Artificial intelligence in Critical care
The chosen healthcare technology is the use of artificial intelligence (AI) in the intensive care unit (ICU). By using AI, it is possible to make disease predictions as well as phenotypes. Gutierrez’s (2020) article has been chosen to support the project. This is because it primarily discusses how AI can be implemented in ICU. If an old man faints and is taken to hospital and the test cannot show the cause, AI can be used to diagnose a disease like cancer accurately.
Explanation and background
According to Gutierrez (2020), AI is simulated human intelligence by machines. In the ICU, it has been used in various ways, including helping to recognize hidden diseases that doctors may need help identifying, especially when the dataset is heterogeneous or noise prone. For AI to work, the system needs to be fed with alot of patient data to learn from past diagnoses and therefore be able to come up with diagnoses that are more accurate and timely than a doctor’s. The AI also develops the right treatment plans that physicians can follow (Komorowski et al., 2018).
According to Meyer et al. (2018), AI can be integrated to work on all patients who step into the ICU, mainly those with diseases that cannot easily be diagnosed. It shortens the diagnosis time and gives precise results together with the right treatment plan. Matheny et al. (2020) argue that despite AI bringing positive changes in medicine, it is simply a system vulnerable to hacking like any other. If no proper security measures are followed, it can lead to the leakage of patient data and corruption of its intelligence, leading to erroneous results that can endanger patient lives. There is also the risk of racial bias; this depends on who created the software and the data it has been fed. AI is also very expensive and may not be afforded by everyone, especially those without insurance.
Risks and Benefits
Patients have a lot to benefit from using AI in the ICU. Gutierrez (2020) has analyzed several benefits, ranging from predicting the ICU length of ICU stay, ICU mortality, and chances of developing complications; AI also improves ventilation among patients who cannot breathe independently. Medical Information Mart for Intensive Care III (MIMIC-III) is an open-source database that can tell how long a patient will spend in the ICU and if there is a chance for the patient to be readmitted. Predicting whether a trauma or pediatric patient admitted to the ICU will survive is possible using decision trees and machine learning algorithms. For this to happen, the AI must interpret a highly complex model using many variables (Thorsen-Meyer et al., 2020).
Matheny et al. (2020) have stated that something extra must be done to promote positive outcomes with AI. It is not just adding more data into the system. A single pixel can make an AI think a bacteria is a virus and can confidently give a wrong diagnosis through such errors. Therefore to improve AI, it needs to be connected with human and scientific inputs. AI needs to be integrated with scientific laws for scientific inputs, and then the data can be augmented with expert human insights.
In addition, it is important to have devices that can show precisely how the AI arrives at its conclusions and have an expert human confirm this before the AI results are applied. This may seem like an additional waste of time; however, it reduces the risks significantly. Again besides AI, other models can be used to make the diagnosis, and therefore, results can then be compared. This way chance of complications can be highly reduced (Challen et al., 2018).
Interdisciplinary team’s Roles and Responsibilities
According to Komorowski et al. (2018), the interdisciplinary team members in the ICU include physicians, dieticians, therapists, clinical pharmacists, and clinical psychologists. The physicians diagnose using AI-equipped machines to determine a treatment plan and quickly and easily understand the patient’s ailment. The physicians also review the appropriate therapeutic intervention required for patient care and make the right decision.
Dietitians are responsible for informing the team on the timing of when to introduce food to patients. They also monitor the patient macro and micronutrient requirements and the fluids and electrolytes (Meyer et al., 2018). Therapists available depend on the patient’s condition. For instance, if a patient cannot breathe independently, the respiratory therapist monitors the breathing with assistive machine support. The respiratory AI helps the therapist determine when the patient is breathing properly. Pharmacists ensure that the right medicine is given at the right doses. The psychologist helps the patient cope by reducing any distress and making the patient feel calm and relaxed.
The nurse plays a critical role in the ICU, especially where AI is used. The AI can process data by reducing the nurse’s work, but then the nurse uses the knowledge to confirm the info and carry out the instructions. The nurse is the main person who evaluates the patient’s condition and administers treatment. When AI is involved, it can inform the nurse of the exact and appropriate time to administer medications and doses. When all the other personnel leave the ICU, the nurses are left with the patient and continue monitoring the changes (Gutierrez, 2020).
According to Meyer et al. (2018), Interdisciplinary teams are known to work fast and effectively because all resources are merged into one body. However, there exist challenges, especially in the ICU. All specialists tend to rely too much on the nurses to the point that they may need help to complete their specific tasks. Sometimes there could be miscommunication since people from different departments come together and may not be used to working together. Poor decision-making because the person making the decision is more senior than the rest.
Nursing Scope of Practice
In order to operate and understand AI in the ICU setup, registered nurses (RN) need empirical knowledge. Challen et al. (2018) argue that this knowledge helps nurses understand abstract and theoretical explanations. This makes the nurse well-equipped with the knowledge of setting and understanding relations between variables.
RN nurses need to be skilled in information technology (IT) to understand the use of AI in the healthcare environment, especially in the ICU. By this, it does not mean they need to be computer experts; basic knowledge of interpreting AI results and feeding data to the system is vital. According to Meyer et al. (2018), AI mainly processes data to make decisions and conclusions.
Farhud and Zokaei (2021) argue that nurses with curious and optimistic attitudes and compassion are better off working in an ICU that has embraced AI. Through curiosity, they can understand new technology and easily embrace and use it.
AI is a new technology that is not easily acceptable to people. Like any other new technology, people will always be hesitant to embrace it (Gutierrez, 2020). Therefore, patients and their families need to be informed about the use of AI, how it works, and its benefits to the treatment. Before making their decision, they also need to understand the cons, like the chance of misdiagnoses, privacy breaches, and the increased cost of their treatment.
In order to teach them, the doctors and nurses need to interact with the patient and their family and have a sit down where they explain how AI is generally used and how it will be applied to their case. This is done by remembering to include the risks together with the benefit. This way, they will easily make informed consent (Matheny et al., 2020).
Because AI is new, some people feel that it will take over doctors’ jobs; others feel there are high chances of data breaches by using AI. However, in an actual sense, according to Komorowski et al. (2018), AI is improving lives by doing activities that humans would take too long to achieve or cannot achieve. Therefore the cultural impact of AI is making life simpler, more efficient, and even safer.
To evaluate the training and teaching of the patient and their families about using AI in ICU, AI can determine how well the patients and families understand its use. Excell can also review the teaching data (Thorsen-Meyer et al., 2020).
AI, which is human intelligence that machines have simulated, uses data sets to help with the diagnosis of diseases and develop the right treatment plans. AI has also been used to predict mortality. With all the benefits, some risks include data breaches or chances of misdiagnosis.
In the ICU, when AI is applied, the following interdisciplinary team members are available physicians, dieticians, therapists, clinical pharmacists, and clinical psychologists. In order to operate and understand AI, the RN nurse needs to have empirical knowledge, IT skills, and a positive, curious attitude toward the technology. When educating patients and their families, they must be informed of the benefits and risks of its use. Farhud and Zokaei’s (2021) article is a very beneficial additional resource that can be used to understand the use of AI in healthcare.
Challen, R., Denny, J., Pitt, M., Gompels, L., Edwards, T., & Tsaneva-Atanasova, K. (2019). Artificial intelligence, bias and clinical safety. BMJ Quality & Safety, 28(3), 231-237. http://dx.doi.org/10.1136/bmjqs-2018-008370
Farhud, D. D., & Zokaei, S. (2021). Ethical Issues of Artificial Intelligence in Medicine and Healthcare. Iranian Journal of Public Health, 50(11), i. https://doi.org/10.18502%2Fijph.v50i11.7600
Gutierrez, G. (2020). Artificial intelligence in the intensive care unit. Critical Care, 24(101), 1-9. https://doi.org/10.1186/s13054-020-2785-y
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. https://doi.org/10.1038/s41591-018-0213-5
Matheny, M. E., Whicher, D., & Israni, S. T. (2020). Artificial intelligence in health care: A report from the National Academy of Medicine. Jama, 323(6), 509-510. https://doi.org/10.1001/jama.2019.21579
Meyer, A., Zverinski, D., Pfahringer, B., Kempfert, J., Kuehne, T., Sündermann, S. H., … & 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. https://doi.org/10.1016/S2213-2600(18)30300-X
Thorsen-Meyer, H. C., Nielsen, A. B., Nielsen, A. P., Kaas-Hansen, B. S., Toft, P., Schierbeck, J., … & Perner, A. (2020). Dynamic and explainable machine learning prediction of mortality in patients in the intensive care unit: a retrospective study of high-frequency data in electronic patient records. The Lancet Digital Health, 2(4), e179-e191. https://doi.org/10.1001/jama.2019.21579