Lung nodules are one of the most common pathological changes, thus early detection of lung nodule is very important for the diagnosis medical treatment of lung eancer. In recent years, as the application of multi-slice spiral CT(MSCT), high-resolution CT(HRCT) and low-dose chest CTCLDCT), computer-aided diagnosis (CAD) system will be more essential and more important. Since CAD system can improve the working efficiency of doctors and provide service to more patients, has become the research hotspot and achievement has been made in relevant area internationally recently. This review summarizes the basic methods and applieations of computer-aided detection and diagnosis of lung nodule based on CT image.

Objective To deal with blind source separation(BSS) more effectively in the field of mixed signal separations of strong and week sources.Methods According to the consistency between array signal processing model and BSS model,the real sources were estimated under linear constrains and least mean square(LMS),based on minimum output energy(MOE).EEG and evoked potential(EP) were used as strong background noise and week signal source separately in our experiment.The mixed signals were separated with the method proposed in this paper.Results The EP could be seperated from the strong noise EEG effectively.Conclusion Compared with typical BSS approaches,this new algorithm need not solve the unmixing matrix,so it runs fast,is of a little low computational complexity and can correctly estimate the weak signal source from low signal/noise(S/N) ratio.

Objective To analyze the texture features of SPIO-enhanced MR imaging in rat models of hepatocellular carcinoma (HCC) and hepatocirrhosis with gray level co-occurrence matrix (GLCM). Methods HCC and hepatocirrhosis models were established in rats. SPIO-enhanced MR images were obtained. A total of 161 regions of interests (ROIs, 81 of HCC and 80 of hepatocirrhosis) were selected manually. Feature values as angular second moment, contrast, correlation, inverse difference moment, entropy, variance were extracted based on GLCM. The differences of feature values between two groups were statistically analyzed. Results In SPIO-enhanced MR images, hypointense signal changes were found in hepatocirrhosis, as well as hyperintensity in HCC nodules and intermixed intensity in larger HCC nodules. Correlation and entropy values of HCC group were higher than that of hepatocirrhosis group, while the angular second moment, contrast, inverse difference moment, and variance values were lower than hepatocirrhosis group. Conclusion The feature values based on GLCM could be used for the further computer aided diagnosis of SPIO-enhanced MR images in rat models of HCC and hepatocirrhosis.

Automatic detection of epileptic events is of significance in clinical application. It is helpful to reduce the electroencephalogram analysts' workload. This paper summarizes and analyzes the detection of epileptic events by traditional and especially advanced methods, including nonlinear filtering, template matching, mimetic, wavelet transform and artificial neural network.
Electroencephalography ; Epilepsy ; diagnosis ; Humans ; Neural Networks (Computer) ; Signal Processing, Computer-Assisted ; Wavelet Analysis

The automatic spike detection in EEG is significant in both diagnosing epilepsy and alleviating the heavy labor force of the doctors. This paper proposes an empirical model decomposition (EMD) based epileptic spike detection method. It extracts the high frequency components related to spikes in EEG signal by EMD, and it detects the spikes by calculating the instantaneous amplitude of the high component with Hilbert transform. The results of experiments show that the method works well.
Algorithms ; Electroencephalography ; methods ; Epilepsy ; diagnosis ; physiopathology ; Humans ; Principal Component Analysis ; methods ; Signal Processing, Computer-Assisted

It is of great importance for the detection of epilepsy in clinical applications. Based on the limitations of the common used approximate entropy (ApEn) in the epilepsy detection, this paper analyzes epileptic EEG signals with the sample entropy (SampEn) approach, a new method for signal analysis with much higher precision than that of the ApEn. Data analysis results show that the values from both ApEn and SampEn decrease significantly when the epilepsy is burst. Furthermore, the SampEn is more sensitive to EEG changes caused by the epilepsy, about 15%-20% higher than the results of the ApEn.
Algorithms ; Data Interpretation, Statistical ; Electroencephalography ; methods ; Entropy ; Epilepsy ; diagnosis ; physiopathology ; Humans ; Nonlinear Dynamics ; Signal Processing, Computer-Assisted

Nowadays the tremendous amount of data has far exceeded our human ability for comprehension, and this has been particularly true for the medical database. However, traditional statistical techniques are no longer adequate for analyzing this vast collection of data. Knowledge discovery in database and data mining play an important role in analyzing data and uncovering important data patterns. This paper briefly presents the concepts of knowledge discovery in database and data mining, then describes the rough set theory, and gives some examples based on rough set.
Artificial Intelligence ; Clinical Medicine ; Data Interpretation, Statistical ; Databases as Topic ; Databases, Factual ; Decision Making, Computer-Assisted ; Diagnosis ; Factor Analysis, Statistical ; Knowledge ; Mathematical Computing ; Medical Records Systems, Computerized

The latency change detection of EPs is of special interest in many clinical applications such as diagnosis of the injury and pathological changes in the nervous system. This paper reviews the adaptive latency change detection approaches under stable noise conditions, based on the fractional lower order statistics. It also evaluates and compares the performances of the presented algorithms.
Algorithms ; Animals ; Brain ; physiology ; Electroencephalography ; methods ; Evoked Potentials ; physiology ; Humans ; Reaction Time ; physiology ; Signal Processing, Computer-Assisted

The extraction of evoked potentials is a main subject in the area of brain signal processing. In recent years, the single-trial extraction of evoked potentials has been focused on by many studies. In this paper, the approaches based on the wavelet transform, the neural network, the high order acumulants and the independent component analysis are briefly reviewed.
Algorithms ; Brain ; physiology ; Electroencephalography ; Evoked Potentials ; physiology ; Humans ; Neural Networks (Computer) ; Signal Processing, Computer-Assisted

Hilbert-Huang transform (HHT) is a new time-frequency analytic method to analyze the nonlinear and the non-stationary signals. The key step of this method is the empirical mode decomposition (EMD), with which any complicated signal can be decomposed into a finite and small number of intrinsic mode functions (IMF). In this paper, a new EMD based method for suppressing the cross-term of Wigner-Ville distribution (WVD) is developed and is applied to analyze the epileptic EEG signals. The simulation data and analysis results show that the new method suppresses the cross-term of the WVD effectively with an excellent resolution.
Algorithms ; Electroencephalography ; methods ; Epilepsy ; physiopathology ; Humans ; Nonlinear Dynamics ; Signal Processing, Computer-Assisted