Artificial Intelligence in Healthcare A Review of Machine Learning Applications
Answer: AI in medicine. AI in medicine has been a massive advance in diagnostics, predictive analytics, and patient care. Despite its potential, however, there are significant barriers to widespread adoption, such as data privacy issues, high computational costs, AI bias, lack of standardized evalua...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2025-01-01
|
| Series: | ITM Web of Conferences |
| Subjects: | |
| Online Access: | https://www.itm-conferences.org/articles/itmconf/pdf/2025/07/itmconf_icsice2025_01012.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Answer: AI in medicine. AI in medicine has been a massive advance in diagnostics, predictive analytics, and patient care. Despite its potential, however, there are significant barriers to widespread adoption, such as data privacy issues, high computational costs, AI bias, lack of standardized evaluation, regulatory barriers, and integration with legacy healthcare systems. At present, the challenges explored highlight the need for federated learning as a new way to train AI without exposing sensitive patient data, bias-aware models which promote equitable and fair healthcare decisions for all patients, cloud and edge AI to ensure that processing is cost effective and appropriate, and Explainable AI (XAI) to promote trust and transparency to patients and communities. Additionally, we introduce an AI middleware framework, developed to integrate AI into existing Electronic Health Records (EHRs), enabling seamless uptake into clinical arenas. Summary: To enable privacy-preserving, fair, efficient, and regulatory-compliant AI and accelerate AI-driven innovations in the healthcare domain this research will develop an AI benchmarking framework where the progress of AI will be monitored and regulated. This will pave the way for scalable, interpretable, and sustainable AI applications that can close the gap between the existing theoretical AI models and their use in real-world clinical settings. |
|---|---|
| ISSN: | 2271-2097 |