- Nurse educators are behind the professional curve of AI due to hesitancy and a lack of research available.
- AI has many tenets with various capabilities and levels of intelligence.
- Nurse educators can incorporate AI to enhance learning and critical thinking.
Many nurse educators entered the field well before AI had a prominent role in healthcare and nursing education curricula. AI is used widely across healthcare settings. However, a limited understanding or competence has stalled its progression, especially for nurses.
In a recent study published in the Journal of Nursing Education, nurse educators Tonya Schneidereith, Ph.D., MBA, and Joseph Thibault, MBA, MEd, established the building blocks for how nursing schools can integrate AI into their curriculum safely, responsibly, and ethically.
“AI offers exciting possibilities; however, AI is a double-edged sword,” the co-authors wrote. “The adoption of this technology offers many benefits but also presents risks to academic integrity and appropriately prepared graduates. Many of today’s nurse educators are from generations that are unlikely to possess an understanding of AI.”
Here are a few key concepts and definitions that nurse educators should be aware of, which have wider applications in the healthcare delivery system.
AI is an umbrella term for any machine that can replace some aspect of human intelligence. The machine learns independently by analyzing data over time, creating intelligent machines, and developing algorithms. The system uses input data to reason, learn, and process.
Computers can learn without programming. Machine learning tries to imitate intelligent human behavior and how people solve problems.
Machine-learning algorithms make predictions after identifying patterns and trends. Ultimately, they are allowed to program themselves through experience. Examples include chatbots, predictive text, and suggestions for you based on previous choices, such as Amazon shopping recommendations and Netflix viewing suggestions.
Deep learning has many layers of “neural” networks and is trained to process information and patterns that are too complex for humans, such as image recognition. The neural networks are modeled after the structure and function of the human brain to process and connect large amounts of data. During training, the connections’ strength can be adjusted.
Healthcare-related benefits from deep learning include creating more efficient processes, such as analyzing diagnostic images and monitoring patient self-care and medication administration.
Natural Language Processing
Natural language processing (NLP) allows computers to interpret and respond to human language. NLP runs the chatbots used for customer services and personal assistants, such as Google Translate, Siri, and Alexa. Search engines also use it to understand a webpage’s content to return results other than with keywords. Text-to-speech, language translation, and chatbots are also NLP. The end goal is for NLP to interpret human language and generate responses.
Prediction models predict outcomes based on data from previous events. It calculates probability based on earlier data on similar events and hidden trends and predicts the best outcomes. Nursing-related predictions include risk assessments, such as falls and skin breakdown risks.
The expert systems’ purpose is to complement human experts. They simulate the judgment and behavior of one or more humans to become the most efficient. It takes relevant knowledge from its database and interprets it according to the user’s problem.
Humans must enter the data, or rules, into the database because it doesn’t learn and needs data updates. Further, expert systems have no common sense or emotion and can’t explain the logic behind the decision.
Given the data inputted, healthcare professionals can utilize expert systems to assist in medical diagnoses. With the skin breakdown example above, an expert system can use the rules created for it to identify and treat pressure ulcers by stage.
Fuzzy logic uses unclear or vague data and provides flexibility in reasoning. Imprecise or uncertain information is used rather than true or false. It allows for shades of gray and partial truth.
An example of fuzzy logic is the weight of the data assigned to the many factors in diabetes management and treatment, such as pre-meal and post-meal blood sugar levels, grams of carbohydrate intake, and the patient’s level of insulin resistance.
Artificial Narrow Intelligence
Artificial narrow intelligence (ANI) has a narrower range of abilities. Examples include speech-to-text capabilities like Apple’s Siri, Amazon’s Alexa, and Microsoft’s Cortana.
In the medical field, ANI helps diagnose cancer and other diseases with extreme accuracy through human behavior, cognition, replication, and reasoning.
Generative AI (most commonly associated with ChatGPT) is a subset of deep learning that can create new artificial content or data. Large sets of data from books, academic journals, and the internet were entered into its system. With a prompt, it can produce new text or content. Generative AI can’t analyze data or think critically, though. Therefore, humans must judge the accuracy of the content it produces.
Timesaving benefits for the nursing field include automatic sentence completion. However, legal patient care documentation requires a nurse’s critical thinking ability to override AI suggestions or auto-completion when necessary.
Recommendations for Nurse Educators and AI
Research shows that the need for nurse educators to incorporate the various subsets of AI is valuable to enhancing and preparing nursing students for professional practice when used responsibly.
However, nurse educators may correlate AI technology with its potential for academic dishonesty rather than its educational advantages (for example, using ChatCPT to write essays). A thorough understanding of AI and its usage can significantly enhance the future of student learning.
There are many opportunities for AI to enhance learning, according to a 2023 article. Chatbots can help individualize students’ training and prepare them for tests with question-and-answer sessions. Human simulators can respond verbally and physically to verbal prompts, including answering questions and moaning in pain. Avatars can answer questions in real time to help them think critically and reason to find solutions. AI also powers telehealth and remote monitoring of patients. What was once time-consuming documentation can be reduced through the power of AI.
Nurses are taught to exercise critical thinking skills. The ability to understand and utilize AI has the advantage of maximizing and safely implementing advanced patient care. Continued research is necessary to create best practices for nurse educators to use AI by nurses legally and ethically.
In their paper, Schneidereith and Thibault called for nurse educators to “understand the basics of AI, judge the appropriateness of integration, and recognize opportunities to embrace future application.”
She also emphasized that “nurses should be involved in the development of patient care technology to help ensure usability and accurate reflections of clinical reality.”
“We have a duty to understand the basics of AI, to judge the appropriateness of integration, and to recognize opportunities to embrace future applications,” they concluded. “It is our role to sift through the hype and hand-wringing and to understand, in short order, the applications of AI to the nursing profession.”