Introduction: This study aims to predict mortality using chest radiographs obtained on the first day of intensive care admission, thereby contributing to better planning of doctors’ treatment strategies and more efficient use of limited resources through early and accurate predictions. Methods: We retrospectively analyzed 510 ICU patients. After data augmentation, a total of 3019 chest radiographs were used for model training and validation, while an independent, non-augmented test set of 100 patients (100 images) was reserved for final evaluation. Seventy-four (74) radiomic features were extracted from the images and analyzed using machine learning algorithms. Model performances were evaluated using the area under the ROC curve (AUC), sensitivity, and specificity metrics. Results: A total of 3019 data samples were included in the study. Through feature selection methods, the initial 74 features were gradually reduced to 10. The Subspace KNN algorithm demonstrated the highest prediction accuracy, achieving AUC 0.88, sensitivity 0.80, and specificity 0.87. Conclusions: Machine learning algorithms such as Subspace KNN and features obtained from PAAC radiographs, such as GLCM Contrast, Kurtosis, Cobb angle, Haralick, Bilateral Infiltrates, Cardiomegaly, Skewness, Unilateral Effusion, Median Intensity, and Intensity Range, are promising tools for mortality prediction in patients hospitalized in the internal medicine intensive care unit. These tools can be integrated into clinical decision support systems to provide benefits in patient management.

Radyoloji başucu serisi tıp fakültesi öğrencileri, radyoloji asistanları, radyoloji uzmanları, radyolojinin geniş ilgi alanında anlatılan konuyla ilgili diğer hekim ve sağlık personelleri için, hap formatında, her konunun en sık görülen tanılarının bulunduğu başucu veya başvuru kitabı olarak hazırlanmıştır.

Breast cancer (BC) is a heterogeneous disease with distinct molecular subtypes that exhibit variable immune responses and metabolic profiles. Recent studies have suggested that immunometabolic pathways play a role in tumor progression and treatment resistance. This study investigates the expression patterns of ALKBH7 and NLRP3 across BC molecular subtypes and explores their relationships with clinicopathological parameters and potential immunometabolic profiles. A total of 118 BC patients were classified into HER2+, TNBC, Luminal A, and Luminal B subtypes. Gene expression levels of ALKBH7 and NLRP3 were analyzed using quantitative real-time PCR (qRT-PCR), and correlations with clinical markers were assessed. ALKBH7 and NLRP3 expression levels varied significantly between subtypes, with the highest expression observed in HER2+ tumors. Strong positive correlations were found between ALKBH7 and NLRP3 in all subtypes, particularly in HER2+ (r = 0.812, p < 0.001). Additionally, NLRP3 correlated with Ki-67 in Luminal B tumors, indicating a link between inflammation and proliferative capacity. These findings suggest that ALKBH7 may function as a dual-role biomarker involved in metabolic adaptation and immune signaling in BC. The strong co-expression of ALKBH7 and NLRP3 suggests a functional association between these molecules that may be critical in shaping the tumor microenvironment. This co-expression set, particularly in aggressive subtypes (HER2+ and TNBC), warrants further mechanistic validation as a potential prognostic marker and a novel therapeutic vulnerability.

Background/Objectives: Brain death determination in children is clinically challenging. When standard clinical examination cannot be completed or reliably interpreted, ancillary testing is required—yet many established methods depend on infrastructure or patient transport that may not be feasible in critically ill pediatric patients. Orbital ultrasonography is bedside-applicable and non-invasive, but remains poorly characterized in children. Methods: We conducted a single-center retrospective study of 28 pediatric patients evaluated for suspected brain death between January 2021 and February 2025. Patients were classified as brain death-positive [BD(+), n = 20] or brain death-negative [BD(−), n = 8] based on clinical criteria independent of imaging findings. Orbital color Doppler parameters (ophthalmic artery, central retinal artery, posterior ciliary artery) and optic nerve sheath diameter (ONSD) were measured under a standardized protocol by a single experienced operator. Ophthalmic artery resistive index (OA-RI) was defined a priori as the primary outcome; ONSD was the secondary outcome. Group comparisons used the Mann–Whitney U test with Cliff’s delta effect sizes; false discovery rate correction was applied to secondary and exploratory comparisons. ROC analyses were performed to assess discriminative performance. The study was reported in accordance with the STARD 2015 guidelines for diagnostic accuracy research. Results: OA-RI was markedly higher in BD(+) patients (0.84 [IQR 0.80–0.90] vs. 0.65 [0.58–0.69]; p < 0.001; δ = 0.975). ROC analysis yielded an AUC of 0.99 (95% CI: 0.96–1.00); at a cut-off of ≥0.77, sensitivity was 95.0% and specificity 100.0%. ONSD also differed significantly between groups (4.75 [4.15–5.08] mm vs. 3.90 [3.40–4.15] mm; p = 0.012; δ = 0.619; AUC = 0.81, 95% CI: 0.62–1.00; cut-off ≥ 4.2 mm; sensitivity and specificity both 75.0%). Across all three orbital vessels, end-diastolic velocity was consistently reduced and resistive indices elevated in BD(+) patients. Systolic velocities did not differ meaningfully between groups. Cut-off values represent cohort-specific statistical optima and should be interpreted as exploratory. Conclusions: Orbital Doppler ultrasonography demonstrates a coherent high-resistance hemodynamic pattern in pediatric brain death. OA-RI showed strong discriminative performance and may serve as a useful bedside adjunct in selected cases where ancillary testing is indicated. ONSD provides complementary anatomical evidence. These findings are exploratory and require prospective validation in larger, multicenter pediatric cohorts.

Background/Objectives: This study investigated the impact of feature type, feature dimensionality, and feature selection strategies on intensive care unit (ICU) mortality prediction using first-day portable frontal chest radiographs and demographic data. In addition, the study evaluated whether comparable predictive performance could be achieved using reduced, clinically interpretable feature sets. Methods: A total of 500 patients were included, comprising 400 cases for model development and 100 independent cases for testing. Two complementary experimental frameworks were evaluated. In Experiment Set-1, a clinically interpretable feature set consisting of 12 radiographic and demographic variables was analyzed. In Experiment Set-2, a high-dimensional feature space consisting of 74 radiographic, radiological, radiomic, and image-derived features was investigated using six feature selection methods: ANOVA, Chi-Square, Kruskal–Wallis, MRMR, ReliefF, and Shapley-based importance analysis. Machine learning models were evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, and F1-score. Results: In Experiment Set-1, feature selection reduced the number of predictors from 12 to 4. Using the complete 12-feature set, the Bagged Trees classifier achieved the highest performance (AUC = 0.95). Following feature reduction, the Subspace KNN classifier achieved an AUC of 0.96 with an accuracy of 0.82, indicating that satisfactory predictive performance could be achieved using a substantially smaller feature set. In Experiment Set-2, comparison of six feature selection strategies demonstrated that MRMR and Kruskal–Wallis produced the highest-performing feature subsets. Cobb angle, bilateral infiltrates, bilateral pleural effusion, and unilateral pleural effusion were consistently identified across all feature selection methods as stable predictors associated with ICU mortality. These predictors were considered stable because they were consistently identified across multiple independent feature selection methods despite differences in the underlying selection algorithms. Conclusions: The findings suggest that appropriate feature selection strategies may reduce model complexity while retaining clinically relevant predictive information. The consistent identification of a small group of radiographic features across multiple feature selection methods in the 74-feature experimental framework suggests that these variables represent robust and clinically meaningful predictors of ICU mortality. Further external validation using larger multicenter datasets is required before clinical implementation.