Residency training in medicine lays the foundation for future medical doctors. In real-world settings, training centers face challenges in trying to create balanced residency programs, with cases encountered by residents not always being fairly distributed among them. In recent years, there has been a tremendous advancement in developing artificial intelligence (AI)-based algorithms with human expert guidance for medical imaging segmentation, classification, and prediction. In this paper, we turned our attention from training machines to letting them train us and developed an AI framework for personalised case-based ophthalmology residency training. The framework is built on two components: (1) a deep learning (DL) model and (2) an expert-system-powered case allocation algorithm. The DL model is trained on publicly available datasets by means of contrastive learning and can classify retinal diseases from color fundus photographs (CFPs). Patients visiting the retina clinic will have a CFP performed and afterward, the image will be interpreted by the DL model, which will give a presumptive diagnosis. This diagnosis is then passed to a case allocation algorithm which selects the resident who would most benefit from the specific case, based on their case history and performance. At the end of each case, the attending expert physician assesses the resident’s performance based on standardised examination files, and the results are immediately updated in their portfolio. Our approach provides a structure for future precision medical education in ophthalmology.

Background/Objectives: Pyoderma gangrenosum (PG) is a rare, autoimmune skin condition characterized by painful, rapidly progressing ulcers, often associated with autoimmune dysregulation. Managing PG following breast surgery presents unique challenges due to its pathergy phenomenon, which complicates surgical interventions. This article outlines the case of PG in a 48-year-old female post-breast surgery and reviews management strategies through a systematic analysis of the literature. Methods: A systematic literature review from 2018 to 2023 identified 24 relevant articles on PG management post-breast surgery. The studies were analyzed to compare the efficacy and complications of conservative versus combined (conservative and surgical) treatment strategies. Results: Results indicate that while conservative management, primarily with corticosteroids, remains preferred, combined strategies, including systemic therapies, vacuum-assisted closure, and surgery, offer significant benefits in select cases. Conclusions: Our findings suggest that a personalized, multifaceted treatment plan is crucial for managing PG effectively, emphasizing the need for early detection, meticulous planning, and comprehensive care to optimize patient outcomes.

Language and cultural differences could be a limiting factor for the international exchange of Virtual Patients (VPs), especially for small countries and languages of limited circulation. Our research evaluated whether it would be feasible to develop a VP based educational program in our Romanian institution, with cases in English and developed in a non-Romanian setting.

Background: Parathyroid neoplasia is a heterogeneous group of tumors, including parathyroid adenoma (PA), atypical parathyroid tumors (aPTs), and parathyroid carcinoma (PC). Differential diagnosis, especially preoperatively, between parathyroid carcinoma and the other two entities is challenging. The purposes of this study were to highlight the main differences between different parathyroid tumors and to evaluate how combined PC suspicion and intraoperative adjuncts can influence surgical decision-making and outcome-related issues. Methods: We performed a retrospective study of a database of patients diagnosed with parathyroid tumors who underwent surgical treatment at our endocrine surgery referral center between June 2019 and July 2024. Demographic, clinical, biochemical, imaging, intraoperative, immunohistochemical, and follow-up data were analyzed. Results: A total of 83 cases were included in our study, divided for analysis into PA (n = 67), aPT (n = 9) and PC (n = 7) subgroups. The clinical profile of the cohort showed a significant difference (p < 0.05) between the PA, aPT, and PC subgroups regarding the presence of palpable tumors (0% vs. 11.11% vs. 14.29%), both bone and kidney involvement (14.93% vs. 44.44% vs. 85.71%), and extensive disease beyond bone and kidney involvement (4.48% vs. 44.44% vs. 71.43%). PTH levels over five times the normal value were present at significantly different rates (p < 0.001), with higher rates in the aPT and PC subgroups (55.56% and 85.71%, respectively) compared with the PA subgroup (7.46%). Also, a significant difference (p < 0.001) was observed when analyzing extreme albumin-corrected serum calcium elevations over 14 mg/dL, with much higher rates in the PC subgroup (71.43%) compared to PA (1.49%) and aPT (33.33%). On preoperative ultrasonography, a significantly higher number of PCs presented diameters ≥ 3 cm (p < 0.001), depth-to-width ratios (D/W) ≥ 1 (p = 0.003), suspicious delineation (p < 0.001), and suspicious echotexture features (p < 0.001), compared to PAs. On preoperative US performed by the surgeon, suspicious features for thyroid cancer were identified in five more patients compared to the four identified by the initial US evaluation, and all (10.84% of all patients) were confirmed on final histopathology as papillary thyroid cancers. Intraoperatively, a significant difference (p < 0.001) regarding parathyroid macroscopic suspicious features, including adhesions to the thyroid gland, was seen between subgroups. When analyzing only cases with en bloc resection, we found that, in all PC cases, a combined preoperative suspicion was present, and in five cases an intraoperative suspicion was raised. Immunohistochemical data showed significantly different median Ki-67 indices between subgroups (1, 2, and 5; p = 0.008) and a different parafibromin staining profile between PC and aPT. Regarding intraoperative neuromonitoring use, a significantly lower incidence of voice changes related to the external branch of the superior laryngeal nerve was observed in the monitoring vs. non-monitoring group (57.14% vs. 12.5%, p = 0.019). Conclusions: Our findings confirm that, in a multimodal and combined diagnostic approach, early pre- and intraoperative PC suspicion can be raised in order to optimize surgical treatment and, thus, favorably influence the outcome. Utilizing all resources available, including intraoperative parathormone determination, laryngeal nerve neuromonitoring, and immunohistochemistry staining, can bring extra benefit to the management of these challenging cases.