
Dermatologist, Associate Professor, @Sahlgrenska @SahlgrenskaAcad @goteborgsuni | Deputy Editor @ActaDV | Tweets are my views and opinions |
Göteborg, Sverige
Joined March 2020
- Tweets402
- Following872
- Followers490
- Likes118
Pinned Tweet
Just 6 out of 17,799 patients with atopic dermatitis (0.3‰) were visually represented across 60 clinical trials, impacting patient engagement.
❗️A call to action for more inclusive research❗️#dermtwitter
@goteborgsuni @KlinvetGu @karolinskainst
👇🏻
doi.org/10.1080/09546634.202…
2
1
386
Full article: It is still a secret disease – clinical images in randomized controlled trials on hidradenitis suppurativa tandfonline.com/doi/full/10.…
3
87
Sam Polesie retweeted
The study: doi.org/10.1111/jdv.20069
Sam Polesie retweeted
What stuns me the most about this story is that a dermatologist would biopsy someone 28 times at a SINGLE appointment. That many biospies at a single appointment is shocking!
After her mom died from skin cancer, she went to get checked and was shocked to receive the same diagnosis. today.com/health/womens-heal…
Increasing melanoma incidence with unchanged mortality: more sunshine, better treatment, increased diagnostic activity, overdiagnosis or lowered diagnostic threshold? academic.oup.com/bjd/article…
Sam Polesie retweeted
A nationwide study found that an AI language model outperformed physicians in medical board residency exams across multiple specialties. Learn more in "GPT versus Resident Physicians — A Benchmark Based on Official Board Scores" by @uriel_katz_, et al. nejm.ai/3Q7hB9j

ALT Figure 1 from an article titled "GPT versus Resident Physicians — A Benchmark Based on Official Board Scores": Distribution of GPT Examination Score Percentiles among Physicians. Generative pretrained transformer (GPT) model percentiles among physicians on 120 attempts at the examination for every medical specialty are shown. The dotted red lines mark the percentile corresponding with a passing score. The graphic was created by the authors using data from the 2022 Israeli board residency examinations and results from the GPT models. OB/GYN denotes obstetrics and gynecology.
Evaluation of an artificial intelligence-based decision support for the detection of cutaneous melanoma in primary care: a prospective real-life clinical trial academic.oup.com/bjd/article…
1
5
339
Sam Polesie retweeted
The results have been published in @BrJDermatol:
Evaluation of an artificial intelligence-based decision support for the detection of cutaneous melanoma in primary care: a prospective real-life clinical trial.
academic.oup.com/bjd/advance…
doi: 10.1093/bjd/ljae021
Sam Polesie retweeted
This article looks at correlation between #melanoma incidence and mortality in the context of #Breslow thickness groups and highlight the need for large studies with extended follow-up while considering timing, frequency, and population of screenings. bit.ly/4cjorBN
Sam Polesie retweeted
Bör individualiserad utbildning för läkarstudenter eftersträvas? dlvr.it/T3M78m
Sam Polesie retweeted
This article sought to test #ChatGPT-4's ability to translate dermatopathology results into patient-friendly language. While results were mixed, it could become a useful tool in dermatology, though it should always be reviewed prior to release to patients. bit.ly/49hT1d6
Sam Polesie retweeted
What is this patient’s normal skin color?
20%The lighter skin
40%The darker skin
40%None of the above
5 votes • Final results Sam Polesie retweeted
Melanoma is overdiagnosed at ‘alarming’ rates. Here’s what to know. on.natgeo.com/4byIEmU appreciate the write up of our work on this underappreciated problem among dermatologists
Sam Polesie retweeted
New paper out 🥳 In an effort led by @AchimHekler and @RomanMaron, we show that using multiple real-world dermoscopic photographs of one lesion improves #melanoma #classification significantly & recommend use in clinical care @JAADjournals: jaad.org/article/S0190-9622(… @DKFZ
Sam Polesie retweeted
Our new paper is out @ JAMA Dermatology! 🎉 Together with six German University Hospitals, we demonstrated that a #federated #learning approach for decentralized #AI in #Melanoma #Diagnostics is superior to centralized approaches. Enjoy reading: jamanetwork.com/journals/jam… @DKFZ
















