24 research outputs found
Prediction of Concurrent Hypertensive Disorders in Pregnancy and Gestational Diabetes Mellitus Using Machine Learning Techniques
No full textGestational diabetes mellitus and hypertensive disorders in pregnancy are serious maternal health conditions with immediate and lifelong mother-child health consequences. These obstetric pathologies have been widely investigated, but mostly in silos, while studies focusing on their simultaneous occurrence rarely exist. This is especially the case in the machine learning domain. This retrospective study sought to investigate, construct, evaluate, compare, and isolate a supervised machine learning predictive model for the binary classification of co-occurring gestational diabetes mellitus and hypertensive disorders in pregnancy in a cohort of otherwise healthy pregnant women. To accomplish the stated aims, this study analyzed an extract (n=4624, n_features=38) of a labelled maternal perinatal dataset (n=9967, n_fields=79) collected by the PeriData.Net® database from a participating community hospital in Southeast Wisconsin between 2013 and 2018. The datasets were named, “WiseSample” and “WiseSubset” respectively in this study. Thirty-three models were constructed with the six supervised machine learning algorithms explored on the extracted dataset: logistic regression, random forest, decision tree, support vector machine, StackingClassifier, and KerasClassifier, which is a deep learning classification algorithm; all were evaluated using the StratifiedKfold cross-validation (k=10) method. The Synthetic Minority Oversampling Technique was applied to the training data to resolve the class imbalance that was noted in the sub-sample at the preprocessing phase. A wide range of evidence-based feature selection techniques were used to identify the best predictors of the comorbidity under investigation. Multiple model performance evaluation metrics that were employed to quantitatively evaluate and compare model performance quality include accuracy, F1, precision, recall, and the area under the receiver operating characteristic curve. Support Vector Machine objectively emerged as the most generalizable model for identifying the gravidae in WiseSubset who may develop concurrent gestational diabetes mellitus and hypertensive disorders in pregnancy, scoring 100.00% (mean) in recall. The model consisted of 9 predictors extracted by the recursive feature elimination with cross-validation with random forest. Finding from this study show that appropriate machine learning methods can reliably predict comorbid gestational diabetes and hypertensive disorders in pregnancy, using readily available routine prenatal attributes. Six of the nine most predictive factors of the comorbidity were also in the top 6 selections of at least one other feature selection method examined. The six predictors are healthy weight prepregnancy BMI, mother’s educational status, husband’s educational status, husband’s occupation in one year before the current pregnancy, mother’s blood group, and mother’s age range between 34 and 44 years. Insight from this analysis would support clinical decision making of obstetric experts when they are caring for 1.) nulliparous women, since they would have no obstetric history that could prompt their care providers for feto-maternal medical surveillance; and 2.) the experienced mothers with no obstetric history suggestive of any of the disease(s) under this study. Hence, among other benefits, the artificial-intelligence-backed tool designed in this research would likely improve maternal and child care quality outcomes
Recommended from our members
Physical Activity of Adults with Intellectual Disabilities: Feasibility and validity of measuring physical activity and physical activity intention and their relationship to the spectrum of Adults with Intellectual Disabilities
Get PDFIn comparison to the wider population people with intellectual disabilities (ID) experience higher rates of morbidity and mortality associated with inactive lifestyles, however little is known about the pattern of their physical activity (PA). To address the inadequacies that currently exist around PA and those with ID, we need to better understand their PA levels to identify those who are inactive and the related factors. The four-part studies in this thesis aimed to establish PA levels of adults with intellectual disabilities (awID) across the spectrum. It also aimed to investigate the readiness to change PA behaviour in this population. The first study, a systematic review of evidence for PA levels among awID found that in 15 eligible studies, comprising 3159 subjects, only 9% achieved PA guidelines. ID severity, living in care, gender, and age were independently significantly correlated with PA levels, the strongest predictor being ID severity (Beta 0.63, p<0.001). These findings were limited in context; none objectively measured PA in people with profound ID. This informed the second feasibility study, investigating assessment of PA across the spectrum using wrist-worn, 7-day accelerometers and the short form of the International Physical Activity Questionnaire (IPAQ-s). This revealed the difficulty in recruiting subjects across the full spectrum, and a difference in response rates between homes (16%) and residential-homes (4%): χ2 (1) = 7.7, p <0.05. Nonetheless, 20 participants were recruited, and both PA measures were used across the ID spectrum, including in those with profound ID (n=5). The third study investigated concurrent validity between the two PA measures; suggesting a substantial agreement between the accelerometer and IPAQ-s (k= 0.6, n = 16, p < .05) in identifying active participants, and a significant positive correlation between the sedentary time by the IPAQ-s and accelerometer (r = 0.50, n = 17, p = 0.04). This has implications both clinically and in the research environment as the study showed for the first time that the IPAQ-s could be used effectively to identify those that are active and measure sedentary behaviour across ID spectrum.
In the final study, PA levels measured across the spectrum using the IPAQ-s suggested low PA (39% active) and high sedentary behaviour (82% sedentary >3 hours/day) among awID (n=82). Multiple linear regression revealed that high ID severity is a significant predictor of low PA (Beta -0.28, p<0.05) and high sedentary behaviour (Beta 0.41, p<0.001). To investigate readiness to change PA behaviour in this group, the author developed a single-item PA intention measure (SPAIM). The construct validity and test-retest reliability of the SPAIM were examined, and the relationship between PA intention and PA levels. The SPAIM was a reliable measure of the PA intention, with a strong, positive correlation between first and repeated measure (rs =0.78, n=35, p<0.001). Additionally, linear regression revealed that PA intention significantly predicts daily sedentary hours (Beta -0.34, p<0.01).
This research showed that it is possible to measure PA levels across the ID spectrum, but the choice of measure is a major factor. It also highlighted a need for better engagement with carers and residential-home managers, to improve research participation by those with severe to profound ID. Moreover, although only a minority of participants were active, the majority had a positive PA intention. The amount of time spent being sedentary is the only variable related to PA intention. Overall, the evidence suggests that intervention studies aimed at reducing sedentary behaviour are urgently needed in this group. The sedentary hours/day could be a primary target for public health efforts in this population
Rebuilding the social fabric: challenging and transforming unwarranted influences in the educational institutions in Nigeria
No full textHow does corruption become socialized? Corruption is a major global problem. The effects are devastating. Corruption undermines rule of law, breaks down the social fabric of society, erodes morality and positive values systems. It significantly undermines public trust in institutions and their leaders. On a societal level, the pervasiveness of corruption within a society can lead to an increase in negative structural and systemic practices as well as encourage individual corrupt behavior that ultimately erodes public morality. Considering these reasons and more, this dissertation analyzes the role academic dishonesty in institutions of education plays in normalizing deviant behavior, which may result in the socialization of corruption.Ph.D.Includes bibliographical referencesby Yetunde A. Odugbesa
QuizMap: Open social student modeling and adaptive navigation support with TreeMaps
Get PDFIn this paper, we present a novel approach to integrate social adaptive navigation support for self-assessment questions with an open student model using QuizMap, a TreeMap-based interface. By exposing student model in contrast to student peers and the whole class, QuizMap attempts to provide social guidance and increase student performance. The paper explains the nature of the QuizMap approach and its implementation in the context of self-assessment questions for Java programming. It also presents the design of a semester-long classroom study that we ran to evaluate QuizMap and reports the evaluation results. © 2011 Springer-Verlag Berlin Heidelberg
Experimental Validation of Contextual Variables for Research Resources Recommender System
No full text
Disinfection By-Product Formation in Drinking Water Treated with Chlorine Following
No full textI hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final revisions, as accepted by my examiners. I understand that my thesis may be made electronically available to the public. ii As far back as the early 1900’s when it was discovered that water could be a mode of transmitting diseases, chlorine was used to disinfect water. In the 1970’s, the formation of disinfection by-products (DBPs) from the reaction of chlorine with natural organic matter was discovered. Since then there have been various studies on alternative disinfectants that could inactivate microorganisms and at the same time form less or no disinfection by-products. More recently the ultraviolet (UV) irradiation has been used to both disinfect and remove organic contaminants in drinking water. Though the use of UV irradiation has been found to be very effective in the inactivation of microorganisms, it does not provide a residual effect to maintain the water’s microbial quality in the distribution system. Due to this,
Exploring young students’ attitude towards coding and its relationship with STEM career interest
Get PDFDATA AVAILABILITY :
The datasets presented in this study are not publicly available for privacy reasons. The data is available from the corresponding author on reasonable request.This paper presents findings of an investigation on students’ attitudes towards coding and its relationship with interest in STEM-related careers. A concurrent mixed-method research design involving a pre-intervention-intervention-post-intervention non-equivalent control group was adopted. A sample of 50 grade seven to nine South African students (21 male and 29 female) from Township schools in Johannesburg, South Africa, participated. Quantitative data was gathered using the elementary students’ coding attitude survey and STEM Career Interest survey, while qualitative data was collected through a focus group interview. Interview data was analysed using content analysis, and quantitative data was analysed using multiple correlation analysis and standardized regression coefficients (β). It was found that students’ attitude towards coding was generally positive. A number of correlations between students’ attitude and their STEM career interests were significant at p < .05. Results also revealed that students’ attitudes in terms of coding confidence, coding interest, the social value of coding and perceptions of coders were found to be significant predictors of their interest in a STEM occupation. Based on this, it is argued that encouraging a positive attitude toward coding in students and increasing their self-efficacy can reinforce STEM learning and increase students’ interest in STEM occupations.Open access funding provided by University of Johannesburg.https://link.springer.com/journal/10639hj2024Education Management and Policy StudiesSDG-04:Quality Educatio
Physical activity levels in adults with intellectual disabilities: A systematic review
Get PDFAbstractDespite evidence that inactivity is a major factor causing ill health in people with intellectual disabilities (pwID) there are gaps in our knowledge of their physical activity (PA). To date, there is no published systematic review of their PA levels. Therefore, we performed a systematic review from January–October 2015, comprising studies from across the globe to establish PA levels, determine how they were measured, and what factors influenced PA in adults with intellectual disabilities (awID). Five databases were searched. Studies were included if written in English, peer-reviewed, had primary research data, and measured PA levels of awID. Quality was assessed using a 19-item checklist. Meta-summary of the findings was performed and a meta-analysis of factors influencing PA using multiple regression.Fifteen studies were included consisting of 3159 awID, aged 16–81years, 54% male and 46% female. Only 9% of participants achieved minimum PA guidelines. PA levels were measured using objective and subjective methods. ID severity, living in care, gender, and age were independently significantly correlated with the number of participants achieving PA guidelines with the strongest predictor being ID severity (Beta 0.631, p<0.001). Findings should be in the context that most of the participants were in the mild/moderate range of ID severity and none of the studies objectively measured PA in people with profound ID. To inform measurement and intervention design for improved PA, we recommend that there is an urgent need for future PA studies in awID population to include all disability severity levels. PROSPERO registration number CRD42015016675
