Session 6/7 Assignment: AI in Healthcare by Rose Slahetka

Overview:

roseslahetka — 2025-10-11 13:54:01 UTC

Healthcare is an extremely staple in our everyday society. Over the past decade, artificial intelligence, also known as AI, has begun to transform many aspects of healthcare. Examples of this diagnostic imaging and clinical decision support to remote monitoring and patient triage. I chose healthcare because I am fascinated by how technology can help improve patient outcomes, reduce medical errors, and expand access to care. In this Padlet wall, I will cover three specific AI applications in healthcare (diagnostic imaging/ radiology, remote patient monitoring, and AI in triage/ chatbots). For each application I’ll describe the underlying technology, benefits and challenges, and reflect on their societal and ethical implications. I will also discuss future trends, ethical considerations, and the societal impact. My personal interest within AI uses in Healthcare and how it may help reduce inequalities in underserved communities. This is especially prominent in rural or low-resource areas. I hope this project will deepen my understanding of both the potential and risks of AI in healthcare.

Butterfly Network: AI in Diagnostic Imaging

roseslahetka — 2025-10-11 13:58:06 UTC

Butterfly Network is changing medical imaging by using artificial intelligence in handheld ultrasound devices. These small scanners use computer vision and deep learning to read ultrasound images in real time and help doctors find possible health problems faster (Nesa et al., 2025). This makes imaging more affordable and available, especially in areas without big hospitals or expensive machines. One challenge is that the accuracy of results depends on image quality, and doctors must still double-check AI findings. I find this technology inspiring because it helps bring life-saving tools to rural or low-income communities.

Biofourmis: AI in Remote Patient Monitoring

roseslahetka — 2025-10-11 14:00:58 UTC

Biofourmis uses artificial intelligence to track patients’ health from home with smart wearables that record data like heart rate and breathing. The system uses machine learning to spot early signs of illness and alert doctors before problems get worse (Shaik et al., 2022). This helps people stay out of the hospital and get care sooner. However, issues like privacy, device reliability, and internet access can make things difficult. I find this technology meaningful because it helps people with chronic illnesses, especially those who cannot travel easily or live far from hospitals.

Babylon Health: AI in Clinical Triage and Decision Support

roseslahetka — 2025-10-11 14:03:09 UTC

Babylon Health uses AI chatbots to help patients describe symptoms and get advice about what kind of care they might need. The system uses natural language processing, also known as NLP, and large language models, also known as LLM, to understand patient messages and give helpful suggestions (Ullah & Ali, 2025). It can work at any time, reducing wait times and helping busy doctors. However, AI can make mistakes or misunderstand information, so human review is still very important. I find this tool interesting because it makes healthcare more accessible, especially in places with few doctors or long wait times.

Future Trends and Ethical Considerations:

roseslahetka — 2025-10-11 14:04:30 UTC

Artificial intelligence continues to reshape healthcare, with new trends promising to enhance both diagnosis and treatment. Federated learning, also known as FL, allows hospitals to train AI models collectively without sharing raw patient data, and has improving performance while maintaining privacy (Nguyen et al., 2024). Explainable AI, also known as XAI, is also becoming popular. This allows for transparency through interpretability tools such as SHAP and LIME, which make AI decisions understandable to clinicians (Nadarajan & Raj, 2023).AI will also connect with genetic data and personalized medicine to better predict health risks, and new regulations will help ensure these systems are safe and reliable for clinical use.

These innovations do raise serious ethical considerations. Which include bias and fairness. These remain pressing issues, as models trained on limited datasets can underperform for minority groups (Banerjee et al., 2023). Privacy and data security are still big concerns because, even with methods like data anonymization and federated learning, sensitive information can still leak (Choudhury et al., 2023).

I believe the most important ethical priorities for AI in healthcare are transparency, fairness, and equal access. AI systems should not only be advanced and effective but also easy to understand, unbiased, and available to everyone, regardless of where they live or what resources their healthcare system has. Ensuring that AI tools are explainable helps build trust among clinicians and patients, while focusing on fairness and accessibility ensures that the benefits of these technologies reach all groups, not just those in well-funded or technologically advanced settings. By addressing these issues, AI can truly support better, more equitable healthcare outcomes for everyone.

Societal Impact:

roseslahetka — 2025-10-11 14:06:03 UTC

Artificial intelligence is changing healthcare and affecting society in many ways. One major impact is on employment. AI can take over repetitive tasks like medical image screening and data entry, which may reduce some traditional roles. There is opportunity to create new ones focused on AI oversight and data management (Jiang et al, 2022). This shift means healthcare workers will need new skills to work alongside technology.

AI also influences equity and access. It can expand care to underserved communities through telemedicine and remote monitoring tools, helping people in rural or low-income areas receive faster diagnoses (Kumar et al, 2023). However, hospitals with fewer resources might not be able to afford or maintain these tools, which could increase existing inequalities.

Privacy is another key issue. AI systems collect large amounts of sensitive data, and even with strong protections, there are risks of misuse by employers or insurers (Choudhury et al., 2023). Maintaining privacy is essential to protect trust in the healthcare system.

The public’s trust and perception also matter. Many people worry about machines replacing doctors or making mistakes without human judgment. Clear rules, transparency, and ethical guidelines are needed to make AI more trustworthy (Gerke et al., 2021).

I feel hopeful that AI can make healthcare more accessible and affordable, but I am concerned about data control and fairness. The benefits of AI should reach everyone, not just those with advanced technology.

Reflection:

roseslahetka — 2025-10-11 14:08:28 UTC

While researching AI in healthcare, I was amazed by how many exciting possibilities exist. I also discovered the many issues that remain, like bias, limited data sharing, and quality of service. One issue I did face was finding peer-reviewed, and open-access articles on some specific topics was challenging. Therefore I utilized systems like chat-GPT to help me find open journals that are accredited, and peer reviewed. This made me realize how important it is to stay hopeful about AI’s potential while also being aware of its risks. This project gave me a deeper understanding of how complicated it is to use AI in real healthcare settings. It also showed me that innovation must go hand in hand with strong governance, transparency, and fairness to truly succeed.

References:

roseslahetka — 2025-10-11 14:18:25 UTC

-Banerjee, I., Singh, R., Li, C., & Desai, A. D. (2023). Unmasking bias in AI: A systematic review of fairness in EHR-based models. arXiv. https://arxiv.org/abs/2310.19917

-Choudhury, O., Das, A., & Khandoker, A. (2023). Privacy-preserving federated learning for healthcare: Challenges and opportunities. International Journal of Environmental Research and Public Health, 20(15), 6539. https://doi.org/10.3390/ijerph20156539

-Gerke, S., Minssen, T., & Cohen, G. (2021). Ethical and legal challenges of artificial intelligence-driven healthcare.Journal of the American Medical Informatics Association, 28(7), 1574–1579.

https://pmc.ncbi.nlm.nih.gov/articles/PMC7332220/

-Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., & Ma, S. (2022). Artificial intelligence in healthcare: Past, present and future. Stroke and Vascular Neurology, 7(1), 11–18.

https://pubmed.ncbi.nlm.nih.gov/29507784/

-Kumar, A., Dey, A., & Sharma, R. (2023). Artificial intelligence in telemedicine and rural healthcare: Opportunities and challenges. Healthcare, 11(6), 835. https://doi.org/10.3390/healthcare11060835

-Nadarajan, T., & Raj, R. (2023). Explainable artificial intelligence in healthcare. International Journal of Engineering Research & Technology, 12(5), 263–271. https://ecejournals.in/index.php/INES/article/view/263

-Nesa, L., Rony, M. K. K., Chowdhury, S., Naznin, M. B., Halder, K., Ara, M. H., Akter, N. N., Mankhin, K., Shabnur, J. M., Alam, J., Parvin, M. R., Alrazeeni, D. M., & Akter, F. (2025). Artificial intelligence in healthcare: A scoping review of medical professionals' acceptance and institutional challenges in implementation. Journal of Evaluation in Clinical Practice, 31(4), e70170. https://doi.org/10.1111/jep.70170

-Nguyen, T. T., Pham, T., & Liu, J. (2024). Federated learning for healthcare predictive modeling: A multicenter study.Journal of Biomedical Informatics, 158, 104589. https://pubmed.ncbi.nlm.nih.gov/38681759/

-Shaik, T., Tao, X., Higgins, N., Li, L., & Gururajan, R. (2022, October 13). Remote patient monitoring using artificial intelligence. arXiv. https://arxiv.org/pdf/2301.10009

-Ullah, W., & Ali, Q. (2025, January 23). Role of artificial intelligence in healthcare settings: A systematic review. Journal of Medical Artificial Intelligence. https://jmai.amegroups.org/article/view/9683/html

-Wang, L., Krishnan, A., & Jones, P. (2022). Equity and access in AI-based healthcare: Challenges and recommendations.Journal of Medical Internet Research, 24(11), e40783. https://pubmed.ncbi.nlm.nih.gov/36450391/

Video Overview- Part 1:

roseslahetka — 2025-10-12 18:46:06 UTC

Video Overview- Part 2:

roseslahetka — 2025-10-12 18:52:15 UTC

Video Overview- Part 3:

roseslahetka — 2025-10-12 19:05:54 UTC