Article
Medicine, General & Internal
Jennifer K. Maratt, Thomas F. Imperiale
Summary: This study identified 5 online colorectal cancer risk calculators and evaluated their consistency in predicting risk for different screening scenarios. The results showed that the online models were more consistent in predicting risk for 5-year and 10-year time frames compared with lifetime risk. The National Cancer Institute's Colorectal Cancer Risk Assessment Tool was found to be a reliable calculator for providing 5-year and lifetime risk estimates in a clinical setting.
AMERICAN JOURNAL OF MEDICINE
(2023)
Article
Cardiac & Cardiovascular Systems
Pau Codina, Elisabet Zamora, Wayne C. Levy, Elena Revuelta-Lopez, Andrea Borrellas, Giosafat Spitaleri, German Cediel, Maria Ruiz-Cueto, Elena Canedo, Evelyn Santiago-Vacas, Mar Domingo, David Buchaca, Isaac Subirana, Javier Santesmases, Rafael de la Espriella, Julio Nunez, Josep Lupon, Antoni Bayes-Genis
Summary: This study assessed the changes in predicted mortality risk after 12 months of management in a multidisciplinary HF clinic. The results showed a significant decline in mortality risk and improvement in cardiac function after 12 months, indicating the importance of repeated risk assessment after optimizing HF treatment.
FRONTIERS IN CARDIOVASCULAR MEDICINE
(2022)
Review
Public, Environmental & Occupational Health
Mohammed Abd ElFattah Mohammed Darw Badawy, Lin Naing, Sofian Johar, Sokking Ong, Hanif Abdul Rahman, Dayangku Siti Nur Ashikin Pengiran Tengah, Chean Lin Chong, Nik Ani Afiqah Tuah
Summary: This review aimed to summarise evidence on the key features, usability and benefits of CVD risk calculators using digital platforms. The review found that different guidelines recommend different algorithms for CVD risk prediction, with QRISK(R) being the most accurate calculator. The key features include variables, predictive accuracy, discrimination index, applicability, understandability, and cost-effectiveness.
Article
Computer Science, Artificial Intelligence
Wenzhi Xi, Zhanfeng Li, Xinyuan Song, Hanwen Ning
Summary: This study introduces a novel online portfolio selection method that improves prediction accuracy by utilizing correlation information among assets, and develops a high-dimensional covariance matrix estimation/prediction method to assess instantaneous risk, leading to improved OPS setting and investment performance.
PATTERN RECOGNITION
(2023)
Review
Clinical Neurology
Weijing Qi, Yongjian Wang, Caixia Li, Ke He, Yipeng Wang, Sha Huang, Cong Li, Qing Guo, Jie Hu
Summary: This study systematically reviews and evaluates the risk of bias and the applicability of postpartum depression (PPD) prediction models. The findings suggest that most models have a high risk of bias and limited clinical applicability. Future research should focus on developing predictive models with excellent performance and clinical relevance.
JOURNAL OF AFFECTIVE DISORDERS
(2023)
Article
Multidisciplinary Sciences
Neeloofar Soleimanpour, Maralyssa Bann
Summary: This study assessed the methodologic quality and applicability of clinical risk scoring tools used to guide hospitalization decision-making. The results showed that these tools often do not meet rigorous methodologic standards, lack pertinent details, and raise concerns about validation and generalizability.
Article
Automation & Control Systems
Michael Maiworm, Daniel Limon, Rolf Findeisen
Summary: This study introduces a combination of an output feedback model predictive control scheme and a Gaussian process-based prediction model, which achieves efficient constraint satisfaction and input-to-state stability through online learning.
INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Ignacio Rodriguez-Rodriguez, Maria Campo-Valera, Jose-Victor Rodriguez, Wai Lok Woo
Summary: This research explores IoMT-based methodologies for comprehensive monitoring and short-term prediction of blood glucose levels in diabetic individuals. By combining machine learning techniques with wearable technology, reliable prediction models can be obtained, providing a new approach for precise diabetes management.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Construction & Building Technology
Abdolmajid Erfani, Qingbin Cui
Summary: This study introduces a data driven framework for risk identification using historical data and artificial intelligence techniques, particularly word embedding models, to address the challenges faced by public agencies in risk identification.
AUTOMATION IN CONSTRUCTION
(2022)
Article
Ergonomics
Huiying Mao, Feng Guo, Xinwei Deng, Zachary R. Doerzaph
Summary: The study suggests a decision-adjusted framework using telematics data to develop optimal driver risk prediction models, which improves prediction precision compared to non-telematics predictors and has broad applications in fleet safety management and connected-vehicle safety technology.
ACCIDENT ANALYSIS AND PREVENTION
(2021)
Article
Public, Environmental & Occupational Health
Yikang Wang, Liying Zhang, Miaomiao Niu, Ruiying Li, Runqi Tu, Xiaotian Liu, Jian Hou, Zhenxing Mao, Zhenfei Wang, Chongjian Wang
Summary: This study aimed to establish an effective T2DM prediction tool using machine learning and genetic risk scores (GRS). The results showed significant improvements in reclassifications for all classifiers when adding GRS, demonstrating the potential clinical use of genetic markers in improving predictive performance for T2DM.
FRONTIERS IN PUBLIC HEALTH
(2021)
Article
Nutrition & Dietetics
Rachael Taylor, Megan E. Rollo, Jennifer N. Baldwin, Melinda Hutchesson, Elroy J. Aguiar, Katie Wynne, Ashley Young, Robin Callister, Clare E. Collins
Summary: This study evaluated the preliminary efficacy, feasibility, and acceptability of the 3-month Body Balance Beyond online program among Australian women with overweight/obesity and recent gestational diabetes mellitus. The results showed that video coaching sessions were associated with improvements in quality of life scores and self-efficacy. Further refinement of the website and text message support could improve the program's acceptability.
INTERNATIONAL JOURNAL OF BEHAVIORAL NUTRITION AND PHYSICAL ACTIVITY
(2022)
Article
Surgery
Jiangming Chen, Zixiang Chen, Xiyang Yan, Xiaoliang Liu, Debao Fang, Xiang Miao, Zhong Tong, Xiaoming Wang, Zheng Lu, Hui Hou, Cheng Wang, Xiaoping Geng, Fubao Liu
Summary: This study aimed to establish two online calculators for predicting anastomotic stricture occurrence and stricture-free survival in patients undergoing hepaticojejunostomy following bile duct injury. Independent risk factors were identified, and two nomogram models with corresponding online calculators were developed. The calculators demonstrated optimal predictive performance in internal and external validation.
INTERNATIONAL JOURNAL OF SURGERY
(2023)
Article
Multidisciplinary Sciences
Claudia Fernanda Giraldo-Jimenez, Javier Gaviria-Chavarro, Milton Sarria-Paja, Leonardo Antonio Bermeo Varon, John Jairo Villarejo-Mayor, Andre Luiz Felix Rodacki
Summary: Recent technological advances have brought both positive and negative impacts on people's lives, particularly in terms of how they interact and the development of new addictions. This study proposes a machine learning-based prediction model for smartphone dependency, which shows promising accuracy in identifying smartphone addiction.
SCIENTIFIC REPORTS
(2022)
Article
Medicine, Research & Experimental
Vishal Vyas, Hazel Blythe, Elizabeth G. Wood, Balraj Sandhar, Shah-Jalal Sarker, Damian Balmforth, Shirish G. Ambekar, John Yap, Stephen J. Edmondson, Carmelo Di Salvo, Kit Wong, Neil Roberts, Rakesh Uppal, Ben Adams, Alex Shipolini, Aung Y. Oo, David Lawrence, Shyam Kolvekar, Kulvinder S. Lall, Malcolm C. Finlay, M. Paula Longhi
Summary: Human EAT is enriched in adaptive immune cells, with significantly elevated expression of immune mediators in overweight/obese and diabetic patients compared to lean controls. The presence of obesity and diabetes, rather than underlying heart disease, alters the inflammatory profile of EAT, leading to significant alterations in metabolic and inflammatory pathways.
Article
Endocrinology & Metabolism
Gregor Stiglic, Fei Wang, Aziz Sheikh, Leona Cilar
Summary: This study developed and validated a 10-year T2DM risk prediction model using large survey data, showing the importance of recalibration and data pooling across multiple countries to increase precision.
PRIMARY CARE DIABETES
(2021)
Article
Environmental Sciences
Gregor Stiglic, Ruth Masterson Creber, Leona Cilar Budler
Summary: This study explores the relationship between internet use and psychosomatic symptoms among university students in Slovenia. It found that computer science students reported more frequent psychological symptoms, while health science students reported more somatic symptoms.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Review
Computer Science, Artificial Intelligence
Jeroen Ooge, Gregor Stiglic, Katrien Verbert
Summary: Healthcare is increasingly applying advanced AI algorithms to make predictions and explore large datasets. In order to make trustworthy decisions, healthcare professionals need ways to understand the outputs of these algorithms, one approach being visual analytics. Despite the development of many visual analytics systems for healthcare, there is a lack of clear overview of their explanation techniques. This study reviews 71 systems and analyzes how they explain advanced algorithms through visualization, interaction, and other methods.
WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY
(2022)
Article
Health Care Sciences & Services
Leona Cilar Budler, Majda Pajnkihar, Ulrike Ravens-Sieberer, Owen Barr, Gregor Stiglic
Summary: This study translated and validated the KIDSCREEN-27 scale among adolescents in Slovenia, finding that the scale has good psychometric properties and reliability, making it suitable for assessing the quality of life of adolescents.
HEALTH AND QUALITY OF LIFE OUTCOMES
(2022)
Article
Health Care Sciences & Services
Simon Kocbek, Primoz Kocbek, Lucija Gosak, Nino Fijacko, Gregor Stiglic
Summary: Type 2 diabetes mellitus is a significant health issue, and developing accurate and interpretable prediction models is crucial. This study proposes a framework that combines logic regression and least absolute shrinkage and selection operator to predict undiagnosed T2DM. The models developed using this framework perform well in terms of predictive accuracy and simplicity, making them suitable for healthcare experts to interpret.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Article
Health Care Sciences & Services
Lucija Gosak, Majda Pajnkihar, Gregor Stiglic
Summary: This study aims to assess the impact of mobile app use on self-care in patients with type 2 diabetes and test the practicality of the forDiabetes app. A double-blind randomized controlled trial will be conducted, collecting data including self-care questionnaire, illness perception questionnaire, as well as measurements of blood sugar, blood pressure, glycated hemoglobin, and weight. The research results will be published in 2023.
JMIR RESEARCH PROTOCOLS
(2022)
Review
Management
Lucija Gosak, Kristina Martinovic, Mateja Lorber, Gregor Stiglic
Summary: The aim of this review is to examine the effectiveness of artificial intelligence in predicting multimorbid diabetes-related complications. Artificial intelligence can accurately predict the risks of diabetes complications and provide an important tool for nurses in preventive healthcare. Using artificial intelligence contributes to better quality of care, better autonomy of patients in diabetes management, and reduction of complications, costs of medical care, and mortality.
JOURNAL OF NURSING MANAGEMENT
(2022)
Article
Environmental Sciences
Darja Korosec, Dominika Vrbnjak, Gregor Stiglic
Summary: This study examined the relationship between working hours and health outcomes in European countries. It found that men worked longer hours compared to women, and that different industries had varying levels of increase in health conditions. Healthcare workers had a slower increase in the prevalence of health conditions compared to workers in other industries, particularly regarding diabetes and hypertension. The study suggests the importance of implementing preventive health activities in the workplace to improve employee health.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Review
Computer Science, Artificial Intelligence
Leona Cilar Budler, Lucija Gosak, Gregor Stiglic
Summary: The use of conversational agents in health care is increasing globally, but their effectiveness is not well-understood. This advanced review aimed to assess the use and effectiveness of conversational agents in various health care fields. Most of the reviewed articles reported the effectiveness, while less information was available on the use. The study findings provide evidence-based knowledge about artificial intelligence-based question-answering systems in health care.
WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY
(2023)
Review
Chemistry, Multidisciplinary
Lucija Gosak, Adrijana Svensek, Mateja Lorber, Gregor Stiglic
Summary: Diabetic foot is a common complication of diabetes, causing a significant economic and societal problem due to the increased risk of lower limb amputation. Artificial intelligence can help predict the development of diabetic foot, leading to more effective preventive treatment. However, further research is needed to determine the most effective technique.
APPLIED SCIENCES-BASEL
(2023)
Article
Computer Science, Interdisciplinary Applications
Leon Kopitar, Peter Kokol, Gregor Stiglic
Summary: This paper proposes a hybrid visualization approach for detecting depression and builds a prediction model using NHANES data, achieving an average AUC score of 0.748. The model identifies crucial features such as chest pain, the ratio of family income to poverty, and smoking status in predicting depressive states in both the original and local environments.
JOURNAL OF BIOMEDICAL INFORMATICS
(2023)
Review
Health Care Sciences & Services
Natasa Mlinar Reljic, Maja Drescek Dolinar, Gregor Stiglic, Sergej Kmetec, Zvonka Fekonja, Barbara Donik
Summary: The aim of this study was to evaluate the advantages and limitations of e-learning for nursing and midwifery students during the COVID-19 pandemic. Through literature review and thematic analysis, e-learning was found to be an effective learning method for nursing and midwifery students during this period, but it also presented challenges related to internet access, technical equipment, finances, and work-family commitments.
Article
Health Care Sciences & Services
Leona Cilar Budler, Lucija Gosak, Dominika Vrbnjak, Majda Pajnkihar, Gregor Stiglic
Summary: Emotional intelligence is an important factor for the success and work performance of nursing students. However, research results on emotional intelligence in nursing students vary. A longitudinal study found that emotional intelligence of nursing students changes over time, with years of education and age as influencing factors. Further research is needed to determine the gendered nature of emotional intelligence in nursing students.
Review
Health Care Sciences & Services
Lucija Gosak, Gregor Stiglic, Leona Cilar Budler, Isa Brito Felix, Katja Braam, Nino Fijacko, Mara Pereira Guerreiro, Mateja Lorber
Summary: This study aims to review the outcomes of digital tools in behavior change support education and to examine existing instruments to assess their effectiveness. A systematic literature review was conducted, and 15 studies were included in the final analysis. The results showed that digital tools can enhance students' knowledge of behavior change techniques in individuals with chronic diseases, leading to increased self-confidence, improved cooperation, and practical experience and skills. However, time and space constraints are the most common limitations identified in using these tools.
Article
Health Care Sciences & Services
Lucija Gosak, Nino Fijacko, Carolina Chabrera, Esther Cabrera, Gregor Stiglic
Summary: This study validated the tool for assessing the educational environment in the Slovenian nursing student population using the DREEM tool. Despite the impact of the coronavirus pandemic on the educational process, students still perceive the educational environment as positive, especially in terms of learning outcomes, teacher evaluation, and academic self-perception.