P. M., Pathania, M., & Rathaur, V. K. (2019). Overview of artificial intelligence in medicine. Journal of Family Medicine and Primary Care, 8(7), 23-28.
Beil, M., Proft, I., van Heerden, D., Sviri, S., & van Heerden, P. V. (2019). Ethical considerations about artificial intelligence for prognostication in intensive care. Intensive Care Medicine Experimental, 7(70).
Futoma, J., Simons, M., Panch, T., Doshi-Velez, F., & Celi, L. A. (2020). The myth of generalisability in clinical research and machine learning in health care. The Lancet Digital Health, 2(9), 489-492.
Komorowski, M., Celi, L. A., Badawi, O., Gordon, A. C., & Faisal, A. A. (2018). The artificial intelligence clinician learns optimal treatment strategies for sepsis in intensive care. Nature Medicine, 24(11), 1716-1720.
Krittanawong, C., Zhang, H., Wang, Z., Aydar, M., & Kitai, T. (2017). Artificial intelligence in precision cardiovascular medicine. Journal of the American College of Cardiology, 69(21), 2657-2664.
Labovitz, D. L., Shafner, L., Reyes Gil, M., Virmani, D., & Hanina, A. (2017). Using artificial intelligence to reduce the risk of nonadherence in patients on anticoagulation therapy. Stroke, 48(5), 1416-1419.
Martin, L., Peine A., Marx G., et al. (2021). Rethinking Critical Care – Use and Challenges of Artificial Intelligence. HealthManagement, 21(1), 38-40.
Meyer, A., Zverinski, D., Pfahringer, B., Kempfert, J., Kuehne, T., Sündermann, S. H., Stamm, C., Hofmann, T., Falk, V., & Eickhoff, C. (2018). Machine learning for real-time prediction of complications in critical care: A retrospective study. The Lancet Respiratory Medicine, 6(12), 905–914.
Nguyen, D., Ngo, B., & van Sonnenberg, E. (2020). AI in the intensive care unit: Up-to-date review. Journal of Intensive Care Medicine, 36(10), 1115–1123.
Sanchez-Pinto, L. N., Luo, Y., & Churpek, M. M. (2018). Big data and data science in critical care. Chest, 154(5), 1239-1248.
Shah, P., Kendall, F., Khozin, S., Goosen, R., Hu, J., Laramie, J.,… & Schork, N. (2019). Artificial intelligence and machine learning in clinical development: a translational perspective. NPJ Digital Medicine, 2(1), 1-5.
Rahmatizadeh, S., Valizadeh-Haghi S., & Dabbagh, A. (2020). The role of artificial intelligence in management of critical COVID-19 patients. Journal of Cellular & Molecular Anesthesia, 5(1), 16-22.
Robert, N. (2019). How artificial intelligence is changing nursing. Nursing Management, 50(9), 30–39.
Yuan, K.-C., Tsai, L.-W., Lee, K.-H., Cheng, Y.-W., Hsu, S.-C., Lo, Y.-S., Chen, R.-J. (2020). The development an artificial intelligence algorithm for early sepsis diagnosis in the intensive care unit. International Journal of Medical Informatics, 141(104176).
Zhang, Z., Ho, K. M., & Hong, Y. (2019). Machine learning for the prediction of volume responsiveness in patients with oliguric acute kidney injury in critical care. Critical Care, 23(1), 1-10.