ORCID Profile
0009-0009-9542-1409
Current Organisations
UiT The Arctic University of Norway
,
University Hospital of North Norway
,
University of Sydney
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Springer Nature Switzerland
Date: 2022
Publisher: JMIR Publications Inc.
Date: 08-11-2021
Abstract: ardiovascular diseases, cancers, chronic respiratory diseases, and diabetes are the 4 main noncommunicable diseases. These noncommunicable diseases share 4 modifiable risk factors (tobacco use, harmful use of alcohol, physical inactivity, and unhealthy diet). Short smartphone surveys have the potential to identify modifiable risk factors for in iduals to monitor trends. e aimed to pilot a smartphone-based information communication technology solution to collect nationally representative data, annually, on 4 modifiable risk factors. e developed an information communication technology solution with functionalities for capturing sensitive data from smartphones, receiving, and handling data in accordance with general data protection regulations. The main survey comprised 26 questions: 8 on socioeconomic factors, 17 on the 4 risk factors, and 1 about current or previous noncommunicable diseases. For answers to the continuous questions, a keyboard was displayed for entering numbers there were preset upper and lower limits for acceptable response values. For categorical questions, pull-down menus with response options were displayed. The second survey comprised 9 yes-or-no questions. For both surveys, we used SMS text messaging. For the main survey, we invited 11,000 in iduals, aged 16 to 69 years, selected randomly from the Norwegian National Population Registry (1000 from each of the 11 counties). For the second survey, we invited a random s le of 100 in iduals from each county who had not responded to the main survey. All data, except county of residence, were self-reported. We calculated the distribution for socioeconomic background, tobacco use, diet, physical activity, and health condition factors overall and by sex. he response rate was 21.9% (2303/11,000 women: 1397/2263 61.7%, men: 866/2263, 38.3% missing: 40/2303, 1.7%). The median age for men was 52 years (IQR 40-61) the median age for women was 48 years (IQR 35-58). The main reported reason for nonparticipation in the main survey was that the sender of the initial SMS was unknown. e successfully developed and piloted a smartphone-based information communication technology solution for collecting data on the 4 modifiable risk factors for the 4 main noncommunicable diseases. Approximately 1 in 5 invitees responded thus, these data may not be nationally representative. The smartphone-based information communication technology solution should be further developed with the long-term goal to reduce premature mortality from the 4 main noncommunicable diseases.
Publisher: MDPI AG
Date: 07-02-2023
DOI: 10.20944/PREPRINTS202302.0117.V1
Abstract: Machine Learning (ML) methods have become important to enhance the performance of decision-support predictive models. However, class imbalance is one of the main challenges for developing ML models, because it limits the generalization of these models, and biases the learning algorithms. In this paper, we consider overs ling methods for generating synthetic categorical clinical data aiming to improve the predictive performance in ML models, and the identification of risk factors for cardiovascular diseases (CVDs). We performed a comparative study of several categorical synthetic data generation methods, including Generative Adversarial Networks (GANs). Then, we assessed the impact of combining overs ling strategies and linear and nonlinear supervised ML methods. Lastly, we conducted a post-hoc model interpretability based on the importance of the risk factors. Experimental results show the potential of GAN-based models for generating high-quality categorical synthetic data, yielding probability mass functions that are highly close to real data, maintaining relevant insights, and contributing to increase the predictive performance. The GAN-based model and a linear classifier outperforms other overs ling techniques, improving 2\\% the area under the curve. These results demonstrate the capability of synthetic data to help both in determining risk factors and building models for CVD prediction.
Publisher: E.U. European Publishing
Date: 31-10-2022
DOI: 10.18332/TPC/155287
Publisher: MDPI AG
Date: 23-03-2023
DOI: 10.3390/APP13074119
Abstract: Machine Learning (ML) methods have become important for enhancing the performance of decision-support predictive models. However, class imbalance is one of the main challenges for developing ML models, because it may bias the learning process and the model generalization ability. In this paper, we consider overs ling methods for generating synthetic categorical clinical data aiming to improve the predictive performance in ML models, and the identification of risk factors for cardiovascular diseases (CVDs). We performed a comparative study of several categorical synthetic data generation methods, including Synthetic Minority Overs ling Technique Nominal (SMOTEN), Tabular Variational Autoencoder (TVAE) and Conditional Tabular Generative Adversarial Networks (CTGANs). Then, we assessed the impact of combining overs ling strategies and linear and nonlinear supervised ML methods. Lastly, we conducted a post-hoc model interpretability based on the importance of the risk factors. Experimental results show the potential of GAN-based models for generating high-quality categorical synthetic data, yielding probability mass functions that are very close to those provided by real data, maintaining relevant insights, and contributing to increasing the predictive performance. The GAN-based model and a linear classifier outperform other overs ling techniques, improving the area under the curve by 2%. These results demonstrate the capability of synthetic data to help with both determining risk factors and building models for CVD prediction.
Publisher: JMIR Publications Inc.
Date: 25-02-2022
DOI: 10.2196/33636
Abstract: Cardiovascular diseases, cancers, chronic respiratory diseases, and diabetes are the 4 main noncommunicable diseases. These noncommunicable diseases share 4 modifiable risk factors (tobacco use, harmful use of alcohol, physical inactivity, and unhealthy diet). Short smartphone surveys have the potential to identify modifiable risk factors for in iduals to monitor trends. We aimed to pilot a smartphone-based information communication technology solution to collect nationally representative data, annually, on 4 modifiable risk factors. We developed an information communication technology solution with functionalities for capturing sensitive data from smartphones, receiving, and handling data in accordance with general data protection regulations. The main survey comprised 26 questions: 8 on socioeconomic factors, 17 on the 4 risk factors, and 1 about current or previous noncommunicable diseases. For answers to the continuous questions, a keyboard was displayed for entering numbers there were preset upper and lower limits for acceptable response values. For categorical questions, pull-down menus with response options were displayed. The second survey comprised 9 yes-or-no questions. For both surveys, we used SMS text messaging. For the main survey, we invited 11,000 in iduals, aged 16 to 69 years, selected randomly from the Norwegian National Population Registry (1000 from each of the 11 counties). For the second survey, we invited a random s le of 100 in iduals from each county who had not responded to the main survey. All data, except county of residence, were self-reported. We calculated the distribution for socioeconomic background, tobacco use, diet, physical activity, and health condition factors overall and by sex. The response rate was 21.9% (2303/11,000 women: 1397/2263 61.7%, men: 866/2263, 38.3% missing: 40/2303, 1.7%). The median age for men was 52 years (IQR 40-61) the median age for women was 48 years (IQR 35-58). The main reported reason for nonparticipation in the main survey was that the sender of the initial SMS was unknown. We successfully developed and piloted a smartphone-based information communication technology solution for collecting data on the 4 modifiable risk factors for the 4 main noncommunicable diseases. Approximately 1 in 5 invitees responded thus, these data may not be nationally representative. The smartphone-based information communication technology solution should be further developed with the long-term goal to reduce premature mortality from the 4 main noncommunicable diseases.
Location: China
No related grants have been discovered for Beyza Klein.