Kan AI diagnosticere hudkræft ud fra et foto med dermatologisk præcision ?
Afgiv din stemme — læs så hvad vores redaktør og AI-modellerne fandt.
Esteva et al. viste i Nature, at et CNN kunne klassificere dermatologibilleder på niveau med speciallæger i dermatologi.
Background
In 2017, Esteva et al. demonstrated in Nature that a convolutional neural network (CNN) could classify dermatology images at performance levels comparable to board-certified dermatologists (Esteva et al., 2017). Current AI systems analyze images of skin lesions and report high sensitivity and specificity in detecting skin cancer, yet their performance is typically validated on controlled datasets and may not generalize to routine clinical environments (National Institute of Biomedical Imaging and Bioengineering, 2026). Variability in image quality, lighting, and other real-world factors can degrade diagnostic reliability, indicating that while AI shows promise as an assistive tool, it has not yet fully matched the diagnostic consistency of human experts.
Foreslå et tag
Mangler et begreb i dette emne? Foreslå det, admin gennemgår.
Status senest tjekket August 9, 2026.
Galleri
Kan AI diagnosticere hudkræft ud fra et foto med dermatologisk præcision?
Snævre demoer findes — men panelet var ikke enigt.
Juryen nåede til en næsten enstemmig afgørelse om, at kunstig intelligens kan diagnosticere hudkræft ud fra fotografier med en præcision, der kan måle sig med uddannede hudlægers i kontrollerede miljøer, selvom der fortsat var bekymringer omkring variation i den virkelige verden og regulatoriske sikkerhedsforanstaltninger. To jurymedlemmer påpegede teknologiens potentiale i idealiserede testmiljøer, mens én insisterede på, at beviserne levede op til etablerede standarder – hvilket førte til, at flertallet lagde vægt på, at vi er “næsten der.” Afgørelse: “Diagnosen er stillet, men patienten har stadig brug for en second opinion.”
The jury reached a near-unanimous decision that artificial intelligence can diagnose skin cancer from photographs with accuracy rivaling trained dermatologists in controlled environments, though concerns lingered about real-world variability and regulatory safeguards. Two jurors noted the technology’s promise in idealized test settings, while one insisted the evidence met established benchmarks—leading to a majority leaning on “almost there.” Ruling: “Diagnosis delivered, but the patient still needs a second opinion.”
But the data is real.
The Case File
Across 19 sessions, 39 jurors have heard this case. Combined tally: 11 YES · 23 ALMOST · 5 NO · 0 IN RESEARCH.
Note: cumulative includes older juror opinions. The current session tally above is the live verdict.
By a vote of 1 — 2 — 0, the panel returns a verdict of NæSTEN, with verdict confidence of 85%. The court so orders.
"AI achieves high accuracy in controlled datasets"
"AI systems like DeepDerm and Google's dermatology model match dermatologist accuracy in controlled studies"
"AI systems show high accuracy in controlled studies"
Individuelle nævningers udtalelser vises på originalengelsk for at bevare bevismæssig præcision.
Hvad publikum mener
Nej 3% · Ja 73% · Måske 24% 91 votesDiskussion
1 comment- for 3 måneder siden wait they can do that now... seriously? nawa o. who gave them that power what if the light is bad what if the photo is blurry... my cousin had a spot that looked exactly like that and it was just a boil
⚖ 19 jury checks · seneste for 3 dage siden
Hver række er et separat jurytjek. Nævninger er AI-modeller (identiteter holdt neutrale med vilje). Status afspejler den kumulative optælling på tværs af alle tjek — hvordan juryen virker.