Background:Stroke is the second leading cause of death and third leading cause of disability worldwide and is the leading cause of death and disability among adults in China,with its incidence rate continuing to rise....Background:Stroke is the second leading cause of death and third leading cause of disability worldwide and is the leading cause of death and disability among adults in China,with its incidence rate continuing to rise.In China,the average age of firsttime stroke patients is 66.4 years,and the intravenous thrombolysis rate using recombinant tissue plasminogen activator within 3 h of onset is only 16%.Given this fact,there is a pressing need for real‐time predictive tools,particularly for elderly individuals at home,that can provide early warnings for potential strokes.Methods:We collected continuous monitoring data from nonintrusive smart beds and multimodal temporal data from electronic medical records at the National Center for Neurological Disorders.The data included smart bed monitoring indicators,laboratory tests,nurse observations,and static data as potential predictors,with stroke as the outcome.We applied feature representation and feature selection techniques and then input the predictors into machine learning models.Additionally,deep learning models were used after preprocessing the irregular temporal data.Finally,we evaluated the performance of the stroke prediction models and assessed the importance of the features.We used continuously updated vital signs and clinical data during hospitalization to generate timely stroke risk alerts during the same period of admission.Results:A total of 37,041 samples were analyzed,of which 7020 patients were diagnosed with stroke.When only the smart bed features were used for prediction,the model achieved an area under the receiver operating characteristic curve(AUROC)of 0.59−0.63,with an accuracy ranging from 60%−65%.Among the four artificial intelligence algorithms,the random forest model demonstrated the best performance.After all the available features were incorporated,the AUROC increased to 0.94,and the accuracy improved to 92%.Conclusions:In this study,the occurrence of stroke was successfully identified by integrating multimodal temporal data from electronic medical records.Noncontact monitoring of respiration and heart rate offers a promising approach for daily stroke surveillance in home‐based populations,particularly for elderly individuals living alone.展开更多
The links of many medical-biological events with high levels of geomagnetic activity (GMA) are widely discussed. In recent years, several medical phenomena were described in inverse distribution by time with GMA. Also...The links of many medical-biological events with high levels of geomagnetic activity (GMA) are widely discussed. In recent years, several medical phenomena were described in inverse distribution by time with GMA. Also a concurrent to GMA and solar activity force-cosmic ray activity (CRA) and closely related high energy neutron and proton fluxes are studied as a forces dominating at low GMA and solar activity in relation to considered medical events. The aim of this study was to explore the distribution of some important medical events on days with “Zero” GMA levels, accompanied by high CRA (neutron activity). Medical event data of the Grand Baku region (more than 3 mln inhabitants), Azerbaijan, with daily distribution on the time 1 Dec. 2002-31 Dec. 2007 was compared to daily GMA Kp indices in general (Kp > 0, 1837 days) and 34 days daily GMA indices Kp = 0. Daily CRA data was also compared using neutron monitoring data from two stations. Daily averaged data and their standard deviations on the mentioned GMA levels were compared and statistical significance was established. Results revealed a significant rise in the number of emergencies (n = 1,567,576) and total deaths number (n = 46,360) at the days of “Zero” GMA level. These days were accompanied by significant rise of CRA (neutron activity). For Sudden Cardiac Deaths (SCD, n = 1615) and cerebral stroke (CVA, n =10,054) the increase achieved strong trend to significance level. Acute Myocardial Infarction occurrence (morbidity) and trauma were also absolutely more registered at days with “Zero” GMA level, despite the small number of such days. The average Infection numbers show an inverse relationship with absolutely high registry at the “Zero” GMA level days. Study linking environmental physical activity levels and the human medical data shows that geomagnetic field variations accompanied by the increased level of cosmic ray activity, can have either direct or indirect adverse effects on human health and physiology, even when the magnitude of the geomagnetic field disturbance is extremely small or even is equal to zero. On days of “Zero” daily Kp indices describing Geomagnetic Activity, accompanied by high Cosmic Ray Activity (neutron activity), more medical emergencies and total death number (daily) occurred. Sudden Cardiac Deaths and Cerebral Stroke numbers show a strong trend to significant rise. Absolute increase of number of Acute Myocardial Infarction and less Infections, not achieving statistical significance, was also observed. These results are additional data for considering Cosmic Ray Activity (neutron activity) as an additional factor involved in time distribution of human medical events.展开更多
In year 2001 a paper in the ANAS considered links between month of birth and longevity. In the following years we published four papers related to “big killers” (cardiac and oncology) that showed some differences in...In year 2001 a paper in the ANAS considered links between month of birth and longevity. In the following years we published four papers related to “big killers” (cardiac and oncology) that showed some differences in birth months distribution of this group and studied by LA, NS Gavrilov’s centenarians. The aim of this study was to study conception and birth months of another modern “big killer”—cerebral stroke (CVA) that is taking a leading role among cardiovascular causes of death in the last decades. Methods: 130,120 deaths of both gender CVA victims in Lithuania at 1989-2013 were studied. In addition to birth month, the months of conception (9 months before birth) were studied. Our data were compared with results of centenarians (birth of LA, NS Gavrilov’s study and transformed by authors also their conception month). Results: The maximum of births were January, March and May for CVA victims, while the analogical conception maximum were in April, June, May and July. The similar data for centenarians were that maximal births were in November, September, October (LA, NS Gavrilov) and conception in December, January, February. These results are similar to data related to cardiac and oncology deaths published in our previous publications. Conclusion: The conception and birth month of victims of CVA is different of similar data obtained by centenarians study. Different environmental conditions at different parts of the year and solar cycle can play a role affecting the embryo at early stages of development, predisposing to some pathologies in coming years of life.展开更多
Background:Studies have shown that heart rate variability(HRV)is a predictor of the prognosis of cardiovascular diseases.Contact heartbeat monitoring equipment is widely used,especially in hospitals,and benefits from ...Background:Studies have shown that heart rate variability(HRV)is a predictor of the prognosis of cardiovascular diseases.Contact heartbeat monitoring equipment is widely used,especially in hospitals,and benefits from the rapidity and accuracy of the detection of physiological health indicators.However,long-term contact with equipment has many adverse effects.The purpose of this study was to improve the accuracy of HRV detection via noncontact equipment,thus enabling HRV to be assessed in various scenarios.Methods:A novel deep learning approach was proposed for measuring heartbeats through camera videos.First,we performed facial segmentation and divided the face into 16 grid cells with different light balance scores.After the trend is filtered by the Hamming window,a transformer-based neural network is used to further filter the signal.Finally,heart rate(HR)and HRV are estimated.Results:We used 1 million synthetic data points for pretraining and a public dataset in combination with a dataset that we constructed for task training.The final results were obtained on a test dataset that we constructed.The accuracy for HR with a low light balance score(0.867-0.983)was greater than that with a high score(0.667-0.750).Our method had higher accuracy in estimating HR than traditional filtering methods(0.167-0.417)and state-of-the-art neural network filtering methods(0.783-0.917)did.The root mean square error of the HRV from the time domain was the lowest,and the correlation index score was the highest for the HRV from the frequency domain estimated by our method compared with those estimated by two neural networks.Conclusions:Light balance,large sample training,and two-stage training can improve the accuracy of HRV estimation.展开更多
Objective:This study addresses the limitations of existing traditional Chinese medicine(TCM)constitution identification techniques for the elderly by proposing an intelligent identification method aimed at enhancing t...Objective:This study addresses the limitations of existing traditional Chinese medicine(TCM)constitution identification techniques for the elderly by proposing an intelligent identification method aimed at enhancing the accuracy,standardization,and formalization of the identification process.Materials and Methods:Leveraging data from the images of the tongue,face,and pulse,this study introduced four image classification models:EfficientNetV2,MobileViT,Vision Transformer,and Swin Transformer.A comparative experimental approach was employed to establish a baseline model.Subsequently,a multi-information fusion model was constructed on this foundation,extracting integrated features from diverse data to further improve identification accuracy.Results:The multi-information fusion model developed in this study achieved an accuracy of 71.32%,effectively enhancing the accuracy of TCM constitution identification for the elderly.Conclusions:The multi-information fusion model developed in this study,by integrating tongue,facial,and pulse data,considerably enhances the accuracy of TCM constitution identification.It effectively addresses the certain limitations inherent in existing TCM constitution identification techniques,offering a novel and efficacious strategy for this domain.展开更多
With the application of integrated sensing and communication,radiated source localization has gradually become a popular research direction.Radiation source localization has more applications in reality,for example,in...With the application of integrated sensing and communication,radiated source localization has gradually become a popular research direction.Radiation source localization has more applications in reality,for example,in earthquake disaster scenarios,entrapped individuals can be found by using terminal devices.The traditional methods suffer from degradation of performance under low signal-to-noise ratio(SNR)conditions and cannot effectively deal with complex propagation environments.A signal direction of arrival(DOA)localization method based on convolutional neural networks is proposed to achieve high resolution localization of single or multiple radio frequency(RF)radiation sources in scenarios with low SNR and adjacent sources.The experiment shows that the proposed method has good performance in single target and multi-target localization.In addition,the proposed method still has good estimation performance in environments with small signal source angle intervals and varying SNR.展开更多
Emergency communication in forested environments encounters extremely complex propagation conditions,whose channel characteristics cannot be accurately captured by conventional models designed for urban scenarios.This...Emergency communication in forested environments encounters extremely complex propagation conditions,whose channel characteristics cannot be accurately captured by conventional models designed for urban scenarios.This paper conducts field measurements and modeling for 380 MHz vehicle-to-infrastructure(V2I)links in the primeval forest of the Greater Khingan Mountains.Through joint analysis of large-scale and small-scale channel characteristics,we systematically reveal the signal attenuation,fading mechanisms,and delay spread behavior in such environments.Results indicate that for path loss modeling,the Alpha-Beta model outperforms the Close-In model in characterizing the rapid attenuation in dense foliage.Small-scale fading is predominantly Rayleigh-distributed,while Ricean fading only dominates in the initial stage,with its K-factor decreasing significantly as distance increases.The root mean square(RMS)delay spread ranges from 0 to 450 ns,follows a log-normal distribution,and is markedly larger than in urban scenarios.Based on these observations,a parametric tapped-delay-line(TDL)model is developed,effectively quantifying multipath delay,power decay,and fading characteristics.The proposed model can reproduce realistic propagation features in simulations,offering theoretical support for emergency communication system planning and optimization,as well as engineering guidance for enhancing communication reliability and coverage in disaster rescue scenarios.展开更多
基金supported by the National Natural Science Foundation of China(72204169,82425101,82271516,81801187)Noncommunicable Chronic Diseases‐National Science and Technology Major Project(2023ZD0504800,2023ZD0504801,2023ZD0504802,2023ZD0504803,2023ZD0504804)+2 种基金Beijing Municipal Science&Technology Commission(Z231100004823036)Capital's Funds for Health Improvement and Research(2022‐2‐2045)National Key R&D Program of China(2024YFC3044800,2022YFF1501500,2022YFF1501501,2022YFF1501502,2022YFF1501503,2022YFF1501504,2022YFF1501505).
文摘Background:Stroke is the second leading cause of death and third leading cause of disability worldwide and is the leading cause of death and disability among adults in China,with its incidence rate continuing to rise.In China,the average age of firsttime stroke patients is 66.4 years,and the intravenous thrombolysis rate using recombinant tissue plasminogen activator within 3 h of onset is only 16%.Given this fact,there is a pressing need for real‐time predictive tools,particularly for elderly individuals at home,that can provide early warnings for potential strokes.Methods:We collected continuous monitoring data from nonintrusive smart beds and multimodal temporal data from electronic medical records at the National Center for Neurological Disorders.The data included smart bed monitoring indicators,laboratory tests,nurse observations,and static data as potential predictors,with stroke as the outcome.We applied feature representation and feature selection techniques and then input the predictors into machine learning models.Additionally,deep learning models were used after preprocessing the irregular temporal data.Finally,we evaluated the performance of the stroke prediction models and assessed the importance of the features.We used continuously updated vital signs and clinical data during hospitalization to generate timely stroke risk alerts during the same period of admission.Results:A total of 37,041 samples were analyzed,of which 7020 patients were diagnosed with stroke.When only the smart bed features were used for prediction,the model achieved an area under the receiver operating characteristic curve(AUROC)of 0.59−0.63,with an accuracy ranging from 60%−65%.Among the four artificial intelligence algorithms,the random forest model demonstrated the best performance.After all the available features were incorporated,the AUROC increased to 0.94,and the accuracy improved to 92%.Conclusions:In this study,the occurrence of stroke was successfully identified by integrating multimodal temporal data from electronic medical records.Noncontact monitoring of respiration and heart rate offers a promising approach for daily stroke surveillance in home‐based populations,particularly for elderly individuals living alone.
文摘The links of many medical-biological events with high levels of geomagnetic activity (GMA) are widely discussed. In recent years, several medical phenomena were described in inverse distribution by time with GMA. Also a concurrent to GMA and solar activity force-cosmic ray activity (CRA) and closely related high energy neutron and proton fluxes are studied as a forces dominating at low GMA and solar activity in relation to considered medical events. The aim of this study was to explore the distribution of some important medical events on days with “Zero” GMA levels, accompanied by high CRA (neutron activity). Medical event data of the Grand Baku region (more than 3 mln inhabitants), Azerbaijan, with daily distribution on the time 1 Dec. 2002-31 Dec. 2007 was compared to daily GMA Kp indices in general (Kp > 0, 1837 days) and 34 days daily GMA indices Kp = 0. Daily CRA data was also compared using neutron monitoring data from two stations. Daily averaged data and their standard deviations on the mentioned GMA levels were compared and statistical significance was established. Results revealed a significant rise in the number of emergencies (n = 1,567,576) and total deaths number (n = 46,360) at the days of “Zero” GMA level. These days were accompanied by significant rise of CRA (neutron activity). For Sudden Cardiac Deaths (SCD, n = 1615) and cerebral stroke (CVA, n =10,054) the increase achieved strong trend to significance level. Acute Myocardial Infarction occurrence (morbidity) and trauma were also absolutely more registered at days with “Zero” GMA level, despite the small number of such days. The average Infection numbers show an inverse relationship with absolutely high registry at the “Zero” GMA level days. Study linking environmental physical activity levels and the human medical data shows that geomagnetic field variations accompanied by the increased level of cosmic ray activity, can have either direct or indirect adverse effects on human health and physiology, even when the magnitude of the geomagnetic field disturbance is extremely small or even is equal to zero. On days of “Zero” daily Kp indices describing Geomagnetic Activity, accompanied by high Cosmic Ray Activity (neutron activity), more medical emergencies and total death number (daily) occurred. Sudden Cardiac Deaths and Cerebral Stroke numbers show a strong trend to significant rise. Absolute increase of number of Acute Myocardial Infarction and less Infections, not achieving statistical significance, was also observed. These results are additional data for considering Cosmic Ray Activity (neutron activity) as an additional factor involved in time distribution of human medical events.
文摘In year 2001 a paper in the ANAS considered links between month of birth and longevity. In the following years we published four papers related to “big killers” (cardiac and oncology) that showed some differences in birth months distribution of this group and studied by LA, NS Gavrilov’s centenarians. The aim of this study was to study conception and birth months of another modern “big killer”—cerebral stroke (CVA) that is taking a leading role among cardiovascular causes of death in the last decades. Methods: 130,120 deaths of both gender CVA victims in Lithuania at 1989-2013 were studied. In addition to birth month, the months of conception (9 months before birth) were studied. Our data were compared with results of centenarians (birth of LA, NS Gavrilov’s study and transformed by authors also their conception month). Results: The maximum of births were January, March and May for CVA victims, while the analogical conception maximum were in April, June, May and July. The similar data for centenarians were that maximal births were in November, September, October (LA, NS Gavrilov) and conception in December, January, February. These results are similar to data related to cardiac and oncology deaths published in our previous publications. Conclusion: The conception and birth month of victims of CVA is different of similar data obtained by centenarians study. Different environmental conditions at different parts of the year and solar cycle can play a role affecting the embryo at early stages of development, predisposing to some pathologies in coming years of life.
基金National Natural Science Foundation of China,Grant/Award Number:72204169Department of Science and Technology of Sichuan Province,Grant/Award Number:2021YFS0393。
文摘Background:Studies have shown that heart rate variability(HRV)is a predictor of the prognosis of cardiovascular diseases.Contact heartbeat monitoring equipment is widely used,especially in hospitals,and benefits from the rapidity and accuracy of the detection of physiological health indicators.However,long-term contact with equipment has many adverse effects.The purpose of this study was to improve the accuracy of HRV detection via noncontact equipment,thus enabling HRV to be assessed in various scenarios.Methods:A novel deep learning approach was proposed for measuring heartbeats through camera videos.First,we performed facial segmentation and divided the face into 16 grid cells with different light balance scores.After the trend is filtered by the Hamming window,a transformer-based neural network is used to further filter the signal.Finally,heart rate(HR)and HRV are estimated.Results:We used 1 million synthetic data points for pretraining and a public dataset in combination with a dataset that we constructed for task training.The final results were obtained on a test dataset that we constructed.The accuracy for HR with a low light balance score(0.867-0.983)was greater than that with a high score(0.667-0.750).Our method had higher accuracy in estimating HR than traditional filtering methods(0.167-0.417)and state-of-the-art neural network filtering methods(0.783-0.917)did.The root mean square error of the HRV from the time domain was the lowest,and the correlation index score was the highest for the HRV from the frequency domain estimated by our method compared with those estimated by two neural networks.Conclusions:Light balance,large sample training,and two-stage training can improve the accuracy of HRV estimation.
文摘Objective:This study addresses the limitations of existing traditional Chinese medicine(TCM)constitution identification techniques for the elderly by proposing an intelligent identification method aimed at enhancing the accuracy,standardization,and formalization of the identification process.Materials and Methods:Leveraging data from the images of the tongue,face,and pulse,this study introduced four image classification models:EfficientNetV2,MobileViT,Vision Transformer,and Swin Transformer.A comparative experimental approach was employed to establish a baseline model.Subsequently,a multi-information fusion model was constructed on this foundation,extracting integrated features from diverse data to further improve identification accuracy.Results:The multi-information fusion model developed in this study achieved an accuracy of 71.32%,effectively enhancing the accuracy of TCM constitution identification for the elderly.Conclusions:The multi-information fusion model developed in this study,by integrating tongue,facial,and pulse data,considerably enhances the accuracy of TCM constitution identification.It effectively addresses the certain limitations inherent in existing TCM constitution identification techniques,offering a novel and efficacious strategy for this domain.
基金funded by the National Key Research and Development Program of China under Grants 2023YFC3011500 and 2024YFC3016000.
文摘With the application of integrated sensing and communication,radiated source localization has gradually become a popular research direction.Radiation source localization has more applications in reality,for example,in earthquake disaster scenarios,entrapped individuals can be found by using terminal devices.The traditional methods suffer from degradation of performance under low signal-to-noise ratio(SNR)conditions and cannot effectively deal with complex propagation environments.A signal direction of arrival(DOA)localization method based on convolutional neural networks is proposed to achieve high resolution localization of single or multiple radio frequency(RF)radiation sources in scenarios with low SNR and adjacent sources.The experiment shows that the proposed method has good performance in single target and multi-target localization.In addition,the proposed method still has good estimation performance in environments with small signal source angle intervals and varying SNR.
基金funded by the National Key Research and Development Program of China under Grants 2023YFC3011500 and 2024YFC3016000.
文摘Emergency communication in forested environments encounters extremely complex propagation conditions,whose channel characteristics cannot be accurately captured by conventional models designed for urban scenarios.This paper conducts field measurements and modeling for 380 MHz vehicle-to-infrastructure(V2I)links in the primeval forest of the Greater Khingan Mountains.Through joint analysis of large-scale and small-scale channel characteristics,we systematically reveal the signal attenuation,fading mechanisms,and delay spread behavior in such environments.Results indicate that for path loss modeling,the Alpha-Beta model outperforms the Close-In model in characterizing the rapid attenuation in dense foliage.Small-scale fading is predominantly Rayleigh-distributed,while Ricean fading only dominates in the initial stage,with its K-factor decreasing significantly as distance increases.The root mean square(RMS)delay spread ranges from 0 to 450 ns,follows a log-normal distribution,and is markedly larger than in urban scenarios.Based on these observations,a parametric tapped-delay-line(TDL)model is developed,effectively quantifying multipath delay,power decay,and fading characteristics.The proposed model can reproduce realistic propagation features in simulations,offering theoretical support for emergency communication system planning and optimization,as well as engineering guidance for enhancing communication reliability and coverage in disaster rescue scenarios.
