AI Applications In Medicine
Introduction
Artificial Intelligence (AI) is an incredibly powerful tool that easily has the potential to transform the field of science and medicine. Healthcare systems currently face many challenges that AI can be trained to solve. Especially during a global pandemic, AI was used in many situations to regulate systems that would have taken another staff member to do, helping with both understaffing and prevention of greater spread of such a disease. AI can be trained to prevent mistakes in diagnosing diseases and providing treatments, such as teaching intelligence using X-ray images of cancerous tumors to help detect tumors in new photos. While Artificial intelligence in the medical field is easily seen in its use in diagnostics, treatment, and its use to improve hospital conditions, there are also many challenges and ethical considerations in implementing AI into the medical workforce.
AI in Diagnostics
With the field of medicine advancing at miraculous speeds in our modern era, fast and consistent disease diagnosis is difficult, especially on a global scale. AI can be used and trained to distinguish certain diseases from other ones in order to effectively diagnose patients without human help. The BioMed Central finds that in using a type of AI called ML (Machine Learning), where one feeds an AI a large quantity of data for the AI to compute and learn from so it can distinguish similar patterns in new data, AI can be used to “assist in decision-making, manage workflow, and automate tasks in a timely and cost-effective manner” (Alowais). This use of AI will not only help with the short staffing crisis in the medical field due the pandemic, but will also allow research in medicine to be conducted at an even faster rate, since less people are needed to help diagnose patients’ diseases.
Los Angeles Pacific University also corroborates this data, writing that ML can and has been used to “analyze medical imaging data, such as X-rays, MRIs, and CT scans, to assist healthcare professionals in accurate and swift diagnoses,” leading to the ability for healthcare systems to be “smarter, faster, and more efficient in providing care to millions of people worldwide” (LAPU). They had performed a study in which they found that an AI model was “The AI-utilized diagnosis was more sensitive to diagnose breast cancer with mass compared to radiologists, 90% vs. 78%, respectively. Also, AI was better at detecting early breast cancer (91%) than radiologists (74%)” (Alowais). Despite being early in development, AI models have already far surpassed humans in many areas of diagnosis. This is only possible due to the outstanding potential that AI has in the medical field, allowing for less error in diagnosis and therefore increased safety and even lowering the current high price of healthcare, transforming the field altogether.
AI in Patient Care and Hospital Management
AI isn’t just usable in fields of research, but can also be applied to fields of patient welfare. ML can be used to alter the course of healthcare entirely for the better. An article from the National Institute of Health (NIH) National Library of Medicine (NLM) archive finds that AI is projected to completely change the way hospitals are viewed. They predict that in the long term, that is, over the course of at least 10 years, “healthcare will shift from the traditional one-size-fits-all form of medicine to a preventative, personalised, data-driven disease management model” that will be able to have “improved patient and clinical experiences of care in a more cost-effective delivery system” (Bajwa).
In addition, AI could enhance patient flow and experience throughout the hospital, being able to monitor and identify problems with sick patients in a timely manner. For example, healthcare experts Elizabeth Donner, Orrin Devinsky, and Daniel Friedman explored the use of wearable AI technology on epilepsy patients, finding that they can “provide data on seizure frequency” and “fill several critical gaps in epilepsy care” (Donner). Further examples of this are AI chatbots such as those used in Babylon or Ada, or recent ambient intelligence systems like Google Nest or MIT’s Emerald. (Bajwa) It is also noted that wearables can be used to assess a patient’s motion, heart rate, pulse rate, etc, which can be used for seizure detection. This suggests that wearables are going to be a necessity in healthcare in the near future, providing both information to scientists and researchers as well as a personalized system for each patient. Seeing this, there is no doubt that Artificial Intelligence is and will continue to play a notable role in not just healthcare, but in the upcoming modern era.
Ethical Considerations and Implementation Challenges
Although AI may have seemingly endless potential within the medical field, not all of its implementations may be worth the time and effort or ethically sound. While some critics anticipate that AI will replace physicians in jobs such as radiology and dermatology, Ted A James, a graduate from Harvard Medical School, finds that a collaboration of AI and human work in these fields outperforms either one on their own. He finds that it is near-impossible for AI to replicate the “human aspects of care, including empathy, compassion, critical thinking, and complex decision-making” which are “invaluable in providing holistic patient care beyond diagnosis and treatment decisions” (James). James also found that, while AI is incredibly useful in providing quick and precise information as well as addressing workforce shortages, participants of James’s digital transformation course said that they would prefer receiving news of a serious medical diagnosis from a doctor rather than a textbook AI. This is likely due to the nature of humans to want to connect with the things around us and form complex societies, which is much harder to do with an AI, who doesn’t understand the intricacies of human emotion. Moreover, a recent Pew Research Center survey highlighted how patients found AI as a cheat or a way around years of studying and learning within the medical field. In fact, “Six-in-ten U.S. adults say they would feel uncomfortable if their own health care provider relied on artificial intelligence to do things like diagnose disease and recommend treatments,” (Tyson) suggesting that AI makes people inclined to believe that the doctor or physician is reliant on the machine and does not understand what they are doing. In accordance with James, Pew Research Center’s survey also underscores the need for patient and health care provider connections, finding that “57% say the use of artificial intelligence to do things like diagnose disease and recommend treatments would make the patient-provider relationship worse” (Tyson). Thus, humans likely won’t be replaced by AI models anytime soon.
However, implementation of these AI models is not a walk in the park. First, the Journal of Medical Internet Research (JMIR) underscores the limitations of AI, such as how ‘“AI methods generally lack “common sense,” making them unable to identify simple mistakes in data or decisions that would otherwise be obvious to a human being”’ (Asan). This lack of common sense may be a complication within implementation of AI models. Second, the collection of unbiased data to provide to an AI model is more difficult than one would expect. Another article from NIH NLM highlights how “if the data collected is biased, that is, it targets a particular race, a particular gender, a specific age group then the resulting model will be biased,” (Basu) leading to the AI model to be essentially useless. Even if the data is unbiased, the AI can still come out biased through the many steps of integration and data reprocessing. Finally, AIs are usually trained to complete a task that is somewhat specific, but that results in a model that “cannot be seamlessly transitioned for immediate use to another organization without recalibration. Due to privacy concerns, data sharing is often inaccessible or limited between healthcare organizations resulting in fragmented data limiting the reliability of a model” (Basu). Consequently, there are many obstacles that medical researchers face when implementing AI models into the field of healthcare. So, while the loss of jobs within the medical field is likely out of question, one must still be incredibly careful when integrating AI models into the medical workspace.
Future Potential
Despite its many challenges and limitations, AI will continue to advance within the field of medicine due to its outstanding potential in diagnostics, research, and patient care. AI’s machine learning capabilities enable it to provide fast and reliable information to doctors more effectively than doctors. It can also be used in wearable pieces of technology or monitoring devices to regulate and respond to patients’ situations, personalizing hospital care. However, taking into account the significant limitations in AI and the risk of accidentally creating a biased AI model that disrupts the workspace and makes mistakes that can seriously alter a patient’s life is important and necessary in taking precautions when integrating these models. All in all, AI models have inconceivable promise within the medical field that will no doubt lead to breakthroughs in the near future that will advance human society in a way that has never been seen before. Artificial intelligence will transform the world as we know it at a rate so remarkable that it will also transform the way humans live as a society as a whole, integrating itself deeper and deeper into the roots of mankind.
Works Cited
Alowais, S.A., Alghamdi, S.S., Alsuhebany, N. et al. Revolutionizing healthcare: the role of artificial intelligence in clinical practice. BMC Med Educ 23, 689 (2023). https://doi.org/10.1186/s12909-023-04698-z
Asan O, Bayrak A, Choudhury A Artificial Intelligence and Human Trust in Healthcare: Focus on Clinicians J Med Internet Res 2020;22(6):e15154
Bajwa, Junaid et al. "Artificial intelligence in healthcare: transforming the practice of medicine." Future healthcare journal vol. 8,2 (2021): e188-e194. doi:10.7861/fhj.2021-0095
Basu, Kanadpriya et al. "Artificial Intelligence: How is It Changing Medical Sciences and Its Future?." Indian journal of dermatology vol. 65,5 (2020): 365-370. doi:10.4103/ijd.IJD_421_20
Donner, Elizabeth, et al. "Wearable Digital Health Technology for Epilepsy." The New England Journal of Medicine, 21 Feb. 2024. The New England Journal of Medicine, https://doi.org/10.1056/NEJMra2301913. Accessed 21 Dec. 2024.
James, Ted A. "How Artificial Intelligence is Disrupting Medicine and What it Means for Physicians." Harvard Medical School, President and Fellows of Harvard College, 13 Apr. 2023, postgraduateeducation.hms.harvard.edu/trends-medicine/how-artificial-intelligence-disrupting-medicine-what-means-physicians. Accessed 21 Dec. 2024.
Los Angeles Pacific University. "Revolutionizing Healthcare: How is AI being Used in the Healthcare Industry?" Los Angeles Pacific University, 21 Dec. 2023, www.lapu.edu/ai-health-care-industry/. Accessed 21 Dec. 2024.
Tyson, Alec, et al. "60% of Americans Would Be Uncomfortable With Provider Relying on AI in Their Own Health Care." Pew Research Center, 22 Feb. 2023, www.pewresearch.org/science/2023/02/22/60-of-americans-would-be-uncomfortable-with-provider-relying-on-ai-in-their-own-health-care/. Accessed 21 Dec. 2024.