The risk prediction of paroxysmal atrial fibrillation (PAF) is a challenge in the field of biomedical engineering. This study integrated the advantages of machine learning feature engineering and end-to-end modeling of deep learning to propose a PAF risk prediction method based on multimodal feature fusion. Additionally, the study utilized four different feature selection methods and Pearson correlation analysis to determine the optimal multimodal feature set, and employed random forest for PAF risk assessment. The proposed method achieved accuracy of (92.3 ± 2.1)% and F1 score of (91.6 ± 2.9)% in a public dataset. In a clinical dataset, it achieved accuracy of (91.4 ± 2.0)% and F1 score of (90.8 ± 2.4)%. The method demonstrates generalization across multi-center datasets and holds promising clinical application prospects.
Humans ; Atrial Fibrillation/diagnosis* ; Machine Learning ; Deep Learning ; Risk Assessment/methods*

To treat the problem of identification performance and the complexity of the algorithm, we proposed a piecewise linear representation and dynamic time warping (PLR-DTW) method for ECG biometric identification. Firstly we detected R peaks to get the heartbeats after denoising preprocessing. Then we used the PLR method to keep important information of an ECG signal segment while reducing the data dimension at the same time. The improved DTW method was used for similarity measurements between the test data and the templates. The performance evaluation was carried out on the two ECG databases: PTB and MIT-BIH. The analystic results showed that compared to the discrete wavelet transform method, the proposed PLR-DTW method achieved a higher accuracy rate which is nearly 8% of rising, and saved about 30% operation time, and this demonstrated that the proposed method could provide a better performance.
Algorithms ; Biometric Identification ; methods ; Electrocardiography ; methods ; Humans ; Patient Identification Systems ; Pattern Recognition, Physiological

Biotic tissues are a kind of highly scattering random media; studies on laser light propagation in biotic tissues play an important role in bio-medical diagnostics and therapeutics. The propagation and distribution of infinitely narrow photon beam in tissues are simulated by Monte Carlo method in this paper. Also presented are the energy distribution with regard to depths, light distribution in tissues, reflection and transmittance on the upper and lower surface. The optical parameters adopted in this study are g, albedo and microa, which have influence on energy distribution. The results show: The energy distribution decreases more quickly with the increase of depths and reveals a peak value close to the surface; g factor plays an important part in the lost energy on the upper surface and lower surface; the decrease of g factor causes weaking of the forward moving ability, so the penetration depth becomes smaller and the energy becomes dispersives variation of albedo has distinct effect on the shallow and deep tissues.
Computer Simulation ; Energy Transfer ; Light ; Models, Biological ; Monte Carlo Method ; Optics and Photonics ; Photochemotherapy ; methods ; Scattering, Radiation

A new transformation method for follow-up coordinate system is used and a matrix equation for transforming spatial transporting coordinate of scattering photons is given. The equation has recursion form and only relates to multiplication and will not lead to overflow in calculation. The equation is applied to Monte Carlo simulation in the light propagation in bio-tissues. The results show it can effectively save CPU time and improve simulation efficiency. The results are also in good agreement with the results from the traditional method.
Animals ; Computer Simulation ; Connective Tissue ; radiation effects ; Humans ; Light ; Models, Biological ; Monte Carlo Method ; Nephelometry and Turbidimetry ; methods ; Photons ; Refractometry ; methods ; Scattering, Radiation

Independent component analysis (ICA) is a new method of signal statistical processing and widely used in many fields. We face several problems such as the different nature of source signals (e.g. both super-Gaussian and sub-Gaussian sources exist), unknown number of sources and contamination of the sensor signals with a high level of additive noise in the analysis of signal. A robust approach was proposed to solve these problems in this paper. Firstly, observations (noisy data) possessing high dimensionality were preprocessed and decomposed into a source signal subspace and a noise subspace. Then the number of sources was got through the cross-validation method, and this solved the problem that ICA could not confirm the number of sources. At last the transformed low-dimensional source signals were further separated with the fast and stable ICA algorithm. Through the analysis of artificially synthesized data and the real-world Magnetoencephalographic data, the efficacy of this robust approach was illustrated.
Algorithms ; Humans ; Magnetoencephalography ; methods ; Principal Component Analysis ; Signal Processing, Computer-Assisted