One study concluded that bots on Reddit could intercept misinformation on dermatology forums that could potentially cause harm
Identifying and responding to health misinformation on Reddit dermatology forums with artificially intelligent bots using natural language processing: design and evaluation study
Reddit is the fifth most popular website in the US. The site has a large community that is active on its dermatology forums, where the public can access free medical information and advice. However, according to the authors, most of the advice that is found on this platform is unvalidated and could lead to inappropriate care.
Initial testing, which was carried out by the authors, has shown that artificially intelligent bots can identify incorrect information on Reddit in regard to tanning and essential oils and may be able to produce responses to posts with misinformation.
The aim of this study was to analyse the bots' ability to find and respond to health misinformation on Reddit's dermatology forums in a controlled test environment. To do this, the authors used natural language processing techniques and trained the bots to target misinformation. They used relevant keywords to post responses.
The results showed that the models used by the authors yielded data test accuracies ranging from 95% to 100%. In the test environment, bots were able to post corrective responses to the misinformation.
The authors concluded that bots can accurately detect examples of incorrect health information within Reddit dermatology forums. It is hoped that the ability of these bots to post prefabricated responses may allow for interception of misinformation that could potentially cause harm. The authors stated that providing accurate information does not mean that users will be receptive or find these interventions persuasive. They also believe that further studies should be carried out.
‘Rainbow pattern’: a dermoscopic sign of invasive melanoma
Initially, the rainbow pattern was described as a common, but specific, feature of Kaposi's sarcoma. Over the years, it has been a feature in many benign and malignant cutaneous tumours, including some melanomas.
The aim of this study was to determine the frequency and presentation of this dermoscopic pattern in primary cutaneous melanomas compared to other cutaneous tumours.
The authors evaluated the presence of rainbow pattern in 1100 dermoscopic sample images. These images were of different melanocytic and non-melanocytic cutaneous neoplasms.
The results found showed that the rainbow pattern was observed in 9.4% of melanomas and 5.1% of non-melanoma neoplasms. The authors also found that melanomas with this feature were thicker than 1 mm and 2 mm. When compared to non-melanomas, rainbow pattern was more commonly associated with more than two dermoscopic structures associated with melanoma.
In conclusion, the authors agreed that the rainbow pattern is a dermoscopic sign that can occasionally be observed in melanomas. In melanomas, this feature is usually associated with other dermoscopic criteria of melanoma.
Fractional carbon dioxide laser for treating hypertrophic scars: a systematic review of randomised trials
Hypertrophic scars present with collagen deposition and an abnormal extracellular matrix that can cause unusual shape changes and restrict normal movement. Over the years, fractional carbon dioxide (CO2) laser therapy has been used to treat hypertrophic scars with promising evidence. However, the improvement of scarring has not been comprehensively reviewed.
The authors of this study carried out a systematic review of published randomised trial articles that they found on PubMed, MEDLINE, EMBASE, Cochrane and Scopus databases. They included five randomised studies and they split the outcomes into two groups. Treatment efficacy was assessed as the primary outcome, while adverse events and patient satisfaction were assessed as the secondary outcome.
The authors noted that all studies showed consistent results that CO2 fractional laser treatment demonstrated statistically significant improvement for various scar scoring methods. The authors also concluded that combination treatments may give better results. They also commented that side effects, such as itching or burning, erythema and oedema, were present, but they appeared to be minimal and well tolerated. Overall, patients reported considerable improvement in their quality of life.
Perceptions of the use of artificial intelligence in the diagnosis of skin cancer: an outpatient survey
Convolutional neural networks (artificial intelligence (AI)) are increasingly being seen in the dermatology field. These technologies are becoming widely available for use by both health professionals and the general public. NHS England realises AI's potential in healthcare, but stresses that patient-centred care should be kept at the forefront of these technological advancements.
The aim of this study was to collect the opinions of patients on the use of AI in a dermatology setting. The authors devised a 14-point questionnaire, which was handed out to patients attending dermatology clinics in two UK secondary care hospitals between March and August 2018.
A total of 603 patient questionnaires were completed. It was found that 47% of respondents were not concerned if AI technology was used by a skin specialist to assist skin cancer diagnosis. Some 81% of patients considered it crucial for a dermatologist to examine and confirm a diagnosis and be present for discussion of a cancer diagnosis.
The authors conclude that, while the majority of patients are not concerned with the use of AI for skin cancer diagnosis, it is still considered hugely important that dermatologists are involved in the diagnosis and/or confirmation of skin cancer. Face-to-face interaction with a clinician is felt to be of huge importance, so AI is not considered a substitute for a dermatologist.
Clinical and pathologic correlation of cutaneous COVID-19 vaccine reactions including V-REPP: a registry-based study
Skin reactions following COVID-19 vaccinations have been widely reported. However, histopathologic features and clinical correlations have not been well documented.
The authors assessed all reports of reactions associated with COVID-19 vaccinations and identified an international registry. When histopathology reports were available, they categorised them by patients' reaction patterns. A total of 803 vaccine reactions were reported, and 58 of these cases had biopsy reports available for review. The most frequent histopathologic reaction pattern was spongiotic dermatitis, with a spectrum of different appearances. The authors coined the acronym vaccine-related eruption of papules and plaques (V-REPP) for this spectrum.
The authors noted that the cases where histopathology was available represented a minority of registry entries. The analysis of registry data cannot measure incidence. The categorisation of cutaneous reactions to the COVID-19 vaccine were allowed by clinical and histopathologic correlation. The authors proposed defining a subset of vaccine-related eruption of papules and plaques following COVID-19 vaccination.