ABSTRACT
Aiming at the problem that the small samples of critical disease in clinic may lead to prognostic models with poor performance of overfitting, large prediction error and instability, the long short-term memory transferring algorithm (transLSTM) was proposed. Based on the idea of transfer learning, the algorithm leverages the correlation between diseases to transfer information of different disease prognostic models, constructs the effictive model of target disease of small samples with the aid of large data of related diseases, hence improves the prediction performance and reduces the requirement for target training sample quantity. The transLSTM algorithm firstly uses the related disease samples to pretrain partial model parameters, and then further adjusts the whole network with the target training samples. The testing results on MIMIC-Ⅲ database showed that compared with traditional LSTM classification algorithm, the transLSTM algorithm had 0.02-0.07 higher AUROC and 0.05-0.14 larger AUPRC, while its number of training iterations was only 39%-64% of the traditional algorithm. The results of application on sepsis revealed that the transLSTM model of only 100 training samples had comparable mortality prediction performance to the traditional model of 250 training samples. In small sample situations, the transLSTM algorithm has significant advantages with higher prediciton accuracy and faster training speed. It realizes the application of transfer learning in the prognostic model of critical disease with small samples.
ABSTRACT
Due to the decline of motor ability and the impact of the diseases, abnormalities in gait is common in the elderly population, which will raise the risk of fall and cause serious injury. This study focuses on the analysis of the gait kinematics parameters of normal adults' gait, aiming to investigate the characteristics of gait parameters in different age groups and to explore the role of gait parameters in motor function assessment and clinical diagnosis. Based on the gait data gained by electronic walkway, the relationship among the toe out angles and their correlation with age and gender etc. were quantitatively analyzed. The results show that most normal subjects walk with positive toe out angles, and the angles increase with age. Such changes are slow in the young and middle age groups. However, the elevations of the left out toe angle and the angles between the feet are statistically significant after entering elder age ( >60 years). The results also suggest that the angle between the feet is a kind of practical gait parameter for varying applications. This study concludes that feet angle analysis is potential to provide a convenient and quantitative tool for the assessment of lower limb motor ability and the diagnosis of knee joint diseases.
ABSTRACT
To solve the problem that mostly gait analysis is independent from the treatment, this work proposes a system that integrates the functions of gait training and assessment for foot drop treatment. The system uses a set of sensors to collect gait parameters and designes multi-mode functional electrical stimulators as actuator. Body area network technology is introduced to coordinate the data communication and execution of the sensors and stimulators, synchronize the gait analysis and foot drop treatment. Bluetooth 4.0 is applied to low the power consumption of the system. The system realizes the synchronization of treatment and gait analysis. It is able to acquire and analyze the dynamic parameters of ankle, knee and hip in real-time, and treat patients by guiding functional electrical stimulation delivery to the specific body locations of patients.
Subject(s)
Humans , Electric Stimulation , Electric Stimulation Therapy , Exercise Therapy , Gait , Wireless TechnologyABSTRACT
Objective To investigate the association between body mass index (BMI) and exercise capacity in patients with chronic systolic heart failure.Methods The elderly patients with chronic systolic heart failure were consecutively recruited from 2008 to 2011 in cardiovascular clinic of Zhejiang Hospital.All the participants underwent height and weight measurements and BMI was calculated with these two parameters.Cardiopulmonary exercise test were performed to achieve peak oxygen uptake (PVO2),oxygen uptake to body mass ratio (PKVO2),oxygen uptake to heart ratio (VO2/HR) and ventilation/carbon dioxide production (VO2/VCO2).Results A total of the 273 patients with chronic systolic heart failure included 6 underweight patients (BMI ＜ 18.5 kg/m2),113 normal weight patients (BMI 18.5-＜ 24.0 kg/m2),116 overweight patients (BMI 24.0-＜ 28.0 kg/m2),and 38 obese patients (BMI ≥ 28 kg/m2).In both NYHA Ⅱ and Ⅲ/Ⅳ patients,unadjusted correlation analyses showed that BMI was positively related to PVO2 and VO2/HR,and was inversely related to PKVO2 and VE/VCO2 (P ＜ 0.05),respectively.Multiple stepwise regression analyses showed age,sex,BMI (P ＜ 0.05) and left ventricular ejection fraction (LVEF) were independent determinants of PKVO2,and age and BMI (P ＜ 0.05) were independent determinants of VE/VCO2.Conclusions BMI is significantly associated with exercise capacity in patients with chronic systolic heart failure,and also independent determinant for the PKVO2 and VE/VCO2,respectively.
ABSTRACT
In this paper, we propose the construction of a fifth-order Windkessel model, and give complete mathematical solutions for this model. Utilizing the diastolic pulse wave analytical methods, we derived the parameters of the mathematical model. The parameters were further applied to estimate arterial compliance, blood flow inertia, peripheral resistance and other indices. With simulation tools we assess the validity of the model, and built a simulation circuit with the model parameters R, C and L. The model parameters were obtained from the high-order Windkessel model. The stroke volume of left ventricle is employed as the input of the simulation circuit. At the end of the circuit, the responding signal was gained. And it in turn was compared with the measured pulse waveform. The results show that the fifth-order Windkessel model is superior to the third-order Windkessel model in the pulse wave fitting and stability, and thus better reflects the role of microvessles in the circulatory system.
Subject(s)
Humans , Algorithms , Blood Vessels , Physiology , Compliance , Computer Simulation , Microcirculation , Physiology , Models, Cardiovascular , Stroke Volume , PhysiologyABSTRACT
In this study ECG signal of unstrained rat was recorded by telemetry device, and heart rate variability (HRV) was analyzed in order to evaluate 24h autonomic nervous activity. The results demonstrated an obvious circadian rhythm in the autonomic nervous activity: sympathetic activity being dominant during wake phase, and parasympathetic activity, dominant during sleep phase. The ratio of the low frequency to high frequency components in HRV power spectrum (LF/HF) fluctuates with the change in the sleep stages. It is concluded that 24h HRV analyses may reveal plentiful information about the behavior of autonomic nervous system and thus facilitate the investigation of its regulating role in physiological and pathological processes.
Subject(s)
Animals , Male , Rats , Autonomic Nervous System , Physiology , Circadian Rhythm , Electrocardiography , Heart Rate , Physiology , Rats, Sprague-Dawley , TelemetryABSTRACT
Objective To assess the modulation of baroreflex activity by exogenous and endogenous nitric oxide (NO) in SD rats. Method The normal baroreflex sensitivity of SD rat was evaluated by intravenous (I.v.) administration of Phenylephrine (PE, 30 μg·kg-1) to elevate blood pressure. Exogenous and endogenous nitric oxide (NO) were respectively produced by I.v. Infusion of sodium nitroprusside (SNP, 10.2 μg·kg-1) and L-arginine (L-Arg, 25 mg·kg-1). And NG-nitro-L-arginine methylester (L-NAME, 10 mg·kg-1) was infused to inhibit the generation of NO. Meanwhile, baroreflex sensitivity fluctuations caused by PE were estimated. Result Compared with the normal state, L-Arg significantly reduced the baroreflex sensitivity (Gmax) and the slope coefficient τ of the blood pressure-heart rate curve [Gmax=-1.469±0.120 beats/(min·mmHg) vs. -0.664±0.080 beats/(min·mmHg), P<0.01; τ=0.0860.005 vs. 0.0490.007, P<0.05]. SNP also led to attenuation of Gmax and τ [Gmax=-1.277±0.392 beats/(min·mmHg) vs. -0.943±0.297 beats/(min·mmHg); τ=0.074豹0.027 vs. 0.071±0.023] but the changes were not significant. In contrast to L-Arg and SNP, L-NAME induced raise of Gmax and τ [Gmax=-0.705±0.231 beats/(min·mmHg) vs. -1.065±0.200 beats/(min·mmHg); τ=0.0450.011 vs. 0.0590.026], however, they were statistically insignificant. Conclusion The results demonstrated that NO imposes effect on baroreflex activity, and the behaviors of exogenous and endogenous NO are different. Endogenous NO may be involved in the regulation of autonomic nervous system via central nervous system or carotid sinus, and inhibits the baroreflex sensitivity.
ABSTRACT
Aim To examine the effect of Puerarin on calcium overload in human umbilical vascular endothelial cells(HUVECs).Methods The HUVECs were incubated with Puerarin,then were exposed to H_2O_2,ATP and high K~+.The changes of intracellular free calcium([Ca~(2+)]i) in HUVECs were determined using fluorescent Ca~(2+)-indicator fluo-3/AM with confocal laser scanning microscope.Results Puerarin inhibited the elevation of [Ca~(2+)]i induced by H_2O_2,ATP and high K~+,and the inhibition ratio were 72.2%,56.5% and 78.2% respectively.The effect of Puerarin inhibiting the elevation of [Ca~(2+)]i induced by high K~+ was similar to that of verapamil.Conclusion The mechanism of puerarin reducing the overload of calcium in HUVECs may be related with blocking both voltage-dependent calcium channels and the release of calcium pool in endoplasm.
ABSTRACT
Objective Normal estimation is the key step for volume visualization. Commonly used methods for normal estimation are based on interpolation and derivative. A novel normal estimation algorithm based on approximation for visualization of medical images was presented in this paper. Method It approximated the density function in local neighborhood with a second-degree polynomial function. The coefficients of the polynomial function were solved by minimizing the error of the approximation and the gradient vector at arbitrary point was obtained directly from the analytical derivative of the density function without interpolation. Because of symmetry, the solution of this equation was simplified.This method was tested in several volume data sets. The results and the generation time by different methods were obtained and compared. Result The results showed that this algorithm produced satisfactory quality images while the computational complexity was not increased. Conclusion This approach is preferable for most applications, especially for medical images reconstruction.