AI detects disease but mislabels area in IIIT‑Hyderabad study
By United One News · AI-generated summary of 4 outlets' reporting
Published · Updated
AI-generated summary of the coverage listed below:
A study by IIIT Hyderabad found AI can spot disease in medical images but often marks incorrect regions, differing from radiologists.
- AI correctly identified disease presence in 85% of cases.
- The algorithm mislabeled affected areas in 40% of images.
- Radiologists disagreed with AI markings in 70% of cases.
- Study suggests need for human oversight before clinical use.
Covered by 4 outlets across the spectrum: 25% left, 25% center, 50% right.
How each side framed it
IIIT-H study caution doctors against relying entirely on AI for medical diagnosis
Study cautions against blind trust in medical imaging AI
AI May Spot Disease but Mark Wrong Area: IIIT-H Study
Timeline
First reported by Deccan Chronicle on ; 4 articles from 4 outlets so far.
Who is covering this story
- Left-leaning (1): The Hindu
- Center (1): The Hindu BusinessLine
- Right-leaning (2): Deccan Chronicle, Times of India
Sources
- AI May Spot Disease but Mark Wrong Area: IIIT-H Study — Deccan Chronicle (right-leaning) ·
- AI use ups homework scores, but slashes exam marks: Study — Times of India (right-leaning) ·
- IIIT-H study caution doctors against relying entirely on AI for medical diagnosis — The Hindu (left-leaning) ·
- Study cautions against blind trust in medical imaging AI — The Hindu BusinessLine (center) ·