Esra Adışen

Introduction: Accurate diagnosis in dermoscopic examination is challenging due to the visual similarity between skin lesion types and the variability within the same group. Early diagnosis and treatment of skin diseases, especially skin cancer, are vital. Objectives: This study aimed to develop and evaluate an explainable multi-architecture framework integrating convolutional, transformer-based, hybrid, and ensemble learning approaches for multi-class skin lesion classification. Methods: Using HAM10000, four architectures (EfficientNet-B0, MobileNetV3-Large, ResNet34, DeiT-Small) were fine-tuned in two phases and combined into a validation-optimized, macro-F1-driven soft-voting ensemble; an independent ResNet34+XGBoost hybrid was evaluated in parallel. Lesion-grouped partitioning was used to prevent overlap between data subsets, with an additional image-level experiment assessing the influence of partitioning strategy on performance. Evaluation used accuracy, macro-averaged precision, recall, F1-score, per-class metrics, nested 5-fold cross-validation, McNemar's test, and lesion-clustered bootstrap effect sizes (95% CIs). Grad-CAM, Grad-CAM++, and LIME provided interpretability, with Grad-CAM localization evaluated against segmentation masks via Intersection over Union (IoU). Results: Under lesion-grouped evaluation, the ensemble achieved the highest performance, with 85.76% accuracy, 75.24% macro-F1, and 0.971 macro-AUC. Paired accuracy gains over comparators ranged from 2.36 to 7.57 percentage points, with all lesion-clustered 95% CIs excluding zero. ResNet34 yielded the highest mean Grad-CAM IoU (0.456). Conclusion: The ensemble achieved improved classification performance under lesion-grouped evaluation. Quantitative localization analysis showed measurable spatial overlap with lesion regions, although clinical usefulness requires reader studies and external validation.

Background and Objectives: This study aimed to characterize the clinical spectrum, diagnostic delay, treatment patterns, and outcomes of pediatric mycosis fungoides (MF) and to identify predictors of early response. Methods: This national multicenter retrospective study included patients diagnosed with histopathologically confirmed MF before 18 years of age. Demographic, clinical, histopathologic, immunophenotypic, staging, treatment, and follow‐up data were collected. Early response at 3–6 months (partial/complete response) was analyzed using logistic regression. Results: A total of 133 patients were included; 60.9% (81/133) were male. Median age at diagnosis was 13 years (IQR, 9–15) and median diagnostic delay was 18 months (IQR, 8–48). Classic morphology was recorded in 78.2% (104/133), and hypopigmented MF in 36.1% (48/133); overlapping morphologies were documented in 42.1% (56/133). Stage IA disease was present in 59.4% (79/133). Phototherapy was used in 60.9% (81/133). Early response occurred in 91.0% (121/133). BSA < 10% was independently associated with higher odds of early response (aOR 8.42, 95% CI 1.71–41.54; p = 0.009), whereas older age at diagnosis predicted lower odds (aOR 0.81 per year, 95% CI 0.67–0.98; p = 0.029). Relapse occurred in 28.5% (37/130) within median follow‐up of 22 months (median time to relapse of 8 months); 67.6% (25/37) relapsed within 12 months. Conclusion: In this cohort, pediatric MF presented at an early stage. Despite favorable early responses, relapse occurred in nearly one third, often within the first year, supporting long‐term surveillance. Higher BSA involvement and older age at diagnosis were associated with reduced early response.

Background and Objective Symptomatic dermographism (SD), the most common chronic inducible urticaria subtype, manifests as strip-shaped, pruritic wheals after skin friction. Conclusive data on its associated factors are limited, and direct comparisons between individuals with and without SD remain scarce. We aimed to identify factors associated with SD internationally. Methods The PREVALENCE-D (Prevalence Estimation of Dermographism) study was an international cross-sectional survey conducted from 2021 to 2024 across 19 countries. An expert-designed questionnaire diagnosed SD and assessed potential associated factors. SD participants were defined as those who self-reported chronic recurrent urticarial dermographism with itch. Factors associated with SD were identified by comparing participants with and without SD. Results Of 68,513 participants, 3101 had SD (female 73.3%). Their mean age was 40.2 ± 16.2 years. Key factors associated with SD included atopic dermatitis (adjusted odds ratio [aOR] 4.20, 95% confidence interval [CI] 3.62‒4.88) and allergic rhinitis (aOR 2.11, 95% CI 1.88‒2.37). Participants with at least one atopic condition (allergic rhinitis, atopic dermatitis, or asthma) were significantly more likely to have SD (aOR 2.70, 95% CI 2.47‒2.95). Those with all three atopic conditions had a further increased likelihood of SD (aOR 7.75, 95% CI 5.31‒11.29). Other associations included working and older age groups, female sex, dyslipidemia, and thyroid disease. Conclusions Atopic dermatitis and allergic rhinitis emerged as the strongest correlates of SD, especially those with all three atopic conditions. Thus, allergic comorbidities should be assessed in SD patients. Further research is needed to clarify the pathophysiological relationship between these conditions and SD. [Graphical abstract available online] Graphical Abstract

In this report a case of cutaneous Scedosporium apiospermum infection in 75-year-old woman with asthma who was receiving long-term therapy with corticosteroids was presented. She was hospitalized in the department of chest diseases for the follow-up of worsening dyspnea, cor pulmonale and stasis dermatitis. Erythema, localized tenderness and swelling developed on her left foot dorsum during her follow-up in the hospital. A few days later, black-colored papules with an erythematous base were observed on her foot dorsum and anterior surface of the tibia. She was diagnosed with cellulitis. Skin biopsy revealed numerous septated fungal hyphae. Preemptive amphotericin B was added to the antibacterial treatment. Within 48 h of incubation, the Scedosporium colony was isolated based on conventional methods. The fungal ribosomal RNA gene internal transcribed spacer sequence analysis revealed S.apiospermum and deposited into GenBank with accession number OM948685. The minimum inhibition concentration was defined as 1 µg/mL for amphotericin B; 16 µg/mL for fluconazole; 0.25 µg/mL voriconazole, 4 µg/mL for caspofungin, 1 µg/mL for itraconazole and 0.125 µg/mL for posaconazole on 48 h of incubation. The patient was transferred to the intensive care unit due to the deterioration of her general condition with the diagnosis of sepsis. Antimicrobial treatment was modified to caspofungin/ voriconazole combination and imipenem administration. Hepatic failure, bone marrow suppression with severe thrombocytopenia and disseminated intravascular coagulation syndrome developed on the third day of intensive care unit stay. The patient died of severe septic shock and multiorgan failure. We also reviewed the literature presenting similar cutaneous scedosporiosis cases and discussed the results.

Pediyatrik hastalarda dermatolojik cerrahi işlemler, tanısal ve tedavi edici amaçlarla sık uygulanmaktadır. Bu işlemler sonrası egzersize dönüş planlaması, sadece yara iyileşmesini değil, çocuğun psikososyal uyumunu da yakından ilgilendirir. Çocuklarda uzun süreli fiziksel kısıtlamalar motivasyon kaybı, irritabilite ve sosyal izolasyona yol açabileceğinden, egzersize dönüş protokolleri yaş, işlem tipi, sütur varlığı ve anatomik bölgeye göre bireyselleştirilmelidir. Punch biyopsi, eksizyonel biyopsi, tırnak cerrahisi ve lazer/kriyoterapi sonrası egzersiz kısıtlamaları farklılık göstermekte olup, özellikle sütur hattının kapanması ve komplikasyon risklerinin azalması beklenmelidir. Çocuk hastaların aileleri, yara bakımı, komplikasyon belirtileri ve egzersize dönüş süreleri konusunda bilgilendirilmeli; hekim-aile iş birliği ile kademeli bir egzersiz programı uygulanmalıdır.

Background: Histamine plays a central role in CSU pathogenesis and may contribute to extracutaneous symptoms such as dysmenorrhea, but this relationship has not been fully examined. Methods: In this prospective, cross‐sectional multicenter study, 425 female CSU patients and 370 age‐matched controls were recruited from 19 UCARE centers in Türkiye. Dysmenorrhea prevalence and severity were assessed using a numerical scale and symptom scores. Histamine skin prick testing (SPT) was performed to evaluate histamine clearance capacity. Associations between dysmenorrhea, CSU phenotype, antihistamine use, and SPT kinetics were analyzed. Results: The prevalence and severity of dysmenorrhea were comparable between CSU patients and controls (81.6% vs. 79.5%, p = 0.244). However, 95/418 (22.7%) of CSU patients reported increased menstrual pain following CSU onset, and 20/95 (21.1%) of these reported partial relief with antihistamines. CSU patients with worsened dysmenorrhea exhibited significantly shorter histamine SPT positivity durations than those without symptom change (35 vs. 45 min; p = 0.015). No correlation was observed between dysmenorrhea severity and total IgE, anti‐TPO levels, or urticaria control/activity scores. A moderate negative correlation was found between UAS7 and elapsed time to histamine SPT negativity ( r = −0.389, p < 0.001). Conclusions: Although overall prevalence and severity of dysmenorrhea were similar in CSU patients and healthy controls, about one in five patients with CSU experienced worsening menstrual pain after disease onset, and only a subset of these patients reported partial improvement with antihistamines. A distinct CSU subgroup with heightened menstrual pain may be characterized by altered histamine dynamics. Whether these patients may benefit from more intensive antihistamine treatment remains uncertain and requires confirmation in future studies; further studies should explore neuroimmune and hormonal mechanisms underlying this overlap.