Background/Objectives: Large language models (LLMs) have shown significant potential to transform various aspects of healthcare. This review aims to explore the current applications, challenges, and future prospects of LLMs in medical education, clinical decision support, and healthcare administration. Methods: A comprehensive literature review was conducted, examining the applications of LLMs across the three key domains. The analysis included their performance, challenges, and advancements, with a focus on techniques like retrieval-augmented generation (RAG). Results: In medical education, LLMs show promise as virtual patients, personalized tutors, and tools for generating study materials. Some models have outperformed junior trainees in specific medical knowledge assessments. Concerning clinical decision support, LLMs exhibit potential in diagnostic assistance, treatment recommendations, and medical knowledge retrieval, though performance varies across specialties and tasks. In healthcare administration, LLMs effectively automate tasks like clinical note summarization, data extraction, and report generation, potentially reducing administrative burdens on healthcare professionals. Despite their promise, challenges persist, including hallucination mitigation, addressing biases, and ensuring patient privacy and data security. Conclusions: LLMs have transformative potential in medicine but require careful integration into healthcare settings. Ethical considerations, regulatory challenges, and interdisciplinary collaboration between AI developers and healthcare professionals are essential. Future advancements in LLM performance and reliability through techniques such as RAG, fine-tuning, and reinforcement learning will be critical to ensuring patient safety and improving healthcare delivery.

Obesity is a disorder identified by an inappropriate increase in weight in relation to height and is considered by many international health institutions to be a major pandemic of the 21st century. The gut microbial ecosystem impacts obesity in multiple ways that yield downstream metabolic consequences, such as affecting systemic inflammation, immune response, and energy harvest, but also the gut–host interface. Metabolomics, a systematized study of low-molecular-weight molecules that take part in metabolic pathways, represents a serviceable method for elucidation of the crosstalk between hosts’ metabolism and gut microbiota. In the present review, we confer about clinical and preclinical studies exploring the association of obesity and related metabolic disorders with various gut microbiome profiles, and the effects of several dietary interventions on gut microbiome composition and the metabolome. It is well established that various nutritional interventions may serve as an efficient therapeutic approach to support weight loss in obese individuals, yet no agreement exists in regard to the most effective dietary protocol, both in the short and long term. However, metabolite profiling and the gut microbiota composition might represent an opportunity to methodically establish predictors for obesity control that are relatively simple to measure in comparison to traditional approaches, and it may also present a tool to determine the optimal nutritional intervention to ameliorate obesity in an individual. Nevertheless, a lack of adequately powered randomized trials impedes the application of observations to clinical practice.

A specific diet regimen is a promising way of managing inflammatory bowel disease (IBD), with the Mediterranean diet (MD) being a likely candidate due to its potential to modulate gut inflammation. Therefore, the aim of this study was to investigate nutritional habits and dietary attitudes of IBD patients, and to assess their adherence to the Mediterranean diet. The study enrolled 50 Crohn’s disease and 44 ulcerative colitis patients, with clinical and laboratory parameters taken. Dietary attitudes were examined, and adherence to MD was assessed using the Mediterranean Diet Service Score (MDSS). Average MDSS score was 6.0 (5.0–7.0), while only nine participants fulfilled criteria for Mediterranean diet adherence. Moreover, all of them were men (p = 0.021). Low percentage of adherence to recommended guidelines was observed for eating olive oil (25.5%), fresh fruit (14.9%), and vegetables (10.6%). Significant positive correlation was observed between total MDSS points and high-density lipoprotein (HDL) cholesterol levels (p = 0.002). The majority of the patients (86.2%) considered that a more controlled diet could reduce their IBD symptoms, while 17% visited a nutritionist for diet advice. The majority of patients (84%) would visit educational programs regarding nutrition. In conclusion, adherence to MD was very low, while IBD patients were willing to extend their nutritional knowledge if proper educational programs were organized.

With the rising global burden of inflammatory bowel disease (IBD) and the rising costs of novel biological drugs, there is an increasing need for dietary approaches and functional foods that could modulate the course of IBD. The Mediterranean diet has proven to be efficacious in managing chronic inflammatory diseases, and recent studies have also shown its benefits in the setting of IBD. Since olive oil and its compounds have been shown to provide a considerable anti-inflammatory effect, in this review, we aim to discuss the latest evidence concerning the impact of olive oil and its bioactive compounds on IBD. Numerous preclinical studies have exhibited solid evidence on the mechanisms by which polyphenol-rich extra-virgin olive oil (EVOO) or specific polyphenols like hydroxytyrosol (HT) provide their anti-inflammatory, antioxidative, antitumour, and microbiota-modulation effects. Accordingly, several human studies that explored the effects of olive oil on patients with IBD further confirmed the evidence brought forward by preclinical studies. Nevertheless, there is a need for larger-scale, multicentric, randomized control trials that would finally elucidate olive oil’s level of efficacy in modulating the course of IBD.

In recent years, sodium-glucose co-transporter 2 inhibitors (SGLT2i) have demonstrated beneficial renoprotective effects, which culminated in the recent approval of their use for patients with chronic kidney disease (CKD), following a similar path to one they had already crossed due to their cardioprotective effects, meaning that SGLT2i represent a cornerstone of heart failure therapy. In the present review, we aimed to discuss the pathophysiological mechanisms operating in CKD that are targeted with SGLT2i, either directly or indirectly. Furthermore, we presented clinical evidence of SGLT2i in CKD with respect to the presence of diabetes mellitus. Despite initial safety concerns with regard to euglycemic diabetic ketoacidosis and transient decline in glomerular filtration rate, the accumulating clinical data are reassuring. In summary, although SGLT2i provide clinicians with an exciting new treatment option for patients with CKD, further research is needed to determine which subgroups of patients with CKD will benefit the most, and which the least, from this therapeutical option.

Background: Due to recent changes in breast cancer treatment strategy, significantly more patients are treated with neoadjuvant systemic therapy (NST). Radiological methods do not precisely determine axillary lymph node status, with up to 30% of patients being misdiagnosed. Hence, supplementary methods for lymph node status assessment are needed. This study aimed to apply and evaluate machine learning models on clinicopathological data, with a focus on patients meeting NST criteria, for lymph node metastasis prediction. Methods: From the total breast cancer patient data (n = 8381), 719 patients were identified as eligible for NST. Machine learning models were applied for the NST-criteria group and the total study population. Model explainability was obtained by calculating Shapley values. Results: In the NST-criteria group, random forest achieved the highest performance (AUC: 0.793 [0.713, 0.865]), while in the total study population, XGBoost performed the best (AUC: 0.762 [0.726, 0.795]). Shapley values identified tumor size, Ki-67, and patient age as the most important predictors. Conclusion: Tree-based models achieve a good performance in assessing lymph node status. Such models can lead to more accurate disease stage prediction and consecutively better treatment selection, especially for NST patients where radiological and clinical findings are often the only way of lymph node assessment.