Uvod/Cilj:Cilj rada je da predstavi metodološki okvir za desetogodišnje predviđanje multimorbiditeta pomoću generativnog, sekvencijalnog modela zasnovanog na transformer arhitekturi, uz potpuno oslanjanje na javno dostupne i sintetičke podatke, bez obrade identifikabilnih podataka pacijenata. Metode:Sprovedena je metodološka studija sa primarnom obukom i evaluacijom na Synthea sintetičkoj kohorti (≥100.000 zapisa), uz tehničke provere na MIMIC-III (de-identifikovani realni podaci intenzivne nege). Ishodi (≥1.000) su definisani mapiranjem ICD-10/ICD-10-CM kodova u PheCode kategorije. Model je transformer sa multitask izlazima (po jedan za svaki cilj) i vremenskim ugradnjama, sa evaluacijom diskriminacije (AUPRC, AUROC), kalibracije (Brier, intercept/nagib) i kliničke korisnosti (Decision Curve Analysis). Sprovedene su analize osetljivosti i osnovne provere pravičnosti (pol, starost). Rezultati:Model je postigao najbolje rezultate u kardiometaboličkom i onkološkom domenu, umerene u respiratornim/bubrežnim, a skromnije u mentalnim/infektivnim ishodima. Kalibracija je bila dobra u srednjim opsezima rizika; DCA je pokazala pozitivnu neto korist u pragovima relevantnim za oportunistički skrining (≈5-15% desetogodišnjeg rizika). Analize osetljivosti potvrdile su stabilnost rang-performansi na promenama pragova retkosti i dužine istorije, bez dokaza o značajnom "label leakage"-u. Zaključak:Prikazan je reproducibilan i etički prihvatljiv pristup dugoročnoj višebolećinskoj prognozi rizika koristeći generativni transformer na javnim/sintetičkim skupovima. Ovaj "barometar zdravlja" može da podrži trijažu i personalizovanu prevenciju, uz preporuku obavezne eksterne validacije, lokalne re-kalibracije i nadzora pravičnosti pre kliničke primene.
Integracija zdravstvene informatike u pametne gradove predstavlja transformativni pristup rešavanju izazova urbanih zdravstvenih sistema. Ovaj rad ispituje potencijal zdravstvene informatike za unapređenje zdravstvene usluge, optimizaciju javnog zdravlja i rešavanje neefikasnosti u urbanim sredinama, uz identifikaciju prepreka u njenoj primeni. Uvod analizira razvoj pametnih gradova i pozicionira zdravstvenu informatiku kao ključni alat za upravljanje urbanim zdravstvenim potrebama, uključujući starenje populacije i rastuću učestalost hroničnih bolesti. Teorijski okvir i pregled literature istražuje tehnologije kao što su IoT, Veliki podaci i veštačka inteligencija, uz primere iz Singapura i Barselone, gde su ove tehnologije poboljšale zdravstvene ishode. Metodologija koristi mešoviti pristup koji obuhvata sistematski pregled literature i analizu studija slučaja u gradovima poput Amsterdama i Tokija. Rezultati ukazuju na smanjenje vremena odgovora hitne pomoći za 40% i hospitalizacija za hronične bolesti za 25%, dok izazovi kao što su privatnost podataka, visoki troškovi i interoperabilnost i dalje predstavljaju prepreke. Diskusija ističe značaj okvira za zaštitu privatnosti, partnerstava između javnog i privatnog sektora i inkluzivnih strategija za povećanje digitalne pismenosti. Rad zaključuje potencijalom zdravstvene informatike za izgradnju otpornijih i efikasnijih urbanih zdravstvenih sistema, uz preporuke za dalja istraživanja.
ACTH zavisni aldosteronizam (Glucocorticoid remediable aldosteronism - GRA) predstavlja retku, autozomno-dominantnu formu primarnog aldosteronizma koja nastaje usled himerizacije gena CYP11B1/CYP11B2 i dovodi do povećane produkcije aldosterona. Supresija aldosterona deksametazonom može ukazivati na ovaj poremećaj, dok genetska analiza ostaje zlatni standard dijagnostike. Prikazana je pacijentkinja sa dugogodišnjom arterijskom hipertenzijom koja je dijagnostikovana u ranoj odrasloj dobi. Zabeležen je povišen bazalni aldosteron uz granične vrednosti odnosa aldosteron/plazma reninska aktivnost. Radiološka dijagnostika nije ukazala na morfološke promene nadbubrežnih žlezda. Učinjeni su potvrdni testovi (kaptoprilski test, infuzioni test, DST). Deksametasonski supresioni test pokazao je značajnu supresiju aldosterona, što je pobudilo sumnju na GRA i dovelo do indikacije za genetsko testiranje. Diferencijalna dijagnoza hiperaldosteronizma obuhvata adenom, hiperplaziju, ektopičnu produkciju i genetske oblike kao što je GRA. Biohemijska supresija aldosterona može ukazivati na familijarni oblik aldosteronizma tip I, ali nije dovoljna za potvrdu dijagnoze. Genetska analiza omogućava definitivno razlikovanje GRA od drugih oblika primarnog aldosteronizma. Kod pacijentkinje je, uprkos biohemijskim nalazima koji su delimično sugerisali GRA, genetskim testiranjem isključeno prisustvo CYP11B1/CYP11B2 himernog gena. Ovaj slučaj ukazuje na važnost genetske potvrde kod sumnje na GRA, naročito u mlađih bolesnika sa opterećenjem u porodičnoj anamnezi.
A 47-year-old female patient with long-standing arterial hypertension and a history of hypertension during pregnancies was referred for evaluation of possible secondary aldosteronism. Initial tests conducted while the patient was on diuretic therapy showed elevated aldosterone and renin levels, suggestive of renin-dependent aldosteronism. Given the potential influence of therapy, a medication washout was performed, and hormonal testing was repeated. The new results showed normal aldosterone and renin levels without hypokalemia, while blood pressure remained stable. The absence of other pathological findings and stable biochemical parameters indicated that the previous hormonal imbalance was most likely due to relative hypovolemia induced by diuretics. Other causes of secondary aldosteronism were ruled out. This case highlights the importance of proper patient preparation for endocrine testing and the need to exclude medications that may affect the interpretation of the RAAS axis. In cases requiring aldosterone-neutral therapy, non-dihydropyridine calcium channel blockers and Alpha-adrenergic blockers are preferred. If heart rate is elevated, verapamil can be used.
Objective: To systematically review available digital tools for self-monitoring of thyroid diseases, with a focus on functionality, quality, and clinical validation. Methods: A systematic literature review was conducted following PRISMA guidelines, using the PubMed, Scopus, Web of Science, and IEEE Xplore databases for the period 2010-2025. The analysis included mobile applications, web platforms, and wearable devices designed for adults. The quality of the tools was assessed using the uMARS scale, which evaluates engagement, functionality, aesthetics, and information quality. Results: A total of 18 digital tools were identified. The most common features included symptom tracking (78%), laboratory result monitoring (67%), and medication logging (56%). The average uMARS score ranged from 3.8 to 4.5 (out of a maximum of 5). Only 28% of the tools were medically validated, while five were integrated with healthcare systems. The main shortcomings included the lack of personalized algorithms (22%) and unclear data protection policies (61%). Discussion: These tools enable real-time monitoring but lack standardization, clinical validation, and interoperability. Multidisciplinary collaboration between clinicians, developers, and regulatory bodies is essential. Conclusion: Digital solutions have the potential to improve self-monitoring of thyroid diseases but require more rigorous validation studies, better data protection, and personalized approaches.
Background: Telemedicine is becoming an increasingly important component of healthcare, particularly in palliative care, where it contributes to improving access, continuity, and personalization of services. For patients with advanced endocrine diseases, the need for continuous monitoring and support makes this technology especially relevant. Objective: The aim of this paper was to identify the key benefits and challenges of implementing telemedicine in palliative endocrine care and to develop practice-based recommendations. Methods: A systematic review of 24 publications published between 2015 and 2024 was conducted, focusing on telemedicine within the context of palliative or endocrine care. The findings related to effectiveness, acceptability, and implementation challenges were analyzed. Results: The most commonly identified benefits of telemedicine included improved access to healthcare services (83.3%), more effective symptom monitoring (72.2%), increased patient satisfaction (66.7%), and reduced need for hospitalization (55.6%). Key challenges included technical issues and limited access to technology (77.8%), insufficient staff training (61.1%), and ethical-legal concerns (50%). Recommendations include staff education, development of specialized platforms, and enhancement of the regulatory framework. Conclusion: Telemedicine holds significant potential to improve the quality of palliative care in endocrine patients. However, successful implementation requires a comprehensive approach involving technical infrastructure, staff training, and supportive healthcare policies.
The way of reacting to potentially stressful situations is individual and depends on two basic factors. The first is the way a person perceives and interprets the situation, and the second factor relates to the state of the body itself, or the organism. Chronic stress, burnout syndrome and post-traumatic stress disorder often lead to the development of diseases. Stress can affect health directly, through the autonomic and endocrine reactions it causes, but also indirectly, through changes in health behavior that can occur due to stress. Physical exercise is often recommended as part of a stress management program. It has been proven that exercise reduces stress hormones and stress reactivity. Exercises adapted to the degree of physical capabilities and the levels of motivation of the person are considered an excellent initial approach to managing psychological stress. Physical activity is important not only as a primary prevention of many chronic diseases, but also as a secondary prevention that slows down and reduces the symptoms of chronic diseases. In addition to its impact on chronic diseases, physical activity also has a beneficial effect on improving self-confidence, social skills, cognitive functioning, and reducing symptoms of stress, which, together with other positive effects, contributes to a better quality of life. The results of existing research show that physical exercise can affect cognitive functioning throughout the lifespan directly, through physiological mechanisms and structural changes in the brain, and indirectly through its impact on mood and stress reduction. Physical exercise, through various mechanisms, has a strong impact on the manifestation of stress levels.
The differential diagnosis of polydipsia and polyuria represents a challenge in clinical practice, as a wide spectrum of diseases and disorders can lead to this syndrome. The most common causes include diabetes mellitus, pituitary disorders, electrolyte imbalance, renal diseases, psychogenic conditions, and drug effects. Our patient developed a polyuric-polydipsic syndrome following embolization and stent implantation of an aneurysm of the left internal carotid artery (ICA). Computed tomography (CT) of the brain showed no pathological changes. Diabetes insipidus (DI) was initially suspected. Over the following year, the daily water balance remained unchanged at 4-5 L/24 h. At our clinic, endocrinological evaluation was performed. The daily water balance with free fluid intake was approximately 4.5 L/24 h, with no electrolyte imbalance detected. Upon fluid restriction, a gradual increase in urine osmolality was observed, with normal urine specific gravity and consistently normal serum osmolality. An infusion test demonstrated maintenance of normal serum electrolytes and osmolality, with an adequate copeptin response. MRI of the sellar region showed no pathological substrate. In primary polydipsia, fluid restriction leads to increased urine osmolality, as seen in our patient, whereas urine osmolality remains low in central and nephrogenic DI. Measurement of basal copeptin is a reliable method for diagnosing nephrogenic DI, and copeptin measurement after a hypertonic saline infusion test helps differentiate central DI from primary polydipsia.
Background: Fragmented care pathways for thyroid nodular disease prolong time to decision and increase costs. “Onestop” clinics (OSCs) integrate clinical evaluation, ultrasound, ultrasoundguided fineneedle aspiration (FNA) and, when appropriate, rapid onsite evaluation (ROSE)/telecytology in a single visit. Objective: To map organizational models of OSCs, define key performance indicators (KPIs), and summarize outcomes (time, visits, adequacy, repeat FNA, costs, patient satisfaction, safety). Methods: Scoping review following PRISMAScR and JBI guidance. MEDLINE, Scopus, and Web of Science were searched (2000–August 2025). We included studies that operationalized OSCs and/or reported KPIs/outcomes. Data extraction covered organizational features, protocols (ACR/EUTIRADS, Bethesda), flow metrics, FNA/ROSE adequacy, economics, satisfaction, and safety. Narrative synthesis was performed. Results: Identified OSC models consistently shorten lead time to decision and reduce the number of visits. Sample adequacy is high–especially with ROSE/telecytology–thereby lowering repeat FNA rates. Patient satisfaction is high; the safety profile of FNA remains favourable. Economic analyses indicate that the costeffectiveness of ROSE is contextdependent and greatest when baseline inadequacy is higher and/or ROSE costs are lower. A KPI set is proposed: lead time, proportion of “singlevisit” completions, Bethesda I rate, ROSE utilization, repeat FNA ≤ 90 days, cost per episode, and satisfaction. Conclusion: OSCs are an applicable, valueoriented model for thyroid diagnostics. Selective use of ROSE/telecytology and KPIdriven management enable efficient and safe implementation across diverse resource settings.
This study investigates how sick leave and socio-demographic and occupational factors affect the choice of general practitioners in Serbia. The objective was to identify the factors influencing the choice of a general practitioner among employees in Serbia, considering incidences of sick leave. The study analyzed data from the 2019 Serbian National Health Survey, including 4,652 participants aged 18 to 65 years, using descriptive statistics and binary logistic regression. Overall, 92.4% of participants, mostly employees in the public sector (85.2%), chose a general practitioner, with a higher prevalence among episodes of prolonged sick leave (92.8%). The choice of a general practitioner is influenced by socio-demographic and health-related factors. Being employed by an employer leads to a more frequent choice of a general practitioner (OR = 2.277). Blue-collar workers (OR = 0.757) and employees of middle and poor wealth status are less likely to choose a general practitioner among participants without sick leave (middle OR = 0.709, poor OR = 0.701). Employees in Vojvodina with sick leave are significantly less likely to choose a general practitioner compared to Belgrade (OR = 0.111). The last visit to a general practitioner is a significant predictor for both groups. Socio-demographic, work-related factors, and health status significantly influence the choice of general practitioner among employees in Serbia, indicating the importance of accessible primary healthcare during periods of sick leave.
Introduction: Rapid and accurate assessment of traumatic injury severity is crucial for effective triage and timely intervention in emergency medicine. Mobile applications based on artificial intelligence (AI) offer an objective assessment of injury severity through photographic analysis; however, their accuracy has not been sufficiently explored. Objectives: To evaluate the accuracy of available AI-based mobile applications for assessing traumatic injury severity in simulated conditions. Methods: This simulation study tested five mobile applications (DermaScore AI, SkinVision, Tissue Analytics, WoundCheck AI, and BurnCare App). A total of 200 simulated images of traumatic injuries, classified by the Abbreviated Injury Scale (AIS) as mild, moderate, severe, and critical, were analyzed by each application. Accuracy, sensitivity, specificity, and ROC analysis were evaluated, along with the impact of photo resolution on app performance. Results: DermaScore AI achieved the highest overall accuracy (89%), sensitivity (92%), and ROC-AUC value (0.91). The lowest accuracy was recorded by BurnCare App (74%). Higher photo resolution (above 12 MP) significantly improved the accuracy of all tested apps (p = 0.0014). Conclusion: AI-based mobile applications can reliably assess traumatic injury severity from photographic analysis, but their performance significantly varies depending on technical and algorithmic factors. Additional clinical research is required to validate these findings in real-world settings.
Thyroid Eye Disease (TED), formerly known as Graves' orbitopathy, is an autoimmune disorder affecting orbital tissues. The Clinical Activity Score (CAS) is used to evaluate TED activity. A score ≥3 indicates active disease, warranting immunomodulatory therapy. Intravenous corticosteroids (IVCS) are the first-line treatment. However, studies show 20-30% of patients with moderately severe active TED have an inadequate response to corticosteroids. For patients with poor response or tolerance to steroids, tocilizumab offers an effective alternative. We present a case of a 49-year-old female with moderately severe, active, corticosteroid-resistant TED. After failure of two corticosteroid regimens (4.5 g and 5 g), biological therapy with tocilizumab (6 cycles of 600 mg every 4 weeks) was initiated. Clinical response was monitored using NOSPECS, CAS, and the Gorman-Bahneman diplopia classification. This case confirms the potential efficacy of tocilizumab in managing corticosteroid-resistant, moderately severe active TED.
Introduction: Pituitary macroadenomas pose a challenge in clinical endocrinology due to their impact on hormonal balance and subsequent clinical complications. Traditional diagnostic methods often suffer from subjectivity, highlighting the need for a more objective approach. Materials and Methods: This study was conducted as a retrospective secondary analysis of publicly available, de-identified data. Digital histopathological images were obtained from a digital pathology repository, while RNA-seq data, including PIT1 gene expression, were retrieved from the NCBI GEO database. Convolutional neural networks (CNN) were applied for tumor tissue segmentation and classification, while differential expression analysis was performed using DESeq2. Results: The model achieved an accuracy of 92.3% in identifying tumor regions, while bioinformatics analysis revealed a significant upregulation of PIT1 expression in adenomas with more pronounced clinical symptoms (log2FC = 1.8, p < 0.01). Integrated analysis confirmed a strong correlation between morphological patterns and PIT1 expression levels, while regression analysis indicated that this gene is an independent predictor of clinical outcomes. Discussion and Conclusion: The integration of digital pathology and bioinformatics analysis has shown promise in improving the diagnosis and classification of pituitary macroadenomas, paving the way for personalized therapy. Further studies on more heterogeneous samples could further validate the utility of this multidisciplinary approach.
