Abstract Identity dissociation is challenging to detect and treat, and its etiology remains incompletely understood. Childhood maltreatment and FKBP5 polymorphisms, which modulate the stress response, may contribute by disrupting the integration of autobiographical experiences essential for identity development. We examined whether gene-environment interactions involving childhood maltreatment and FKBP5 polymorphisms predict clinically significant identity dissociation. In a cohort of N = 377 participants, we assessed childhood maltreatment and identity dissociation using validated questionnaires and genotyped CATT haplotypes within FKBP5 linked to stress reactivity. Identity dissociation was dichotomized using an established clinical threshold. An elastic net regularized logistic regression model incorporating maltreatment subtypes, CATT haplotype count, and their interactions was trained ( N = 194) and validated ( N = 183). Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and Matthews correlation coefficient. Decision curve analysis assessed clinical utility across varying risk thresholds. The model demonstrated fair discrimination (AUC = 0.709) with 58.6% sensitivity and 79.9% specificity. While the positive predictive value was modest (35.0%) due to the low prevalence of identity dissociation (15.9%), decision curve analysis revealed a net clinical benefit across a broad range of threshold probabilities (6–76%), indicating practical utility for risk stratification in clinical settings. The negative predictive value was 0.91. These findings provide initial evidence that gene-environment interactions between childhood maltreatment and FKBP5 variation may contribute to the risk of identity dissociation. While predictive precision remains limited, this study demonstrates the feasibility of applying machine learning approaches to dissociation and highlights the need for further research into their traumatic and biological underpinnings to improve detection, prevention, and treatment.

This study aimed to evaluate whether β-thalassemia trait can be diagnosed without hemoglobin electrophoresis by assessing the diagnostic performance of complete blood count parameters (mean corpuscular volume [MCV], red cell distribution width [RDW], and Mentzer index), transferrin saturation, and derived indices in anemic pediatric patients, using hemoglobin electrophoresis as the reference standard. A retrospective observational cross-sectional study was conducted on 558 children (1 month-18 years) who underwent hemoglobin electrophoresis between 2019 and 2025. Hematological parameters, iron status, and family history were analyzed. Patients were classified as normal electrophoresis, β-thalassemia carrier, major, or intermedia according to guidelines. Receiver operating characteristic analysis determined cutoffs, and a logistic regression model was developed and validated (80% training, 20% testing, Hosmer-Lemeshow P = .62, bootstrap n = 1000). Among 550 analyzed patients, β-thalassemia carrier prevalence was 25.1%. Carriers showed significantly lower MCV (57.07 ± 4.93 vs 70.08 ± 7.08 fL) and higher RDW (18.57 ± 2.62 vs 15.00 ± 2.39%) compared to controls (P < .001). The Mentzer index (≤11.97) achieved optimal performance with 93.5% sensitivity and 88.1% specificity. A combined diagnostic rule using 3 parameters (MCV ≤ 62.45 fL, Mentzer ≤ 11.97, and RDW > 16.45%) markedly improved specificity to 98.3% and positive predictive value to 93.5%. The logistic regression model achieved 95.5% overall accuracy. Positive family history was significantly associated with carrier status (23.1% vs 7.5%, P < .001). Simple complete blood count-derived indices offer reliable, cost-effective screening for β-thalassemia carriers, potentially reducing unnecessary electrophoresis testing by 60% to 70% in resource-limited settings.

BACKGROUND This study evaluated the real-world diagnostic performance and limitations of noninvasive prenatal testing (NIPT) in detecting fetal copy number variations (CNVs) within a large multicenter cohort in Shaanxi Province. MATERIAL AND METHODS This retrospective observational study analyzed 18 525 cases of NIPT at the First Affiliated Hospital of Xi'an Jiaotong University, a referral center for NIPT, from June 2023 to November 2024. Karyotype analysis and CNV sequencing were conducted on the fetuses and/or parents, with follow-up on pregnancy outcomes. RESULTS Abnormal CNVs were detected in 218 cases (1.18%; 218/18525), of which 129 women (59.17%; 129/218) opted for invasive diagnostic confirmation from 38 hospitals in 7 prefectural-level cities. The positive predictive value (PPV) for aberrant CNVs following NIPT was only 48.06% (62/129; 95% CI, 39.4-56.7%), with 28.57% (18/62) possessing pathogenic CNVs. We noted that PPV estimates were based on self-selected confirmatory testing, which might inflate or deflate performance estimates. The detection efficiency varied significantly by chromosomal location; chromosome 18 showed the highest PPV at 83.33% (15/18; P<0.05), notably within the 18p11.23-p11.31 segment. Furthermore, smaller CNVs (<5 Mb) demonstrated a higher concordance rate (PPV 54.74%; 52/95) than larger fragments (>10 Mb). Regional analysis indicated Hanzhong and Xi'an demonstrated elevated PPVs, while Yulin showed the highest incidence of pathogenic CNVs. CONCLUSIONS NIPT demonstrates moderate performance for fetal CNV detection, with a PPV of approximately 48%. Its clinical utility is maximized when combined with ultrasound findings, which significantly increase the predictive value. The stakeholders should be aware of this limitation when interpreting results.

Digital breast tomosynthesis (DBT) increases sensitivity and specificity compared to digital mammography (DM) in the early detection of breast cancer. The potential of artificial intelligence (AI) for mammographic interpretation resulting in workload reduction is increasingly being reported.Presentation of the current evidence on the efficacy of DBT versus DM in population-based breast cancer screening and its support by AI. Narrative review with topic-led literature search of comparative studies of DBT and DM and review of the use of AI in PubMed from 01/2016 to 09/2025.In 42 international studies, the breast cancer detection rate was significantly higher with DBT + DM (6.4‰) and DBT + synthetic mammography (SM) (7.4‰) than with DM (4.7‰). Concordantly, the randomized TOSYMA study reported a higher invasive breast cancer detection rate with DBT + SM (7.1‰) versus DM (4.8‰) with a lower false-positive recall rate (first round -15.6‰). The positive predictive value (PPV) of recall was - consistent with meta-analyses - higher (+4.9%), as was the reading time. With a smaller number of AI studies on DBT than DM, a DBT meta-analysis reported a higher sensitivity of AI alone (89%) than by readers (78%), with a lower specificity of AI. DBT with AI-assisted reading compared to human reading alone increased the detection rate in a prospective study by +3.8‰ without a marked change in the recall rate (+0.8%).DBT increases breast cancer detection compared to DM with more favorable process parameters. AI reporting strategies can further increase sensitivity and reduce human workload. The influence of DBT and AI on reducing the interval cancer rate as a measure of efficiency has not yet been proven. · DBT implementation in population-based breast cancer screening is being reviewed internationally.. · In case of DM replacement, AI reading concepts will become more important.. · Evidence of the long-term effectiveness of DBT and AI is limited.. · Weigel S, Wunderlich P, Sommer A et al. Digital breast tomosynthesis and reading with artificial intelligence. Rofo 2026; DOI 10.1055/a-2796-5225.

Herein, RATSGO (Real-time Automated Training and Sensing for Gas Odor), a fully automated live-animal olfactory training platform, for the detection of GBL as a sexual assault-facilitating drug is reported. The system integrates four distinct operant conditioning-based training paradigms, all executed without human intervention, to enhance learning speed, consistency, and scalability. Using this fully automated framework, four rats were trained to identify γ-butyrolactone (GBL). Three of the four animals successfully reached the predefined learning completion criterion, whereas one failed to meet the criterion. Across 320 automated trials, the GBL rats achieved a mean detection accuracy of 90%, with sensitivity and specificity values of 97% and 82%, respectively. The corresponding positive and negative predictive values (PPV and NPV) were 85% and 96%. When challenged with GBL diluted in drinking water (180 trials), performance remained high, yielding 88% accuracy, 89% sensitivity, 87% specificity, 85% PPV, and 90% NPV. Similarly, in experiments involving GBL mixed with whisky (200 trials), the rats demonstrated robust recognition capability, achieving 90% overall accuracy, perfect sensitivity (100%), 84% specificity, 79% PPV, and 100% NPV. Importantly, odor discrimination performance was preserved when reassessed four months after the completion of training, indicating strong long-term retention of the learned odor representations. Collectively, these findings confirm that the RATSGO system supports rapid, stable, and precise odor learning, underscoring its promise as a practical and extensible biological sensing platform for chemical detection applications.

Circulating tumor cells (CTCs) are detectable in early-stage cancer and may enable early cancer detection. We evaluated a CTC-based assay as a complementary biomarker for breast cancer detection in an Asian population with a high prevalence of dense breast tissue. In this single-center, prospective, blinded study, peripheral blood from Taiwanese women with breast cancer and healthy controls was analyzed using a CTC-enumeration platform (CMx) based on biomarker expression (cytokeratin 18 [CK18], mammaglobin [MGB], CD45), cell morphometry, and nuclear features. A machine-learning model integrating CTC biomarkers with age, white blood cell (WBC) count, and platelet count was developed to assess classification performance, providing proof-of-concept for combining CTC-derived and routine blood parameters in breast cancer risk assessment. A total of 228 breast cancer patients and 170 healthy controls were included. Age and CK18- and MGB-positive CTC counts differed significantly between groups, whereas WBC and platelet counts did not. An ensemble linear support vector machines model incorporating age and CTC features achieved an area under the curve of 0.85 (95% CI, 0.73-0.96) in the independent test cohort, with high sensitivity (0.93), positive predictive value (0.74), and negative predictive value (0.86), but modest specificity (0.57). In the exploratory BI-RADS 3/4 subgroup, the model identified all cancer cases (sensitivity 1.00), with a specificity of 0.44 and overall accuracy of 0.79. This study demonstrates the feasibility of combining CTC enumeration with machine learning for breast cancer detection and supports the need for future large-scale, multicenter, multiethnic prospective external validation.

Introduction: Community-acquired urinary tract infection (CA-UTI) is the second most common infection in the community. The gold standard for diagnosing urinary tract infections (UTIs) is urine culture. A urine dipstick can also be used to diagnose UTIs. This study was designed to compare the urine dipstick test with urine culture in the diagnosis of community-acquired urinary tract infections in a tertiary care hospital in the Colombo district, Sri Lanka.Methods: Four hundred and eighty-nine urine samples were inoculated into cysteine-lactose-electrolyte-deficient agar. They were tested for the presence of nitrite and leukocytes using urine dipsticks, following the manufacturer's instructions.Results: In the entire group, the sensitivity, specificity, positive predictive value, and negative predictive value of the leukocyte esterase test were 91.92%, 76.15%, 91.41%, and 77.34%, respectively, and those of the nitrite test were 91.36%, 75.38%, 91.11%, and 75.97%, respectively.Conclusion: The dipstick is a low-cost, user-friendly test that can be used in an outpatient setting to identify patients who require antibiotics. This will help in establishing antibiotic stewardship by limiting unnecessary prescriptions.

Background Patients admitted with viral respiratory tract infections are at risk of bacterial co-infections, often requiring antibiotics. Standard bacterial cultures may take days to yield results, making early indicators of bacterial co-infection a potential asset. Methods In a retrospective regionwide cohort study, we included all patients admitted to a hospital in the Central Denmark Region with COVID-19, influenza, or respiratory syncytial virus (RSV) in the period February 2019 to February 2024. Further inclusion criteria were having a blood sample taken for procalcitonin measurement and blood cultures within 72 h of admission. We evaluated the diagnostic value of procalcitonin compared to other inflammatory markers. Results There were 1,670 patients fulfilling the inclusion criteria. The majority 1,556 (92.2%) were infected with SARS-CoV-2, 51 (3.7%) with influenza A, 1 (0.1%) with influenza B, 45 (2.7%) with RSV, and 17 (1.0%) with multiple viruses. Blood cultures were positive in 43 (2.6%) of the patients after a median of 67.4 h from time of admission. Median procalcitonin levels were higher in patients with bacteraemia (1.7 µg/L) than in patients without bacteraemia (0.2 µg/L. Overall, procalcitonin cut-off at >0.25 µg/L and >0.50 µg/L tended to have higher sensitivity, specificity, positive and negative predictive values than C-reactive protein (CRP) > 100 mg/L and total leukocyte count >10 × 109 cells/L, although the numerical differences were small. Overall diagnostic performance for bacterial pneumonia was lower. Conclusions All biomarkers had relatively low sensitivity for bacteraemia and bacterial pneumonia in patients with viral respiratory tract infection.

Introduction: Neoadjuvant systemic and immunotherapy strategies in non-metastatic colon cancer have demonstrated high pathological response rates, raising interest in surgery-sparing approaches. Circulating tumour DNA (ctDNA) is an emerging biomarker for treatment response and minimal residual disease, but its role in guiding surgical omission in colon cancer remains unclear. This systematic review evaluates the diagnostic and prognostic accuracy of ctDNA in predicting pathological response following neoadjuvant therapy in non-metastatic colon cancer. Methods: A systematic review was conducted in accordance with PRISMA guidelines. PubMed, Embase/MEDLINE, Scopus, and the Cochrane Register were searched from inception to 21 October 2025. Eligible studies included adults with non-metastatic colon cancer treated with neoadjuvant therapy who had serial ctDNA assessment prior to surgery. Results: Three cohort studies comprising 100 patients met inclusion criteria. Baseline ctDNA detection ranged from 42% to 84%. Across studies, ctDNA clearance following neoadjuvant therapy was consistently associated with major pathological response or pathological complete response, whereas persistent ctDNA strongly predicted residual viable tumour at resection. In the largest prospective cohort, 5 of 26 patients (19%) achieved ctDNA clearance prior to surgery; all were pathological responders, while 19 of 26 patients (73%) with persistent ctDNA demonstrated no pathological response. No study reported pathological complete response in the presence of persistently positive ctDNA. No prospective trial formally evaluated ctDNA-guided surgical omission. Conclusions: Current evidence does not support the use of ctDNA alone to guide omission of surgery after neoadjuvant therapy in non-metastatic colon cancer—even in patients who show complete pathological response. While persistent ctDNA reliably identifies patients with residual disease, ctDNA clearance lacks sufficient positive predictive value to safely forego surgery. Prospective trials with standardised ctDNA platforms and predefined non-operative management protocols are required before ctDNA-guided organ preservation can be recommended.

INTRODUCTION. Evaluation of endodontic treatment outcomes through radiographic assessment is subject to interobserver variability and depends heavily on clinician experience. Artificial intelligence (AI) platforms offer potential for standardized, objective assessment of periapical healing. MATERIALS AND METHODS. This retrospective study analyzed 400 panoramic radiographs from patients who underwent root canal treatment between January 2023 and December 2024. An AI platform developed using TensorFlow and Keras, with model training in PyTorch and validation in MATLAB Deep Learning Toolbox, was employed. Three blinded expert endodontists independently assessed all radiographs, with consensus serving as the gold standard. Outcomes were classified as healed, healing, or diseased based on periapical index criteria. Diagnostic performance metrics including sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) were calculated. RESULTS. The AI platform demonstrated overall accuracy of 89.8% in classifying treatment outcomes. For detecting healed cases, sensitivity was 92.3%, specificity 87.6%, PPV 88.9%, and NPV 91.5%. For diseased / persistent pathology detection, sensitivity was 88.7%, specificity 93.2%, PPV 84.3%, and NPV 95.1%. Agreement between AI and expert consensus was substantial (Cohen’s κ = 0.834, p &lt; 0.001). AI performance was superior in anterior teeth (93.2% accuracy) compared to molars (86.4% accuracy, p = 0.008). Processing time per radiograph averaged 2.3 ± 0.4 seconds. CONCLUSIONS. The AI-driven platform demonstrated high diagnostic accuracy comparable to expert assessment, with potential for standardized, rapid evaluation of endodontic treatment outcomes. Further prospective validation and clinical integration studies are warranted.

The Joint Commission recommends universal suicide screening in emergency departments (EDs), which emphasizes the need to identify at-risk individuals. Existing suicide risk prediction models rely primarily on clinical data and demonstrate limited performance. The potential of incorporating psychosocial information to enhance predictive performance remains understudied. To evaluate whether augmenting clinical data-based risk scores with psychosocial factors improves the prediction of suicide attempt (SA). This retrospective prognostic study based on electronic health record data included 4661 ED patients discharged after presentation for suicidal ideation (SI) from middle Tennessee hospitals between June 1, 2018, and February 27, 2024. The primary outcome was SA within 90 days of ED admission and time-to-event in days. Clinical data-based Vanderbilt Suicide Attempt and Ideation Likelihood (VSAIL) score and 6 psychosocial factors (homelessness, financial insecurity, chronic stress, social isolation, loneliness, and adverse childhood experiences) derived from clinical notes were integrated using a Cox proportional hazards regression model. Performance metrics included area under the receiver operating curve (AUROC), area under the precision-recall curve (AUPRC), positive predictive value (PPV), negative predictive value, sensitivity, and specificity. Performance was evaluated for models trained on (1) VSAIL, (2) psychosocial factors, and (3) VSAIL plus psychosocial factors. This study included 3382 Vanderbilt University Hospital (VUH) (mean [SD] age, 26.1 [15.6] years; 1751 males [51.8%]) and 1279 Regional Health Systems (RHS) (mean [SD] age, 34.5 [18.0] years; 715 males [55.9%]) ED visits for SI. Within 90 days, SAs were reported in 160 (4.7%) VUH and 34 (2.7%) RHS ED visits for SI. Compared with VSAIL alone, VSAIL plus psychosocial factors was associated with significantly increased median AUROC (VUH: 0.645 [IQR, 0.645-0.645] vs 0.734 [IQR, 0.719-0.747]; P < .001; RHS: 0.547 [IQR, 0.547-0.547] vs 0.680 [IQR, 0.672-0.687]; P < .001), AUPRC (VUH: 0.083 [IQR, 0.083-0.083] vs 0.122 [IQR, 0.111-0.137]; P < .001; RHS: 0.029 [IQR, 0.029-0.029] vs 0.054 [IQR, 0.052-0.058]; P < .001), and PPV (VUH: 0.093 [IQR, 0.082-0.094] vs 0.143 [IQR, 0.123-0.161]; P < .001; RHS: 0.042 [IQR, 0.040-0.043] vs 0.112 [IQR, 0.096-0.129]; P < .001) while maintaining specificities above 0.90. Chronic stress emerged as the strongest predictor of SA (β = 0.643 [95% CI, 0.427-0.859]; P < .001). In this prognostic study of patients discharged from the ED after presentation for SI, augmenting a clinical data-based suicide risk prediction model with clinical note-extracted psychosocial factors was associated with significantly higher predictive performance. These findings suggest that psychosocial factors can enhance risk stratification and support targeted interventions, such as therapies addressing chronic stress.

Pancreatic neuroendocrine tumor (PNET) behavior depends on tumor grade and genomics. Radiomics may identify these factors noninvasively. This study developed an automated pipeline from segmentation to radiomics modeling to preoperatively predict tumor grade. Patients resected from 2003 to 2021 with adequate preoperative arterial phase computed tomography (CT) scans were divided into training and test cohorts. The training cohort underwent manual pancreas and tumor region segmentation to train an auto-segmentation model; radiomic features extracted from tumor regions were used to develop a radiomics model for grade prediction (I versus II/III), which was evaluated in the automatically segmented test cohort. Associations between radiomic and genomic features were assessed. In total, 182 patients were divided into training (n = 140) and test (n = 42) cohorts. Grade I and II/III lesions were in 113 (62%) and 69 (38%) patients, respectively. Median tumor size was 24 mm (6, 200). The auto-segmentation model segmented tumor regions in 90% (38) of the test cohort. In this group (n = 38), the radiomics model produced receiver operating characteristic curve (AUC) of 0.85 (0.73, 0.96). Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 0.88 (0.78,0.98), 0.62 (0.46, 0.77), 0.65 (0.50, 0.80), and 0.87 (0.76,0.97), respectively. DAXX loss was associated with three radiomic features, and ATRX loss with one. In < 2cm lesions, auto-segmentation was successful in 75% (9/12) of the test cohort, with accurate grade prediction in 67% (6/9) of cases. The auto-segmentation model correctly identified tumor regions, and the radiomics model accurately predicted grade; new associations between genomic and radiomic features were identified. This automated pipeline can incorporate a radiomics model into preoperative PNET decision-making.

To assess the performanceofthree-dimensional T2-weighted fast field echo imaging (3D-T2-FFE) in the visualization of the intraparotid facial nerve (IFN) and localization of tumors. Magnetic resonance imaging data from sixty-four patients who underwent 3D-T2-FFE (Time of repetition 8.30ms, Time of echo 4.10ms, Voxel 0.65 × 0.65 × 1.00mm, Field of view 220 × 220 × 65mm, Matrix 340 × 339 × 130, Number of signal average 2, Flip angle 30°) were retrospectively enrolled. Finally, 64 cases of tumors were included (including 55 benign tumors and 9 malignant tumors). The identification certainty of IFN on 3D-T2-FFE was scored with an arbitrary scale of 0-3. The parotid gland was divided into superior and inferior parts, with the level of the earlobe (approximately at the level of the external auditory canal). The tumor location was categorized as deep or superficial directly on 3D-T2-FFE images and indirectly by the facial nerve line (FNL) and the retromandibular vein line (RMVL). Surgical localization was considered the reference standard. The accuracy, sensitivity, and specificity of each method for localizing parotid lesions were compared using the McNemar test. The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value for deep lobe lesions in the superior part of the parotid gland using the direct method were 97.8%, 92.3%, 100.0%, 100.0%, and 96.9%, respectively. The 3D-T2-FFE method showed significantly higher sensitivity and specificity than those of FNL (p < 0.05) in the superior part of the parotid gland, and there was no significant difference in sensitivity, specificity, and accuracy between 3D-T2-FFE and RMVL (p > 0.05). The relationship between the tumor and the main trunk of the IFN was correctly predicted in 93.3% and 100% of 3D-T2-FFE images in the superior and inferior parts of the parotid glands, respectively. 3D-T2-FFE can provide detailed morphological information on the nerve in relation to adjacent parotid gland structures and tumors before surgery.

Development and Validation of a Risk Prediction Model for Postoperative Pulmonary Infection in Renal Transplant Patients.

This study aim to analyze the AAC predicts TIMI and Killip scores based on chest X-ray examinations in patients with STEMI. This study is an observational analytical research with a diagnostic test design on 100 STEMI patients who had chest X-rays at Adam Malik Hospital and were evaluated for aortic arch calcification (AAC) scores semi-quantitatively. Analyzing the prediction of TIMI and Killip scores based on aortic arch calcification. Based on the analysis results, AAC in predicting TIMI score has a sensitivity of 48.6% and a specificity of 72.3% (AUC: 0.644). The Killip score has a sensitivity of 37.1% and a specificity of 78.5% (AUC: 0.562). The Positive Predictive Values of TIMI and Killip, respectively, are 64.6% and 64.2% and Negative Predictive Values are 57.5% and 54.5%. AAC is inadequate in predicting TIMI and Killip scores for patients with STEMI.

Accuracy of ICD and SNOMED search strategies for adverse events following COVID-19 vaccination: Analysis of hospital administrative data.

Intelligent identification of osteoporosis on hip X-rays using vision transformer.

Refining the Baveno VII criteria for clinically significant portal hypertension: An individual patient data meta-analysis.

A Simple Laboratory-based Machine Learning Model Accurately Predicts Advanced Liver Fibrosis in Metabolic Dysfunction-associated Steatotic Liver Disease Patients.

Performance of the Australian hospital-acquired complication algorithm for detecting hospital-onset bloodstream infections.