OuluREPO University of Oulu repository
Not a member yet
54977 research outputs found
Sort by
"Ja eikähän sulle tuu minkäällaista käsiallaa" − kirjoittamiskäsitykset Kalle Päätalon Iijoki-sarjassa
No full textCognitive Measures in Developmental Language Disorder Classification in Monolingual and Bilingual Children: A Machine Learning Approach
No full textAbstract
Purpose:
We used machine learning (ML) models to investigate the accuracy of a cognitive assessment battery to differentiate developmental language disorder (DLD) and typical development (TD) in monolingual and sequential bilingual children. Additionally, we tested how the model trained on monolingual children can classify bilingual children and examined the relative importance of the different linguistic and nonlinguistic tasks in the classifications.
Method:
The participants were 4- to 7-year-old monolingual and sequential bilingual children with DLD (n = 167) or TD (n = 127) from the Helsinki longitudinal SLI study. The assessment battery included standardized tasks used to measure different domains of language and other cognition. To investigate the ability of the tasks to classify mono- and bilingual children into having DLD or TD, we used a random forest ML classification model.
Results:
The cognitive assessment battery classified DLD/TD well in the monolingual group (91.3%) and fairly well in the bilingual group (84.7%). However, the model trained with monolingual data was not accurate in the bilingual group (66.0%). The best tasks for classifying DLD and TD reflected language processing and verbal reasoning in both mono- and bilingual children. The nonlinguistic tasks did not considerably improve the classification.
Conclusions:
This study is among the first to employ ML methods for DLD classification and presents a cognitive assessment battery for detecting DLD in mono- and bilingual children. The current results show that bilingual children's performance should not be compared to monolingual standards. The role of the nonlinguistic functions remains unclear.Abstract
Purpose:
We used machine learning (ML) models to investigate the accuracy of a cognitive assessment battery to differentiate developmental language disorder (DLD) and typical development (TD) in monolingual and sequential bilingual children. Additionally, we tested how the model trained on monolingual children can classify bilingual children and examined the relative importance of the different linguistic and nonlinguistic tasks in the classifications.
Method:
The participants were 4- to 7-year-old monolingual and sequential bilingual children with DLD (n = 167) or TD (n = 127) from the Helsinki longitudinal SLI study. The assessment battery included standardized tasks used to measure different domains of language and other cognition. To investigate the ability of the tasks to classify mono- and bilingual children into having DLD or TD, we used a random forest ML classification model.
Results:
The cognitive assessment battery classified DLD/TD well in the monolingual group (91.3%) and fairly well in the bilingual group (84.7%). However, the model trained with monolingual data was not accurate in the bilingual group (66.0%). The best tasks for classifying DLD and TD reflected language processing and verbal reasoning in both mono- and bilingual children. The nonlinguistic tasks did not considerably improve the classification.
Conclusions:
This study is among the first to employ ML methods for DLD classification and presents a cognitive assessment battery for detecting DLD in mono- and bilingual children. The current results show that bilingual children's performance should not be compared to monolingual standards. The role of the nonlinguistic functions remains unclear
Dental students' knowledge, perceptions, and educational needs regarding artificial intelligence: a multinational cross-sectional survey
No full textAbstract
Aims:
To explore undergraduate dental students’ AI knowledge, perceptions, and concerns, and to identify their educational needs based on these findings.
Methods:
A cross-sectional, anonymous survey was conducted using a 30-item online questionnaire distributed to dental schools across multiple countries. The survey employed an exploratory, observational approach with convenience and snowball sampling methods. The population included dental students from all academic semesters, and participation in the survey was voluntary. The questionnaire consisted of multiple-choice and Likert-scale questions organized into five sections: consent form, demographic data, knowledge/awareness, perceptions/attitudes, and ethics-related questions. Data were analysed using Jamovi and R. Descriptive statistics summarised the demographic characteristics and responses to survey questions. Non-parametric correlation analysis was employed as a primary measure of association for relationships between ordinal variables. For regression analyses, ordinal logistic regression models were constructed to identify predictors for specific outcomes. For each Likert scale question, an ordinal logistic regression model was constructed (dependent variable), with the knowledge questions score as a covariate and the nominal answered questions as factors.
Results:
508 students completed the questionnaire. Most students (76.2%) agreed they understood what AI entails, and 67.4% were familiar with generic AI tools; however, only 34.7% knew AI's dental applications. 70.3% supported AI education during undergraduate studies, favoring case-based teaching, and 53.7% felt their current education had not adequately prepared them for AI technologies. Students declared that AI would be beneficial in diagnostic analysis (64.5%), enhance clinical practice (69.5%), and improve patient care (60.4%). Also, 41.7% believed that AI would cause a reduction in professionals’ skills and dehumanize healthcare (29.2%). 3/4 students agreed that AI ethics should be taught from a multidisciplinary perspective, and 65.3% declared AI in healthcare should be legally regulated.
Conclusions:
This study establishes baseline data on dental students' AI knowledge and educational requirements across multiple countries. Despite general AI familiarity, understanding of dental applications remains limited. The results highlight the need for structured AI education programs tailored to students’ knowledge gaps and learning preferences. Dental students’ understanding and perceptions of AI can effectively guide the identification of their learning needs and inform curriculum integration.Abstract
Aims:
To explore undergraduate dental students’ AI knowledge, perceptions, and concerns, and to identify their educational needs based on these findings.
Methods:
A cross-sectional, anonymous survey was conducted using a 30-item online questionnaire distributed to dental schools across multiple countries. The survey employed an exploratory, observational approach with convenience and snowball sampling methods. The population included dental students from all academic semesters, and participation in the survey was voluntary. The questionnaire consisted of multiple-choice and Likert-scale questions organized into five sections: consent form, demographic data, knowledge/awareness, perceptions/attitudes, and ethics-related questions. Data were analysed using Jamovi and R. Descriptive statistics summarised the demographic characteristics and responses to survey questions. Non-parametric correlation analysis was employed as a primary measure of association for relationships between ordinal variables. For regression analyses, ordinal logistic regression models were constructed to identify predictors for specific outcomes. For each Likert scale question, an ordinal logistic regression model was constructed (dependent variable), with the knowledge questions score as a covariate and the nominal answered questions as factors.
Results:
508 students completed the questionnaire. Most students (76.2%) agreed they understood what AI entails, and 67.4% were familiar with generic AI tools; however, only 34.7% knew AI's dental applications. 70.3% supported AI education during undergraduate studies, favoring case-based teaching, and 53.7% felt their current education had not adequately prepared them for AI technologies. Students declared that AI would be beneficial in diagnostic analysis (64.5%), enhance clinical practice (69.5%), and improve patient care (60.4%). Also, 41.7% believed that AI would cause a reduction in professionals’ skills and dehumanize healthcare (29.2%). 3/4 students agreed that AI ethics should be taught from a multidisciplinary perspective, and 65.3% declared AI in healthcare should be legally regulated.
Conclusions:
This study establishes baseline data on dental students' AI knowledge and educational requirements across multiple countries. Despite general AI familiarity, understanding of dental applications remains limited. The results highlight the need for structured AI education programs tailored to students’ knowledge gaps and learning preferences. Dental students’ understanding and perceptions of AI can effectively guide the identification of their learning needs and inform curriculum integration
Hydrogen reduction kinetics of different cobalt compounds
No full textAbstract
Hydrogen reduction kinetics of cobalt oxides, carbonates, and hydroxides were investigated under both isothermal (300, 400, and 500 °C) and non-isothermal (30 °C to 700 °C) conditions, using thermogravimetric analysis (TGA). Isothermal data was fitted to various solid-state kinetic models, revealing that the most suitable model in all temperatures for the reduction of studied compounds was R2 and AE1 in the case of Co2CO3(OH)2 at 300 °C. The calculated activation energies ranged between 39.8 and 48.4 kJ/mol for the two Co3O4, 11.5–26.3 kJ/mol for CoCO3, 46.8–91.7 kJ/mol for Co2CO3(OH)2, 6.9–10.7 for Co(OH)2 with a uniform PSD, and 24.5–28.9 kJ/mol for Co(OH)2 with a non-uniform PSD. Isoconversional methods (Vyazovkin, Friedman, Kissinger-Akahira-Sunose, Flynn-Wall-Osawa) were applied to the non-isothermal data of four different heating rates (5, 10, 15, and 20 °C/min). The derived activation energies, as a function of conversion extent, indicated that the thermal decomposition of cobalt carbonates and hydroxides generally required higher activation energies than the subsequent reduction of CoO (approximately 40–60 kJ/mol). Decomposition of carbonates and hydroxides was found to precede the reduction of resulting CoO under a dynamic heating regime. The study showed that reduction kinetics are highly influenced by characteristics of the cobalt compounds, as well as the used process parameters.Abstract
Hydrogen reduction kinetics of cobalt oxides, carbonates, and hydroxides were investigated under both isothermal (300, 400, and 500 °C) and non-isothermal (30 °C to 700 °C) conditions, using thermogravimetric analysis (TGA). Isothermal data was fitted to various solid-state kinetic models, revealing that the most suitable model in all temperatures for the reduction of studied compounds was R2 and AE1 in the case of Co2CO3(OH)2 at 300 °C. The calculated activation energies ranged between 39.8 and 48.4 kJ/mol for the two Co3O4, 11.5–26.3 kJ/mol for CoCO3, 46.8–91.7 kJ/mol for Co2CO3(OH)2, 6.9–10.7 for Co(OH)2 with a uniform PSD, and 24.5–28.9 kJ/mol for Co(OH)2 with a non-uniform PSD. Isoconversional methods (Vyazovkin, Friedman, Kissinger-Akahira-Sunose, Flynn-Wall-Osawa) were applied to the non-isothermal data of four different heating rates (5, 10, 15, and 20 °C/min). The derived activation energies, as a function of conversion extent, indicated that the thermal decomposition of cobalt carbonates and hydroxides generally required higher activation energies than the subsequent reduction of CoO (approximately 40–60 kJ/mol). Decomposition of carbonates and hydroxides was found to precede the reduction of resulting CoO under a dynamic heating regime. The study showed that reduction kinetics are highly influenced by characteristics of the cobalt compounds, as well as the used process parameters
Building working life cooperation in the working life studies of the degree program in educational sciences
No full textTässä pro gradu -tutkielmassa käsitellään työelämäopintojen merkitystä työelämäyhteistyön rakentajina Oulun yliopiston yleisen kasvatustieteen ja oppimistieteiden tutkinto-ohjelmissa. Tutkimuksen tavoitteena on selvittää, miten työelämäyhteistyö rakentuu opetussuunnitelman ja opiskelijoiden kokemusten kautta sekä missä määrin opetussuunnitelman tavoitteet vastaavat opiskelijoiden kokemuksia työelämäopinnoista. Aihe on ajankohtainen kasvatustieteilijöiden työllistymistä sekä koulutuksen suunnittelua ajatellen.
Tutkimuksen pääkysymystä ”Miten työelämäyhteistyö rakentuu kasvatustieteiden tutkinto-ohjelman työelämäopinnoissa ja opiskelijoiden kokemuksissa?” lähestytään yliopistokoulutuksen, työelämäyhteistyöhön, koulutussuunnitteluun sekä ennakointiin pohjautuvan teoreettisen viitekehyksen pohjalta. Työelämäopinnoilla viitataan tässä tutkimuksessa työharjoitteluun sekä niihin teoreettisiin opintojaksoihin, jotka tarjoavat konkreettista yhteyttä työelämään esimerkiksi työpaikkavierailuiden kautta.
Tutkimus on toteutettu laadullisena tutkimuksena. Tutkimusaineisto muodostuu dokumenttiaineistosta sekä sähköisestä kyselylomakkeesta. Aineistona toimii Oulun yliopiston opetussuunnitelma lukuvuodelta 2024–2025 sekä marraskuussa 2025 kerätty Microsoft Forms -kysely. Kyselyyn vastasi Oulun yliopistosta 11 yleisen kasvatustieteen tai kasvatuspsykologian pääaineen opiskelijaa tai valmistunutta. Tutkimusaineisto analysoitiin teoriaohjaavalla sisällönanalyysillä, jossa opiskelijoiden kokemuksia tarkasteltiin suhteessa opetussuunnitelman tavoitteisiin ja teoreettisiin lähtökohtiin.
Tutkimuksen tulokset osoittivat, että työelämäopinnot koetaan opiskelijoiden näkökulmasta merkitykselliseksi osaksi opintoja, mutta niiden määrä ja jatkuvuus koetaan osin riittämättömiksi. Erityisesti työharjoittelu näyttäytyi keskeisenä opintojaksona teorian ja käytännön yhdistämisessä ja asiantuntijuuden kehittymisessä. Tulosten perusteella työelämäopintojen kehittämisessä tulisi kiinnittää huomiota niiden systemaattisuuteen, varhaisempaan ajoittumiseen opinnoissa sekä opiskelijoiden kokemusten hyödyntämiseen opetussuunnitelmatyössä. Tutkimuksen tulokset eivät ole laajasti yleistettävissä, mutta ne tarjoavat merkityksellistä ennakointitietoa koulutuksen kehittämisen tueksi erityisesti geneerisille aloille, joissa työelämäyhteyksien merkitys korostuu
Sequential coupling of HBV-light and MODFLOW models to assess water table variations under the future climate in agricultural peatlands
No full textAbstract
Understanding water table (WT) variation is essential in peatland agriculture as the WT position is a major determinant of cultivation conditions and greenhouse gas emissions. To simulate past and future WT, we developed a sequential modelling approach by coupling two well-known models: HBV-light for cold climate hydrologic processes and MODFLOW for saturated subsurface flow conditions. We tested this approach in an agricultural peatland on Norway's west coast under two drainage configurations: a traditional pipe-drained field and an adjacent inverted peatland with a complex drainage system. In peat inversion, mineral soil excavated from below the peat is subsequently placed on top of the peat to form a protective cover layer aimed at reducing peat decomposition. Our modelling approach captured the daily WT dynamics of the fields between July 2022 and January 2025. The modelling results indicate that lateral flow dominates over the vertical flow in both drainage systems. Under future climate conditions between 2025 and 2100, the buried peat in the inverted peatland is expected to remain saturated for 25% of the time, whereas the peat in the pipe-drained field will be waterlogged rarely. Our modelling approach to simulate WT in managed peatlands can be set up relatively easily.Abstract
Understanding water table (WT) variation is essential in peatland agriculture as the WT position is a major determinant of cultivation conditions and greenhouse gas emissions. To simulate past and future WT, we developed a sequential modelling approach by coupling two well-known models: HBV-light for cold climate hydrologic processes and MODFLOW for saturated subsurface flow conditions. We tested this approach in an agricultural peatland on Norway's west coast under two drainage configurations: a traditional pipe-drained field and an adjacent inverted peatland with a complex drainage system. In peat inversion, mineral soil excavated from below the peat is subsequently placed on top of the peat to form a protective cover layer aimed at reducing peat decomposition. Our modelling approach captured the daily WT dynamics of the fields between July 2022 and January 2025. The modelling results indicate that lateral flow dominates over the vertical flow in both drainage systems. Under future climate conditions between 2025 and 2100, the buried peat in the inverted peatland is expected to remain saturated for 25% of the time, whereas the peat in the pipe-drained field will be waterlogged rarely. Our modelling approach to simulate WT in managed peatlands can be set up relatively easily
Brief Diagnostic Criteria for Temporomandibular Disorders (Tmd): Enhancing Sensitivity in Diagnosing Headache Attributed to Tmd (A Multi-Centre Study)
No full textAbstract
Background:
The brief Diagnostic Criteria for Temporomandibular Disorders (bDC/TMD) was developed to simplify the original DC/TMD for wider clinical use. While its diagnostic accuracy for most painful TMDs is acceptable, the sensitivity for headache attributed to TMD (HaTMD) was reported to be poor.
Objectives:
To improve the diagnostic sensitivity of HaTMD within the bDC/TMD framework by reintroducing selected examination items from the original DC/TMD protocol.
Methods:
This retrospective multicentre study used data from Finland and Israel. The cohort included 114 individuals previously diagnosed with HaTMD according to the DC/TMD and with myalgia and/or arthralgia diagnoses in both DC/TMD and bDC/TMD. Four examination items excluded from the bDC/TMD—E1b (headache location in temple), E4c (familiar headache on assisted opening), E5a/b (lateral movements) and E5c (protrusive movements)—were reintroduced individually. Four calibrated examiners reassessed each modified dataset. Inter-examiner reliability (Cohen's kappa) and diagnostic sensitivity were calculated using DC/TMD as the gold standard.
Results:
Inter-examiner reliability for HaTMD diagnosis was almost perfect (κ = 0.81–1.00) across all items. Sensitivity improved markedly from the previously reported 0.16–0.38 to 0.82 (E1b)–0.90 (E5c). Item E1b (temple headache confirmation) was present in 95% of Finnish and 86% of Israeli cases, identifying it as the most representative finding.
Conclusion:
Reintroducing item E1b into the bDC/TMD examination substantially increases the diagnostic sensitivity for HaTMD while maintaining brevity. Refinement of the painful TMD diagnostic decision tree and prospective validation of the modified bDC/TMD are recommended to ensure reliability and clinical applicability.Abstract
Background:
The brief Diagnostic Criteria for Temporomandibular Disorders (bDC/TMD) was developed to simplify the original DC/TMD for wider clinical use. While its diagnostic accuracy for most painful TMDs is acceptable, the sensitivity for headache attributed to TMD (HaTMD) was reported to be poor.
Objectives:
To improve the diagnostic sensitivity of HaTMD within the bDC/TMD framework by reintroducing selected examination items from the original DC/TMD protocol.
Methods:
This retrospective multicentre study used data from Finland and Israel. The cohort included 114 individuals previously diagnosed with HaTMD according to the DC/TMD and with myalgia and/or arthralgia diagnoses in both DC/TMD and bDC/TMD. Four examination items excluded from the bDC/TMD—E1b (headache location in temple), E4c (familiar headache on assisted opening), E5a/b (lateral movements) and E5c (protrusive movements)—were reintroduced individually. Four calibrated examiners reassessed each modified dataset. Inter-examiner reliability (Cohen's kappa) and diagnostic sensitivity were calculated using DC/TMD as the gold standard.
Results:
Inter-examiner reliability for HaTMD diagnosis was almost perfect (κ = 0.81–1.00) across all items. Sensitivity improved markedly from the previously reported 0.16–0.38 to 0.82 (E1b)–0.90 (E5c). Item E1b (temple headache confirmation) was present in 95% of Finnish and 86% of Israeli cases, identifying it as the most representative finding.
Conclusion:
Reintroducing item E1b into the bDC/TMD examination substantially increases the diagnostic sensitivity for HaTMD while maintaining brevity. Refinement of the painful TMD diagnostic decision tree and prospective validation of the modified bDC/TMD are recommended to ensure reliability and clinical applicability