| 1. |
- Javeed, Ashir, 1989-, et al.
(författare)
-
Predictive Power of XGBoost_BiLSTM Model : A Machine-Learning Approach for Accurate Sleep Apnea Detection Using Electronic Health Data
- 2023
-
Ingår i: International Journal of Computational Intelligence Systems. - : Springer Nature. - 1875-6891 .- 1875-6883. ; 16:1
-
Tidskriftsartikel (refereegranskat)abstract
- Sleep apnea is a common disorder that can cause pauses in breathing and can last from a few seconds to several minutes, as well as shallow breathing or complete cessation of breathing. Obstructive sleep apnea is strongly associated with the risk of developing several heart diseases, including coronary heart disease, heart attack, heart failure, and stroke. In addition, obstructive sleep apnea increases the risk of developing irregular heartbeats (arrhythmias), which can lead to low blood pressure. To prevent these conditions, this study presents a novel machine-learning (ML) model for predicting sleep apnea based on electronic health data that provides accurate predictions and helps in identifying the risk factors that contribute to the development of sleep apnea. The dataset used in the study includes 75 features and 10,765 samples from the Swedish National Study on Aging and Care (SNAC). The proposed model is based on two modules: the XGBoost module assesses the most important features from feature space, while the Bidirectional Long Short-Term Memory Networks (BiLSTM) module classifies the probability of sleep apnea. Using a cross-validation scheme, the proposed XGBoost_BiLSTM algorithm achieves an accuracy of 97% while using only the six most significant features from the dataset. The model’s performance is also compared with conventional long-short-term memory networks (LSTM) and other state-of-the-art ML models. The results of the study suggest that the proposed model improved the diagnosis and treatment of sleep apnea by identifying the risk factors.
|
|
| 2. |
- Javeed, Ashir, 1989-, et al.
(författare)
-
Predicting Dementia Risk Factors Based on Feature Selection and Neural Networks
- 2023
-
Ingår i: Computers, Materials and Continua. - : Tech Science Press. - 1546-2218 .- 1546-2226. ; 75:2, s. 2491-2508
-
Tidskriftsartikel (refereegranskat)abstract
- Dementia is a disorder with high societal impact and severe consequences for its patients who suffer from a progressive cognitive decline that leads to increased morbidity, mortality, and disabilities. Since there is a consensus that dementia is a multifactorial disorder, which portrays changes in the brain of the affected individual as early as 15 years before its onset, prediction models that aim at its early detection and risk identification should consider these characteristics. This study aims at presenting a novel method for ten years prediction of dementia using on multifactorial data, which comprised 75 variables. There are two automated diagnostic systems developed that use genetic algorithms for feature selection, while artificial neural network and deep neural network are used for dementia classification. The proposed model based on genetic algorithm and deep neural network had achieved the best accuracy of 93.36%, sensitivity of 93.15%, specificity of 91.59%, MCC of 0.4788, and performed superior to other 11 machine learning techniques which were presented in the past for dementia prediction. The identified best predictors were: age, past smoking habit, history of infarct, depression, hip fracture, single leg standing test with right leg, score in the physical component summary and history of TIA/RIND. The identification of risk factors is imperative in the dementia research as an effort to prevent or delay its onset.
|
|
| 3. |
- Behrens, Anders, et al.
(författare)
-
Sleep disturbance predicts worse cognitive performance in subsequent years : A longitudinal population-based cohort study
- 2023
-
Ingår i: Archives of gerontology and geriatrics (Print). - : Elsevier. - 0167-4943 .- 1872-6976. ; 106
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Poor sleep is a potential modifiable risk factor for later life development cognitive impairment. The aim of this study is to examine if subjective measures of sleep duration and sleep disturbance predict future cognitive decline in a population-based cohort of 60, 66, 72 and 78-year-olds with a maximal follow up time of 18 years. Methods: This study included participants from the Swedish National Study on Ageing and Care – Blekinge, with assessments 2001–2021. A cohort of 60 (n = 478), 66 (n = 623), 72 (n = 662) and 78 (n = 548) year-olds, were assessed at baseline and every 6 years until 78 years of age. Longitudinal associations between sleep disturbance (sleep scale), self-reported sleep duration and cognitive tests (Mini Mental State Examination and the Clock drawing test) were examined together with typical confounders (sex, education level, hypertension, hyperlipidemia, smoking status, physical inactivity and depression). Results: There was an association between sleep disturbance at age 60 and worse cognitive function at ages 60, 66 and 72 years in fully adjusted models. The association was attenuated after bootstrap-analysis for the 72-year-olds. The items of the sleep scale most predictive of later life cognition regarded nightly awakenings, pain and itching and daytime naps. Long sleep was predictive of future worse cognitive function. Conclusion: Sleep disturbance was associated with worse future cognitive performance for the 60-year-olds, which suggests poor sleep being a risk factor for later life cognitive decline. Questions regarding long sleep, waking during the night, pain and itching and daytime naps should be further explored in future research and may be targets for intervention.
|
|
| 4. |
- Quintana, María, et al.
(författare)
-
Feasibility-usability study of a tablet app adapted specifically for persons with cognitive impairment—SMART4MD (Support monitoring and reminder technology for mild dementia)
- 2020
-
Ingår i: International Journal of Environmental Research and Public Health. - : MDPI. - 1661-7827 .- 1660-4601. ; 17:18, s. 1-21
-
Tidskriftsartikel (refereegranskat)abstract
- Population ageing within Europe has major social and economic consequences. One of the most devastating conditions that predominantly affects older people is dementia. The SMART4MD (Support Monitoring and Reminder Technology for Mild Dementia) project aims to develop and test a health application specifically designed for people with mild dementia. The aim of this feasibility study was to evaluate the design of the SMART4MD protocol, including recruitment, screening, baseline examination and data management, and to test the SMART4MD application for functionality and usability before utilization in a full-scale study. The feasibility study tested the protocol and the app in Spain and Sweden. A total of nineteen persons with cognitive impairment, and their informal caregivers, individually performed a task-based usability test of the SMART4MD app model in a clinical environment, followed by four-week testing of the app in the home environment. By employing a user-centered design approach, the SMART4MD application proved to be an adequate and feasible interface for an eHealth intervention. In the final usability test, a score of 81% satisfied users was obtained. The possibility to test the application in all the procedures included in the study generated important information on how to present the technology to the users and how to improve these procedures. © 2020 by the authors. Licensee MDPI, Basel, Switzerland.
|
|
| 5. |
- Alsolai, Hadeel, et al.
(författare)
-
Employing a Long-Short-Term Memory Neural Network to Improve Automatic Sleep Stage Classification of Pharmaco-EEG Profiles
- 2022
-
Ingår i: Applied Sciences. - : MDPI. - 2076-3417. ; 12:10
-
Tidskriftsartikel (refereegranskat)abstract
- An increasing problem in today's society is the spiraling number of people suffering from various sleep disorders. The research results presented in this paper support the use of a novel method that employs techniques from the classification of sleep disorders for more accurate scoring. Applying this novel method will assist researchers with better analyzing subject profiles for recommending prescriptions or to alleviate sleep disorders. In biomedical research, the use of animal models is required to experimentally test the safety and efficacy of a drug in the pre-clinical stage. We have developed a novel LSTM Recurrent Neural Network to process Pharmaco-EEG Profiles of rats to automatically score their sleep-wake stages. The results indicate improvements over the current methods; for the case of combined channels, the model accuracy improved by 1% and 3% in binary or multiclass classifications, respectively, to accuracies of 93% and 82%. In the case of using a single channel, binary and multiclass LSTM models for identifying rodent sleep stages using single or multiple electrode positions for binary or multiclass problems have not been evaluated in prior literature. The results reveal that single or combined channels, and binary or multiclass classification tasks, can be applied in the automatic sleep scoring of rodents.
|
|
| 6. |
- Guzman-Parra, Jose, et al.
(författare)
-
Attitudes and use of information and communication technologies in older adults with mild cognitive impairment or early stages of dementia and their caregivers : cross-sectional study
- 2020
-
Ingår i: Journal of Medical Internet Research. - : JMIR Publications. - 1438-8871. ; 22:6
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Information and communication technologies are promising tools to increase the quality of life of people with dementia or mild cognitive impairment and that of their caregivers. However, there are barriers to their use associated with sociodemographic factors and negative attitudes, as well as inadequate knowledge about technologies. OBJECTIVE: The aim of this study was to analyze technophilia (attitudes toward new technologies) and the use of smartphones and tablets along with associated factors in people with dementia/mild cognitive impairment and their caregivers. METHODS: Data from the first visit of the Support Monitoring and Reminder for Mild Dementia (SMART4MD) randomized multicenter clinical trial were used for this analysis. Data were obtained from two European countries, Spain and Sweden, and from three centers: Consorci Sanitari de Terrassa (Catalonia, Spain), Servicio Andaluz de Salud (Andalusia, Spain), and the Blekinge Institute of Technology (Sweden). Participants with a score between 20 and 28 in the Mini Mental State Examination, with memory problems (for more than 6 months), and who were over the age of 55 years were included in the study, along with their caregivers. The bivariate Chi square and Mann-Whitney tests, and multivariate linear and logistic regression models were used for statistical analysis. RESULTS: A total of 1086 dyads were included (N=2172). Overall, 299 (27.53%) of people with dementia/mild cognitive impairment had a diagnosis of dementia. In addition, 588 (54.14%) of people with dementia/mild cognitive impairment reported using a smartphone almost every day, and 106 (9.76%) used specific apps or software to support their memory. Among the caregivers, 839 (77.26%) used smartphones and tablets almost every day, and 181 (16.67%) used specific apps or software to support their memory. The people with dementia/mild cognitive impairment showed a lower level of technophilia in comparison to that of their caregivers after adjusting for confounders (B=0.074, P=.02) with differences in technology enthusiasm (B=0.360, P<.001), but not in technology anxiety (B=-0.042, P=.37). Technophilia was associated with lower age (B=-0.009, P=.004), male gender (B=-0.160, P<.001), higher education level (P=.01), living arrangement (living with children vs single; B=-2.538, P=.01), country of residence (Sweden vs Spain; B=0.256, P<.001), lower depression (B=-0.046, P<.001), and better health status (B=0.004, P<.001) in people with dementia/mild cognitive impairment. Among caregivers, technophilia was associated with comparable sociodemographic factors (except for living arrangement), along with a lower caregiver burden (B=-0.005, P=.04) and better quality of life (B=0.348, P<.001). CONCLUSIONS: Technophilia was associated with a better quality of life and sociodemographic variables in people with dementia/mild cognitive impairment and caregivers, suggesting potential barriers for technological interventions. People with dementia/mild cognitive impairment frequently use smartphones and tablets, but the use of specific apps or software to support memory is limited. Interventions using these technologies are needed to overcome barriers in this population related to sociodemographic characteristics and the lack of enthusiasm for new technologies. TRIAL REGISTRATION: ClinicalTrials.gov NCT03325699; https://clinicaltrials.gov/ct2/show/NCT03325699. ©Jose Guzman-Parra, Pilar Barnestein-Fonseca, Gloria Guerrero-Pertiñez, Peter Anderberg, Luis Jimenez-Fernandez, Esperanza Valero-Moreno, Jessica Marian Goodman-Casanova, Antonio Cuesta-Vargas, Maite Garolera, Maria Quintana, Rebeca I García-Betances, Evi Lemmens, Johan Sanmartin Berglund, Fermin Mayoral-Cleries.
|
|
| 7. |
- Zhang, Jialin, et al.
(författare)
-
Males are more sensitive to reward and less sensitive to loss than females among people with internet gaming disorder : FMRI evidence from a card-guessing task
- 2020
-
Ingår i: BMC Psychiatry. - : BioMed Central. - 1471-244X. ; 20:1
-
Tidskriftsartikel (refereegranskat)abstract
- Background: Many studies have found an interesting issue in the Internet gaming disorder (IGD): males are always observed to be the majority. However, there are little research to exploring the differences in the neural mechanisms between males and females in decision-making process among people with IGD. Therefore, explore the reward/loss processing between different gender with IGD could help in understanding the underlying neural mechanism of IGD. Methods: Data from functional magnetic resonance imaging (fMRI) were collected from 111 subjects (IGD: 29 males, 25 females; recreational internet game user (RGU): 36 males, 21 females) while they were performing a card-guessing task. We collected and compared their brain features when facing the win and loss conditions in different groups. Results: For winning conditions, IGD group showed hypoactivity in the lingual gyrus than RGU group, male players showed hyperactivity in the left caudate nucleus, bilateral cingulate gyrus, right middle frontal gyrus (MFG), right precuneus and inferior parietal lobule relative to the females. And significant sex-by-group interactions results showed higher brain activities in the thalamus, parahippocampal gyrus and lower brain activities in Inferior frontal gyrus (IFG) were observed in males with IGD than females. For losing conditions, IGD group showed hypoactivity in the left lingual gyrus, parahippocampal gyrus and right anterior cingulate cortex (ACC) compared to the RGU group, male players showed hyperactive left caudate nucleus and hypoactive right middle occipital gyrus relative to females. And significant sex-by-group interactions results showed that compared to females with IGD, males with IGD showed decreased brain activities in the IFG and lingual gyrus. Conclusions: First, there appeared to be no difference in reward processing between the IGD and RGU group, but IGD showed less sensitivity to loss. Secondly, male players showed more sensitivity to rewards and less sensitivity to losses. Last but not least, males and females showed opposite activation patterns in IGD degree and rewards/losses processing. And male IGD subjects are more sensitive to reward and less sensitive to loss than females, which might be the reason for the gender different rates on IGD. © 2020 The Author(s).
|
|
