A 10-year-old CNN outperforms frontier LLMs for skin cancer diagnosis
The author tested two frontier LLMs and a fine-tuned Inception v3 CNN on 200 biopsy-proven skin lesions from the public MILK10k dataset, giving each clinical and dermoscopic images. The CNN named the correct diagnosis (out of six options) for 125 lesions (62%), versus 84 and 85 (42%) for the LLMs. The CNN also recognised more melanomas and harmless lesions, while the LLMs showed favourite wrong answers. The author notes the CNN is an older baseline, not state of the art, and the sample is small.
Open original source lynchlab.netOne primary source; other links are context, not independent corroboration.
- Primary sourceHacker News · LLM
- Discovery linkHacker News
- Linked in sourceMILK10k
- Linked in sourceISIC Archive
- Linked in sourcetry the CNN yourself
- Linked in sourcein one download
- Linked in sourceFull profile
A community post points to the original link. Discussion and upvotes do not establish accuracy.
- Community discussion+5
- Fetched source page+30
- No editor link review0
- Community link0
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
The author tested two frontier LLMs and a fine-tuned Inception v3 CNN on 200 biopsy-proven skin lesions from the public MILK10k dataset, giving each clinical and dermoscopic images. The CNN named the correct diagnosis (out of six options) for 125 lesions (62%), versus 84 and 85 (42%) for the LLMs. The CNN also recognised more melanomas and harmless lesions, while the LLMs showed favourite wrong answers. The author notes the CNN is an older baseline, not state of the art, and the sample is small.
A READER'S EVIDENCE CHECKHow to read this signalSix questions about claims, sources, risks, and open checks
This is a signal-level guide until claim-level evidence is available. “Not recorded” means the evidence has not been assessed here; it does not mean a result is false. How we check AI news →
- 01Who is making each claim?
The claimant and whether each claim was measured or asserted are not recorded in this public brief.
- 02What was actually tested?
Not recorded per claim. Check the source for sample size, tasks, model and benchmark versions, and reproducibility.
- 03Where are the sources?
Original link: lynchlab.net. Brief basis: fetched source page. Primary page date and quote locations are not established in this panel.
- 04What could mislead?
Small sample of 200 lesions (30 melanomas), each LLM case run once, and the author's own methodology and clinical rulings have not been independently checked.
- 05What is at risk if wrong?
Not assessed per claim. Consider safety, privacy, cost, and adoption decisions before relying on the result.
- 06What would change the view?
No claim-specific check is recorded. An inspectable primary method or independent replication could change the assessment.
Key findings from the stored brief
- The fine-tuned Inception v3 CNN gave the correct first-choice diagnosis for 125 of 200 lesions, versus 84 and 85 for the two LLMs.
- The CNN named 13 of 30 melanomas correctly, compared with 8 for GPT-6 Astra and 11 for Claude Fable 5.1.
- The author states the CNN is a deliberately non-state-of-the-art baseline and that 200 lesions give only indicative results.
Small sample of 200 lesions (30 melanomas), each LLM case run once, and the author's own methodology and clinical rulings have not been independently checked.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
A community post points to the original link. Discussion and upvotes do not establish accuracy.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
A publication revision history is not stored for this signal. New engagement snapshots do not mean the source itself changed.
Why this may matter
The result suggests general-purpose LLMs may underperform specialised image models on dermatologic diagnosis, a relevant finding for medical AI evaluation. Its small sample and author-run design mean it should be read as preliminary.
Limits and open questions
Small sample of 200 lesions (30 melanomas), each LLM case run once, and the author's own methodology and clinical rulings have not been independently checked.
Organization-specific financial, security, and operational effects require the organization's own context and review.
Discussion
Comments are reviewed before publication. Discuss the evidence, not the person.
No published comments yet.
Reader discussion opens with accounts. You can read the original source and published comments now.