Intraocular foreign bodies (IOFBs) are an important cause of visual loss within the group of working age population. We aim to present the clinical features and the algorithm according to which we manage the foreign bodies that are located in the posterior segment of the eye. We define the outcomes and the prognostic factors that influenced the final visual acuity and globe survival in patients with IOFBs that we extracted by pars plana vitrectomy (PPV) over a 5-year period. We reviewed the medical records of all the cases with IOFBs that we removed by PPV, over 5 years (2009–2013). We extracted the following parameters: age, gender, wound anatomy, IOFB characteristics, ocular lesions, initial and final visual acuities. We used the program SPSS version 20.0.0. for the statistical analysis of our data. During 5 years, we treated 21 IOFBs by PPV, representing 12.20 % of all the open globe injuries. All the patients were males with the median age of 36 years. The foreign body was located in the vitreous - 11 cases (52.38 %), retina - seven cases (33.33 %) and perforating - three cases (14.28 %). Retinal detachment (RD) at presentation was identified in eight cases (38.09 %) and endophthalmitis, in six cases (28.57 %). The visual outcome was significantly worse in patients with RD at presentation (p = 0.012) and with IOFBs larger than 3 mm (p = 0.042). Endophthalmitis did not influence the visual outcome. The worse prognostic factors were: RD at presentation and large foreign body. IRCT2015040418966N3 / Apr. 9/2015

Digital twins (DTHs) and virtual twins (VTHs) in healthcare represent emerging technologies towards precision medicine, providing opportunities for patient-centric healthcare. Our scoping review aimed to map the current DTH and VTH technologies in oncology, summarize their technical solutions, and assess their credibility. A systematic search was conducted in the main bibliographic databases, identifying 441 records, of which 30 were included. The studies covered a wide range of cancers, including breast, lung, colorectal, and gastrointestinal malignancies, with DTH and VTH applications focusing on diagnosis, therapy, and monitoring. The results revealed heterogeneity in targeted topics, technical approaches, and outcomes. Most twining solutions use synthetic or limited real-world data, raising concerns regarding their reliability. Few studies have integrated real-time data and machine learning for predictive modeling. Technical challenges include data integration, scalability, and ethical considerations, such as data privacy and security. Moreover, the evidence lacks sufficient clinical validation, with only partial credibility in most cases. Our findings underscore the need for multidisciplinary collaboration among end-users and developers to address the technical and ethical challenges of DTH and VTH systems. Although promising for the future of personalized oncology, substantial steps are required to move beyond experimental frameworks and to achieve clinical implementation.

Background: Melanoma patients stop responding to targeted therapies mainly due to mitogen activated protein kinase (MAPK) pathway re-activation, phosphoinositide 3 kinase/the mechanistic target of rapamycin (PI3K/mTOR) pathway activation or stromal cell influence. The future of melanoma treatment lies in combinational approaches. To address this, our in vitro study evaluated if lower concentrations of Celecoxib (IC₅₀ in nM range) could still preserve the chemopreventive effect on melanoma cells treated with trametinib. Materials and Methods: All experiments were conducted on SK-MEL-28 human melanoma cells and BJ human fibroblasts, used as co-culture. Co-culture cells were subjected to a celecoxib and trametinib drug combination for 72 h. We focused on the evaluation of cell death mechanisms, melanogenesis, angiogenesis, inflammation and resistance pathways. Results: Low-dose celecoxib significantly enhanced the melanoma response to trametinib. The therapeutic combination reduced nuclear transcription factor (NF)–kB (p < 0.0001) and caspase-8/caspase-3 activation (p < 0.0001), inhibited microphthalmia transcription factor (MITF) and tyrosinase (p < 0.05) expression and strongly down-regulated the phosphatidylinositol-3-kinase/protein kinase B (PI3K/AKT) signaling pathway more significantly than the control or trametinib group (p < 0.0001). Conclusion: Low concentrations of celecoxib (IC₅₀ in nM range) sufficed to exert antineoplastic capabilities and enhanced the therapeutic response of metastatic melanoma treated with trametinib.

Background: In recent years, numerous artificial intelligence applications, especially generative large language models, have evolved in the medical field. This study conducted a structured comparative analysis of four leading generative large language models (LLMs)—ChatGPT-4o (OpenAI), Grok-3 (xAI), Gemini-2.0 Flash (Google), and DeepSeek-V3 (DeepSeek)—to evaluate their diagnostic performance in clinical case scenarios. Methods: We assessed medical knowledge recall and clinical reasoning capabilities through staged, progressively complex cases, with responses graded by expert raters using a 0–5 scale. Results: All models performed better on knowledge-based questions than on reasoning tasks, highlighting the ongoing limitations in contextual diagnostic synthesis. Overall, DeepSeek outperformed the other models, achieving significantly higher scores across all evaluation dimensions (p < 0.05), particularly in regards to medical reasoning tasks. Conclusions: While these findings support the feasibility of using LLMs for medical training and decision support, the study emphasizes the need for improved interpretability, prompt optimization, and rigorous benchmarking to ensure clinical reliability. This structured, comparative approach contributes to ongoing efforts to establish standardized evaluation frameworks for integrating LLMs into diagnostic workflows.

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.

The present study aimed to evaluate the students’ progress in evaluating ultrasound (US) and cadaveric cardiac images and long-time retention of information. First-year medical students were invited to participate in four two-hour online lectures during one week voluntarily. The students were trained to recognize cardiovascular anatomical structures on US and cadaveric images during the intervention. The participants’ abilities to identify specific anatomical structures were tested before, immediately after and six months after the training. A group of second-year students without US training participated as a control group and filled the same test once. Ninety-one first-year students agreed to participate, and forty-nine completed all three tests. The performances in the correct identification of cardiovascular structures on the US images significantly improved after the training but significantly decreased after six months. In the intervention group, the accurate identification of cardiovascular structures was significantly higher on cadaveric images (80% vs. 53%, p-value < 0.0001, n = 91 at post-training; 70% vs. 33%, p-value < 0.0001, n = 49 at 6 months after training). The correct answers percentage score in the control group varied from 6.7% to 66.7% for US cardiovascular anatomical without a significant difference than the intervention group (p-value = 0.7651). First-year students’ knowledge of heart US anatomy proved less effective than cadaveric images, significantly improved after training and decreased over time, indicating the need for repetition reinforcement.