University of Turku

This study presents a machine learning (ML) framework for predicting heat transfer in magnetohydrodynamic (MHD) mixed convection within a complex V-shaped cavity filled with a nano-enhanced phase change material (NEPCM) suspension. Accurate computational fluid dynamics (CFD) simulations are essential for understanding heat transfer mechanisms in such systems, but generating comprehensive data through high-fidelity models remains computationally expensive. To address this challenge, we develop an integrated ML approach that combines synthetic data generation, physics-informed feature engineering, and optimized ensemble boosting. The methodology first augments a limited 34-sample CFD dataset to 2034 samples using Latin Hypercube Sampling with Radial Basis Function interpolation. Next, 14 physics-based features are engineered to encode the underlying physical phenomena. Finally, hyperparameters of three gradient boosting models—eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), and Categorical Boosting (CatBoost)–are optimized via cross-validation. The framework predicts the average Nusselt number ( N u avg ) and average kinetic energy ( K E avg ) from seven geometric and operational inputs. CatBoost achieved optimal performance for N u avg ( R 2 = 0 . 9745 , mean absolute percentage error = 1.27%), while XGBoost excelled for K E avg ( R 2 = 0 . 9920 , mean absolute percentage error = 1.04%). The novelty of this work lies in its ability to generalize across different output variables and significantly reduce computational cost, enabling rapid design optimization and in-depth parametric analysis. This generalizable approach reduces computation time from hours to milliseconds, facilitating efficient design optimization and in-depth parametric studies. • ML framework predicts heat transfer in MHD mixed convection of NEPCM suspension. • Combines physics-informed feature engineering with optimized ensemble boosting. • CatBoost best for Nusselt number (R 2 = 0.9745), XGBoost for kinetic energy (R 2 = 0.9920). • Synthetic data generation expands limited CFD dataset from 34 to 2034 samples. • Reduces simulation time from hours to milliseconds for efficient design exploration.

Bladder exstrophy and epispadias complex is a rare congenital anomaly with significant medical and psychosocial implications. We employed a dual approach to evaluate the prevalence and risk factors of psychiatric morbidity in bladder exstrophy and epispadias patients, by combining a systematic review with a retrospective national registry study. To evaluate the prevalence and risk factors of psychiatric morbidity in bladder exstrophy and epispadias complex. We conducted a systematic literature review following Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines, including studies on psychiatric disorders or symptoms and quality of life in bladder exstrophy and epispadias patients of all ages. Article quality was assessed using the Newcastle-Ottawa Scale. Prevalence data for psychiatric conditions and scores from psychiatric and quality-of-life questionnaires were extracted. Additionally, all bladder exstrophy and epispadias patients born in Finland 2001-2006 were identified from national registries. A matched and randomized control group without congenital malformations was selected from the same population. We evaluated the prevalence and risk factors for psychiatric diagnoses in this cohort. Of the 3850 retrieved results, 30 studies with 1179 participants were included. The overall prevalence of psychiatric morbidity was 31 %, [95 % CI 17 %-47 %]. Psychiatric morbidity was highest in adolescents, while morbidity in adults was comparable to the general population. In the Finnish cohort, 80 % (n = 16/20) of bladder exstrophy and epispadias patients had a psychiatric diagnosis compared to 26 % (n = 21/80) of the controls (OR 11.2, [95 % CI 3.37-37.4], p < 0.001). The severity of the anomaly, number of surgeries, maternal unemployment, or maternal psychiatric diagnosis were not significantly associated with increased psychiatric morbidity. Bladder exstrophy and epispadias is associated with an elevated risk of psychiatric morbidity. Our dual approach highlights the need for systematic mental health assessment and support in this population.

To assess the effectiveness of a six‑week online exercise program on self‑assessed urinary incontinence (UI) symptoms and their impact on daily life among postpartum women immediately and six months after the intervention. A quasi-experimental pre-post design without a control group was used. Participants (n=297) completed a six-week online exercise program focused on core and pelvic floor strengthening. Data were collected via electronic questionnaires at baseline, post-intervention, and six months later. UI symptoms were measured with the UDI-6 and symptoms impact on daily life with the IIQ-7. Changes across timepoints were analyzed using linear mixed models. Attrition increased across measurement points, with 89 women completing the six‑month follow‑up questionnaire. UI symptoms decreased statistically significantly over time (p<0.001) from baseline (mean 18.8; 95% CI 16.9, 20.7) to posttest (13.2; 95% CI 11.1, 15.4). At the 6-month follow-up, the scores had increased slightly (14.1; 95% CI 11.4, 16.9) but were still lower than at the baseline. For the perceived impact of UI on daily activities, the overall change was not statistically significant (p=0.269). Mean (95% CI) IIQ-7 scores were 8.5 (6.9, 10.0) at baseline, 6.6 (4.7, 8.5) at posttest, and 6.7 (3.9, 9.4) at 6‑month follow‑up. Time since childbirth, BMI, mode of birth, and dysfunctions or symptoms related to pelvic floor and core muscles were associated with variation in studied outcomes. A six‑week self‑directed online exercise program was associated with a reduction in postpartum UI symptoms, with improvements largely maintained at six months. Individual characteristics were associated with symptom severity and responsiveness, highlighting the need for tailored postpartum rehabilitation and further controlled studies.

Learning to teach during the teaching practicum is a cognitively demanding, emotionally charged, and identity-shaping process for student teachers, requiring meaningful changes that necessitate mentoring from both university-based teacher educators (UBTEs) and school-based teacher educators (SBTEs). Grounded in Transformative Learning Theory, this study examined the mentoring experiences of two groups of teacher educators in supporting learning to teach. Using a “difference-within-similarity” analytical approach, we identified nine common traits of mentoring competence (MC) across three dimensions (cognitive, social, and emotional). We revealed distinct tendencies in how UBTEs and SBTEs enact their MC. The findings not only highlight the richness and complexity of MC as a multidimensional construct, challenging the binary view that characterizes UBTEs' mentoring as theoretical and SBTEs’ as practical, but also offer practical implications for optimizing labor division and fostering more synergistic, complementary triadic mentoring. • Mentoring practice can be seen as supporting transformative learning in practicum. • Mentoring competence (MC) in cognitive, social, emotional dimensions was enacted in practice. • University- and school-based teacher educators' MC were analyzed and compared. • Nine common traits of MC between the two groups were identified. • The distinctions in MC between the two groups are not theory-practice dichotomous.

Gastric cancer is the fifth most common cancer and the fifth leading cause of cancer-related death worldwide. Its poor prognosis primarily results from a late diagnosis and the lack of effective treatments for advanced disease. Thus, we aimed to identify new prognostic serum biomarkers to aid clinical decision-making. Our patient cohort consisted of 240 individuals who underwent surgery for histologically verified gastric adenocarcinoma in the Department of Surgery at Helsinki University Hospital between 2000 and 2009. To determine the serum protein concentrations of cytokines and growth factors, we utilized Bio-Rad’s premixed Bio-Plex Pro Human Cytokine 27- and 21-plex assay kits. Among the 48 biomarkers we analyzed, three emerged as statistically significant prognostic markers for disease-specific survival using the Cox proportional hazards univariate analysis: C-X-C motif chemokine ligand 12 (CXCL12) (hazard ratio [HR] 0.39, 95% confidence interval [CI] 0.23–0.63, p < 0.001), stem cell factor (HR 0.38, 95% CI 0.19–0.77, p = 0.007), and eotaxin (HR 0.57, 95% CI 0.37–0.89, p = 0.013). Our multivariate survival analysis revealed that, among the 48 biomarkers analyzed, CXCL12 and eotaxin served as independent prognostic markers among gastric cancer patients. The prognostic effect of inflammatory serum biomarkers in gastric cancer may provide new insights into the immunological microenvironment of disease.

Cardiac organ damage (OD) is associated with increased risk of cardiovascular disease. However, limited knowledge exists on cardiac OD in young patients with cryptogenic ischaemic stroke (CIS). To explore prevalence and covariates of cardiac OD in patients with CIS compared to controls participating in the SECRETO study. We analysed data from 427 patients with CIS aged <50 years and 361 age- and sex-matched controls. OD was defined as presence of abnormal left ventricular (LV) geometry (LV hypertrophy or concentric remodelling) or left atrial enlargement (LAE) assessed by echocardiography, using sex-specific threshold values. Compared to controls, patients had higher prevalences of obesity and tobacco smoking, patent foramen ovale (PFO) (52% vs. 25%) and abnormal LV geometry (12% vs. 7%, all p < 0.01), while presence of LAE did not differ. In multivariable analyses, CIS was associated with presence of abnormal LV geometry (odds ratio 1.86 [95% confidence interval 1.09-3.20], p=0.024). In separate multivariable analyses in patients only, obesity was associated with both abnormal LV geometry and LAE (both p < 0.01), and higher systolic blood pressure only with presence of abnormal LV geometry (p < 0.001). No significant association with PFO was found. In young patients with CIS participating in the SECRETO study, abnormal LV geometry was more prevalent compared to age- and sex-matched controls and associated with presence of higher blood pressure and obesity. The results point to the importance of blood pressure and weight control in CIS to prevent progression of cardiac OD and recurrent cardiovascular events.