Artificial Intelligence in Cancer Care: Ensuring Safe Use of Algorithms
Artificial intelligence (AI) is transforming cancer care through advanced algorithms, yet effective evaluation is crucial for safe clinical implementation. Amy Abernethy, MD, cofounder of Highlander Health, highlighted this during a presentation at ESMO Congress 2024. With hundreds of AI algorithms cleared for clinical use, Abernethy noted that less than half have been thoroughly evaluated in real-world clinical contexts.
The sophistication of these AI tools relies on complex datasets combined with genomics, necessitating comprehensive assessments to ensure efficacy in real-time clinical settings. Abernethy identified three ongoing trends facilitating scalable evaluations: the rise of high-quality, organized multimodal data, more confident evaluation approaches for algorithms, and advancements in evidence generation.
As healthcare innovations continue, oncologists must understand what constitutes effective AI algorithms and how to foster confidence in their application for patient care.
This article was prepared using information from open sources in accordance with the principles of Ethical Policy. The editorial team is not responsible for absolute accuracy, as it relies on data from the sources referenced.