Can AI really predict cancer survival better than a doctor? I have my doubts | Devi Sridhar
The short version
AI's potential in predicting cancer survival using facial analysis raises significant doubts regarding its practical utility and biases. While promising, the technology may not consistently outperform medical expertise.
Key takeaways
- AI developed from 40 million images aims to predict cancer survival.
- Results indicate effectiveness varies for different cancer types.
- Racial biases complicate AI's predictive capabilities.
- Question remains if AI adds real value beyond clinician experience.
Overall sentiment
MixedThe article expresses skepticism about AI's promises while acknowledging the potential for innovation.
AI classification of the article’s tone. Limited coverage focuses on the implications of AI in clinical settings and biases, rather than detailed methodologies.
Sources cited by the article
- Devi Sridhar ↗
- recent headlines ↗
- linked with dying ↗
- Dr Raymond Mak ↗
- Prof Miroslaw Bober ↗
- growing concern ↗
- cardiac magnetic resonance (CMR) scans ↗
- AI (artificial intelligence) ↗
- Cancer research ↗
- Medical research ↗
- comment ↗
- Reuse this content ↗
These references were linked or listed by the publisher, then selected for relevance. They are distinct from our original-source citation.