Sanem Kayhan

Background/Objectives: Vitamin B12 (cobalamin) plays an essential role in DNA synthesis, erythropoiesis, and cellular metabolism. Its deficiency is common in the general population and may influence various biochemical markers. Tumor markers such as CA 125, CA 15-3, CA 19-9, CEA, and AFP are widely used in oncology, yet their levels can also be affected by non-malignant conditions. This study aimed to investigate the relationship between serum vitamin B12 levels and tumor marker concentrations in non-cancerous patients. Methods: This retrospective study included 250 adult participants (141 with low serum vitamin B12 < 200 ng/L and 109 with normal B12 ≥ 200 ng/L) who presented to Etlik City Hospital between June 2022 and June 2025. All participants were non-smokers and had no history of chronic kidney disease or malignancy. Hematologic, biochemical, thyroid, and iron parameters were analyzed alongside tumor markers (CA 15-3, CA 125, CA 19-9, CEA, and AFP) using standard immunoassay techniques. Statistical analyses included group comparisons, correlation testing, and logistic regression to identify factors independently associated with low vitamin B12 levels. Results: Patients with low vitamin B12 levels exhibited significantly lower hemoglobin, hematocrit, white blood cell, and neutrophil counts (p < 0.01 for all). Free T4 and serum iron were independently associated with low B12 levels in multivariable analysis (p < 0.001). Among tumor markers, CA 125 and CEA were significantly reduced in patients with vitamin B12 deficiency (p < 0.0001), while CA 15-3 showed a moderate inverse correlation with B12 concentration (r = –0.36, p < 0.0001). Conclusions: Vitamin B12 deficiency is associated with alterations in hematologic, thyroid, and iron parameters and may be associated with serum tumor marker levels even in non-cancerous individuals. Recognizing the potential association between B12 status and tumor marker variability may contribute to more cautious interpretation of tumor marker results in clinical practice.

Point-of-care ultrasound (POCUS) is increasingly recognized as an essential tool in internal medicine; however, its real-world adoption remains variable. This study aimed to evaluate the current landscape of POCUS use, training, and barriers among internal medicine physicians in Türkiye. We conducted a nationwide, cross-sectional survey between January and June 2025, including 1678 physicians across diverse healthcare settings. The questionnaire assessed POCUS availability, training, clinical use, perceptions, and barriers. POCUS was perceived as necessary by 98.7% of participants; however, 73.4% reported no personal implementation, and only 50.3% reported any lifetime POCUS experience. Prior training was reported by 12.8% and was associated with higher self-reported competency (median score 3 [2–5] vs. 1 [1, 2], p < 0.001). A substantial perception–practice gap was observed across all domains, particularly in abdominal (− 62.7%) and thyroid applications (− 58.0%). Clinical observation was limited, with 54.4% of participants never having witnessed POCUS being performed. When present, it was positively correlated with both personal implementation (r = 0.500, p < 0.001) and self-reported competency (r = 0.450, p < 0.001). The most frequently reported barriers were lack of qualified trainers (65.8%), limited access to equipment (56.4%), and time constraints (41.5%). Multivariable analysis linked prior training and career stage to integration readiness, whereas actual implementation was associated with clinical observation, prior training, equipment availability, and inversely with seniority. These findings suggest that similar gaps may exist in other healthcare systems undergoing POCUS integration, highlighting the importance of structured education and system-level support.

Background/Objectives: Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the leading cause of chronic liver disease globally, mirroring the increasing prevalence of obesity, insulin resistance, and type 2 diabetes. Early detection of hepatic steatosis is vital for cardiometabolic risk assessment; however, conventional imaging is costly and impractical for population screening. This study aimed to develop interpretable machine-learning models to predict ultrasound-detected hepatic steatosis within the MASLD spectrum using routinely available clinical and biochemical data. Methods: We analyzed data from 644 adults, 50% of whom had ultrasound-detected hepatic steatosis. Preprocessing, imputation, and feature selection were implemented within a single scikit-learn pipeline to avoid information leakage. An Elastic Net-regularized logistic regression identified the top 20 predictors, which were subsequently used across nine supervised machine learning (ML) classifiers. Model performance was evaluated via repeated stratified 5-fold cross-validation (25 resamples) using accuracy, F1 score, sensitivity, specificity, Youden’s J, balanced accuracy, and Area Under the Receiver Operating Characteristic Curve (AUROC). Interpretability was assessed using SHapley Additive exPlanations (SHAP). Results: Participants with ultrasound-detected hepatic steatosis exhibited greater adiposity, insulin resistance, and dyslipidemia compared with controls [p < 0.05 for body mass index (BMI), waist circumference, glucose, glycated hemoglobin (HbA1c), triglycerides]. Elastic Net selection highlighted Weight, Ponderal Index, Fibrosis-4 Index (FIB-4), blood urea nitrogen (BUN)/Creatinine ratio, Aspartate Aminotransferase to Platelet Ratio Index (APRI), and Visceral Adiposity Index as the strongest predictors. Logistic Regression and Gradient Boosting achieved the best performance (accuracy = 0.65 ± 0.03; AUROC = 0.71 ± 0.04; balanced accuracy = 0.66 ± 0.06), outperforming rule-based indices such as Fatty Liver Index (FLI) and Hepatic Steatosis Index (HSI) reported in the literature. SHAP analysis confirmed clinically coherent feature effects, with higher anthropometric and hepatic injury indices increasing the predicted probability of ultrasound-detected hepatic steatosis. Conclusions: Routinely available clinical and biochemical parameters can predict hepatic steatosis with moderate accuracy using transparent, interpretable ML models. Logistic Regression and Gradient Boosting provided best discrimination and robust internal performance, offering a pragmatic, low-cost approach for early identification of ultrasound-detected hepatic steatosis within the MASLD spectrum in primary and metabolic care settings.

Background/Objectives: Metabolic dysfunction-associated steatotic liver disease (MASLD) is now the leading cause of chronic liver disease globally, mirroring the in-creasing prevalence of obesity, insulin resistance, and type 2 diabetes. Early detection of hepatic steatosis is vital for cardiometabolic risk assessment; however, conventional imaging is costly and impractical for population screening. This study aimed to devel-op interpretable machine-learning models to predict ultrasound-detected hepatic ste-atosis within the MASLD spectrum using routinely available clinical and biochemical data. Methods: We analyzed data from 644 adults, 50% of whom had ultra-sound-detected hepatic steatosis. Preprocessing, imputation, and feature selection were implemented within a single scikit-learn pipeline to avoid information leakage. An Elastic Net–regularized logistic regression identified the top 20 predictors, which were subsequently used across nine supervised machine learning (ML) classifiers. Model performance was evaluated via repeated stratified 5-fold cross-validation (25 resamples) using accuracy, F1 score, sensitivity, specificity, Youden’s J, balanced accu-racy, and Area Under the Receiver Operating Characteristic Curve (AUROC). Inter-pretability was assessed using SHapley Additive exPlanations (SHAP). Results: Par-ticipants with ultrasound-detected hepatic steatosis exhibited greater adiposity, insu-lin resistance, and dyslipidemia compared with controls [p < 0.05 for body mass index (BMI), waist circumference, glucose, glycated hemoglobin (HbA1c), triglycerides]. Elastic Net selection highlighted Weight, Ponderal Index, Fibrosis-4 Index (FIB-4), blood urea nitrogen (BUN)/Creatinine ratio, Aspartate Aminotransferase to Platelet Ratio Index (APRI), and Visceral Adiposity Index as the strongest predictors. Logistic Regression and Gradient Boosting achieved the best performance (accuracy = 0.65 ± 0.03; AUROC = 0.71 ± 0.04; balanced accuracy = 0.66 ± 0.06), outperforming rule-based indices such as Fatty Liver Index (FLI) and Hepatic Steatosis Index (HSI) reported in the literature. SHAP analysis confirmed clinically coherent feature effects, with higher anthropometric and hepatic injury indices increasing the predicted probability of ul-trasound-detected hepatic steatosis. Conclusions: Routinely available clinical and bi-ochemical parameters can predict hepatic steatosis with moderate accuracy using transparent, interpretable ML models. Logistic Regression and Gradient Boosting pro-vided best discrimination and robust internal performance, offering a pragmatic, low-cost approach for early identification of ultrasound-detected hepatic steatosis within the MASLD spectrum in primary and metabolic care settings.

Aims: Diabetes mellitus is a well-recognized risk factor for rotator cuff pathology, yet the specific clinical and metabolic determinants of tendinopathy in this population remain poorly defined. This study aimed to identify the ultrasonographic, biochemical, and clinical factors associated with the development of supraspinatus tendinopathy among patients with diabetes. Methods: We performed a single-center, retrospective cross-sectional study of 63 patients with diabetes mellitus. All participants underwent shoulder ultrasonography to measure supraspinatus tendon thickness and assess tendinitis. Patients were grouped according to ultrasonographic evidence of tendinitis. Demographic data, glycemic parameters [hemoglobin A1c (HbA1c), fasting glucose], renal function indices [estimated glomerular filtration rate (eGFR), creatinine], lipid profiles, and diabetic complications (neuropathy, retinopathy, nephropathy) were compared between groups. Logistic regression and ROC curve analyses were used to identify and evaluate diagnostic predictors.Results: Supraspinatus tendinopathy was present in 29 patients (46%). Compared with patients without tendinitis, affected individuals had significantly greater supraspinatus tendon thickness (7.66±1.29 vs. 6.95±1.22 mm; p=0.03), lower eGFR (p=0.005), higher serum creatinine (p=0.03), and a higher prevalence of diabetic neuropathy (65% vs. 35%; p=0.02). Although tendon thickness correlated positively with body mass index, HbA1c, and C-reactive protein, these metabolic and inflammatory markers did not differ significantly between groups. ROC analysis showed supraspinatus tendon thickness had the highest diagnostic accuracy (AUC=0.70).Conclusion: In our study, supraspinatus tendinopathy in diabetes mellitus was associated with tendon thickening, renal dysfunction, and neuropathy rather than glycemic or lipid measures, highlighting microvascular complications as key drivers of diabetic rotator cuff tendinopathy.

Background/Objectives: Metabolic dysfunction–associated steatotic liver disease (MASLD) is now the leading cause of chronic liver disease globally, mirroring the increasing prevalence of obesity, insulin resistance, and type 2 diabetes. Early detection of hepatic steatosis is vital for cardiometabolic risk assessment; however, conventional imaging is costly and impractical for population screening. This study aimed to develop interpretable machine-learning models to predict ultrasound-detected MASLD using routinely available clinical and biochemical data. Methods: We analyzed data from 644 adults (50% with MASLD on ultrasonography). Preprocessing, imputation, and feature selection were implemented within a single scikit-learn pipeline to avoid information leakage. An Elastic Net–regularized logistic regression identified the top 20 predictors, which were subsequently used across nine supervised machine learning (ML) classifiers. Model performance was evaluated via repeated stratified 5-fold cross-validation (25 resamples) using accuracy, F1 score, sensitivity, specificity, Youden’s J, balanced accuracy, and Area Under the Receiver Operating Characteristic Curve (AUROC). Interpretability was assessed using SHapley Additive exPlanations (SHAP). Results: Participants with MASLD exhibited greater adiposity, insulin resistance, and dyslipidemia compared with controls [p < 0.05 for body mass index (BMI), waist circumference, glucose, HbA1c, triglycerides). Elastic Net selection highlighted Weight, Ponderal Index, Fibrosis-4 Index (FIB-4), blood urea nitrogen (BUN)/Creatinine ratio, Aspartate Aminotransferase to Platelet Ratio Index (APRI), and Visceral Adiposity Index as the strongest predictors. Logistic Regression and Gradient Boosting achieved the best performance (accuracy = 0.65 ± 0.03; AUROC = 0.71 ± 0.04; balanced accuracy = 0.66 ± 0.06), outperforming rule-based indices such as Fatty Liver Index (FLI) and Hepatic Steatosis Index (HSI) reported in the literature. SHAP analysis confirmed clinically coherent feature effects, with higher anthropometric and hepatic injury indices increasing predicted MASLD probability. Conclusions: Routinely available clinical and biochemical parameters can predict hepatic steatosis with moderate accuracy using transparent, interpretable ML models. Logistic Regression and Gradient Boosting provided the best discrimination and generalizability, offering a pragmatic, low-cost approach for early MASLD screening in primary and metabolic care settings.

Introduction: Metabolic dysfunction-associated steatotic liver disease (MASLD) prevalence data in Türkiye is limited. We aimed to determine the nationwide prevalence of MASLD and advanced hepatic fibrosis risk in subjects with cardiometabolic risk factors (CMRF). Despite recent international consensus redefining fatty liver disease terminology, no nationwide MASLD study has been reported in Türkiye. Methods: This cross-sectional study included 14,371 adults with ≥1 CMRF from 44 centers across 31 cities. MASLD was diagnosed using liver ultrasonography plus cardiometabolic criteria. Advanced fibrosis risk was assessed by fibrosis-4 (FIB-4) score (≥1.3 for ≤65 years; ≥2.0 for >65 years). Logistic regression was used to identify independent predictors of high FIB-4. Results: A total of 61.4% of participants were women, the mean age was 51.3 ± 14.4 years, and the mean BMI was 31.4 ± 6.0 kg/m². MASLD prevalence was 75.7% (n = 10,873), rising with the number of CMRFs (56.5% with one factor vs. 83.4% with all). The prevalence of high FIB-4 scores was 12.0% overall, being lower in MASLD patients than non-MASLD patients (11.2% vs. 14.4%, p < 0.001). FIB-4 scores decreased with increasing BMI (28.1% underweight vs. 8.7% class III obesity). Male sex, T2DM, and hypertension independently predicted high FIB-4 scores, while smoking, higher BMI, and MASLD were inversely associated. Conclusions: Three-quarters of Turkish adults with CMRF have MASLD. Standard FIB-4 thresholds may underestimate fibrosis risk in obese and smoking populations, underscoring the need for adjusted cut-offs or alternative tools. This study is the first to provide nationwide MASLD prevalence data in Türkiye.

Amaç: Kolorektal kanserler, kanserle ilgili mortalite ve morbiditenin en önemli sebeplerindendir. Tüm dünyada erkeklerde tanı alan en sık 3., kadınlarda ise en sık 2. kanserdir. Kolon kanseri erkek ve kadında hemen hemen eşit görülürken, rektal kanser erkeklerde K:E = 1:1.3 oranında daha sık görülür. Medyan tanı yaşı 70 olan kolorektal kanserler 50 yaş altında nadir görülür. Kolorektal tanısı ile takipli hastaların genel nitelikleri ve bu niteliklerin literatür ile uyumlu olup olmadığı araştırıldı. Gereç ve Yöntemler: Çalışma, kesitsel, retrospektif kohort çalışması olarak tasarlandı. Temmuz 2011- Eylül 2015 tarihleri arasında Ankara Dışkapı Eğitim ve Araştırma Hastanesi Onkoloji Kliniği’nde takip edilen, kolonoskopi yapılarak tanısı konulan ve histopatolojik olarak tanısı doğrulanan kolorektal kanserli vakalar değerlendirildi. Sonuç: Çalışmaya evre I-IV kolorektal kanserli 172 hasta alındı ve genel nitelikleri değerlendirildi. Hastalık literatür ile uyumlu şekilde erkek cinsiyette kadınlara göre daha sık tespit edildi Hastaların medyan yaşı 62 olarak saptandı. En genç hasta 27, en yaşlı hasta ise 89 yaşındaydı. KRK tanısıyla takip edilen erkek hastaların %37’sinde daha çok rektum kanserleri, %61’inde daha proksimal yerleşimli kolon kanserleri görülürken, kadınlarda %36 oranında rektum kanserleri ve buna yakın olarak %38 oranında daha proksimal yerleşimli kolon kanserleri görüldüğü saptanmıştır. Bu çalışmamız ile dünyada olduğu gibi ülkemizde de önemli morbidite ve mortalite nedenlerinden biri olan kolorektal kanserlerin demografik yapısını ve risk faktörlerini belirleyerek literatüre katkı sağlamayı amaçladık.

Journal of Health Sciences and Medicine | Volume: 7 Issue: 2

Aims: The primary purpose of this study is to make a retrospective evaluation of adrenal incidentaloma (AI) prevalence in patients performed thorax computed tomography (TCT), along with determining whether the diagnosed AIs were assessed functionally for adrenal gland hyperfunction and to detect the rate of hyperfunctional AI. Methods: The patients who applied to with the suspected COVID-19 and performed TCT between January 2020 and December 2021, were included in the present study. However, the patients who were followed-up due to a known adrenal mass and a malignant tumor were excluded. Results: TCT imaging, including adrenal glands, was performed on 2580 patients. The mean age for these patients was 54±16.8. However, when the patients were separated into groups in terms of adrenal pathology (AP), the mean age for the patients with and without AP was 61.4±11 and 53.8±16.9, respectively (p<0.001). The number of patients detected with AP, AI, and adrenal hyperplasia was 68 (2.6%), 60 (88.2%), and 7 (10.3%), respectively. The mean mass diameter was 17 (11-41). Bilateral adrenal hyperplasia was not observed in any patients; however, 13 of the patients (18.1%) detected with AI and adrenal hyperplasia were evaluated functionally. Hyperfunction was not observed in any of the assessed patients. Conclusion: The prevalence of AI found in our study was similar to other studies in the literature; however, functional evaluations of AIs detected via TCT performed due to suspected COVID-19 remained low. Although they are rare, it is important to define whether AIs are functional or malignant due to the comorbid conditions they create. For this reason, we believe patients with AI should be directed to an endocrinology clinic for a practical examination and follow-up plan.