Biomedical Engineering: Applications, Basis and Communications

Sleep deprivation adversely affects the psychological and physiological functions, but little information about the effects of poor sleeping on the emotional responses of individuals are available. In the present study, the effect of one night insufficient sleep on the emotional responses to visual stimuli was appraised using electrocardiogram (ECG) measures. We also tested the hypothesis that men and women differ in affective response, considering the sleep quantity. ECG signals of university students (20 men and 20 women) were recorded while viewing affective pictures. Images were selected from the International Affective Picture System and categorized as happy, fear, sadness, and relax. The data were characterized using 14 time- and frequency-based features. The Wilcoxon statistical test was applied to examine significant differences between two groups of insufficient and normal sleep. A significant effect of ECG parameters was observed in deficient sleep on the emotional responses. Mode, RMS, and mean power show the largest effect size in all affective states. Among the affective states, the most significant differences between the two groups were perceived while watching sad and relaxed pictures. In addition, analysis of affective ECG measures in men and women distinctly revealed that lack of sleeping is more pronounced in men. Sleep is associated with the emotional responses. Our results confirmed the effect of gender on affective physiological reactions.

Content-based image retrieval (CBIR) has turned into an important research field with the advancement in multimedia and imaging technology. The term CBIR has been widely used to describe the process of retrieving desired images from a large collection on the basis of features such as color, texture and shape that can be automatically extracted from the images themselves. Considering the gap between low-level image features and the high-level semantic concepts in the CBIR, we proposed an image retrieval system for brain magnetic resonance images based on saliency map. First, the proposed approach exploits the ant colony optimization (ACO) technique to measure the image’s saliency through ants’ movements on the image. The textural features are then calculated from the saliency map of the images. The image retrieval of the proposed CBIR system is based on textural features and the fuzzy approach using Adaptive neuro-fuzzy inference system (ANFIS). Regarding the various categories of images in a database, we define some membership functions in the ANFIS output in order to determine the membership values of the images related to the existing categories. In online image retrieval, a query image is introduced to the system and the relevant images can be retrieved based on query membership values into different classes including normal or tumoral. The experimental results indicate that the proposed method is reliable and has high image retrieval efficiency compared with the previous works.

Early and correct diagnosis of cardiac arrhythmias is an important step in the treatment of patients. In the recent decades, a wide area of bio-signal processing is allocated to cardiac arrhythmia classification. Unlike other studies, which have employed Electrocardiogram (ECG) signal as a main signal to classify the arrhythmia and sometimes they have used other vital signals as an auxiliary signal to fill missing data and robust detections. In this study, the Arterial Blood Pressure (ABP) is used to classify six types of heart arrhythmias. In other words, in this study for first time, the arrhythmias are classified according ABP signal information. Discrete Wavelet Transform (DWT) is used to de-noise and decompose ABP signal. On feature extraction stage, three types of features including frequency, power, and entropy are extracted. In classification stage, Least Square Support Vector Machine (LS-SVM) is employed as a classifier. The accuracy, sensitivity, and specificity rates of 95.75%, 96.77%, and 96.32% are achieved, respectively. Currently, the classification of cardiac arrhythmias is based on the ABP signal which has some advantages. The recording of ABP signal is done by means of one electrode and therefore it has resulted in lower costs compared with the ECG signal. Finally, it has been shown that ABP has very important and valuable information about the heart performance and can be used in arrhythmia classification.

Purpose: This paper presented a new approach to noninvasively evaluate the upper airway stenosis for obstructive sleep apnea syndrome (OSAS) patients. Methods: In the proposed method, thoracic and abdominal movements were selected to calculate the respiratory movement (RM) and to indicate the change of lung volume. Due to the cumulative effect of thermal sensor, the oronasal-thermistor signal (T_flow) is applied to estimate the air mass change in lung. Based on the mathematical relationship, the “RM–T_flow curves”, drawn by RM and T_flow, together with the correlation coefficients ($r_{\mathrm{\text{RM}}\text{-}\mathrm{\text{TF}}})$ were used to analyze the upper airway stenosis. Results: This method was verified through portable monitoring (PM) based experiments, and numerical analysis of the polysomnography (PSG) data from 20 OSAS patients and 15 non-OSAS controls. Our results indicate the $r_{\mathrm{\text{RM}}\text{-}\mathrm{\text{TF}}}$ values decrease with the narrowing of the upper airway. At each sleep stage, the $r_{\mathrm{\text{RM}}\text{-}\mathrm{\text{TF}}}$ mean values of OSAS subjects are significantly ($p<0.01$) smaller than those of the controls. These facts demonstrate that the $r_{\mathrm{\text{RM}}\text{-}\mathrm{\text{TF}}}$ value can be used to quantify the upper airway stenosis and the analysis of “RM–T_flow curves” is an efficient way to assess upper airway condition associated with the breathing phase. Conclusions: As this method can be used in spontaneous sleep and home sleep testing, we believe it will benefit the popularity of the diagnosis and evaluation of OSAS.

Electrophoretic mobility (EPM) measurement on biological particles in fluids is well established. The current method in measuring EPM is using laser which the target particles are not visible. Additional morphology information is critical for the EPM measurement. Image processing is a promising method to obtain the EPM together with the morphology information. In this study, a setup of micro electrophoresis system with a compact CCD microscope was constructed. This setup was equipped with image processing method for capturing the images of the moving particles in an electric field. With the image processing method (Horn–Schunck method), the images captured were processed in real time to obtain the EPM of the particle. Velocity of the particles was then measured and the particles’ EPM was obtained. With the captured images of the particles in real time, the system can present the image of the targeted particle together with the EPM value. The setup of this prototype was calibrated with discrete particles (Polystyrene microsphere size of 10$\mu$m$\pm$ 5%) and with a magnification value of 125X. This system is suitable for the surface charge measurement of discrete particle with size in between 4$\mu$m and 20$\mu$m. Comparison of commercialized device with our laboratory setup for calibration on EPM of polystyrene beads had a variance of solely 13%. Measurement on yeast cells, normal (hFob 1.19) and cancer bone cells (U2OS) indicated that the EPM of yeast became highly negative in the pH value of 4.5 and 6.5. The negative EPM of the cancer cell is slightly larger than that of the normal cell for pH ranging from 4.4 to 5.0. In conclusion, the real-time EPM measurement set up for this study is able to display the real-time images of the moving particles in fluid suspension during measurement.

The purpose of this work is to describe a chest radiography computer-aided diagnostic (CAD) scheme designed to analyze the chest radiographs performed in the framework of the Galician (Spain) Health Service (GHS), including the radiographs that are not reported by the radiologists. The final goal of this CAD system is its integration in the GHS daily clinical environment, with a feasible RIS-PACS-CAD and EHR-CAD integration model. The database of the study included 55 chest radiographies with 64 nodules/lung cancer. This database was used to develop and test the CAD system in our research laboratory. Free-Response Receiver Operating Characteristic (FROC) curves were employed to evaluate the performance of the CAD system. An independent database was employed to evaluate the performance of the CAD system by external radiologists. After the application of a linear classifier, our CAD system achieved a sensitivity of 70% with a false positive rate between 4 and 6 per image depending on the testing database. When compared with other commercial systems, our CAD scheme achieved similar performance results. Therefore, our CAD scheme could be utilized to help radiologists in the detection of lung nodules in chest radiography, and therefore, it can be integrated in the clinical practice.

Biomedical image compression plays an important role in the medical field. Mammograms are medical images used in the early detection of breast cancer. Mammogram image compression is a challenging task because these images contain information that occupies huge size for storage. The aim of image compression is to reduce the image size and the time taken for recovering the original image without any loss. In this paper, two different techniques of mammogram compression are introduced. The proposed algorithm includes two main steps. First, a preprocessing step is applied to enhance the image, and then a compression algorithm is applied to the enhanced image. The algorithm is tested using 322 mammogram images from the online MIAS database. Three parameters are used to evaluate the performance of the compression techniques; compression ratio (CR), Peak Signal to Noise Ratio (PSNR) and processing time.

According to the results, Haar wavelet-based compression for enhanced images is better in terms of CR of 26.25% and PSNR of 47.27dB.

Ventricular tachycardia (VT) is a fast heart rate that arises from improper electrical activity in the ventricular of the heart. VT may eventually lead to lethal ventricular fibrillation (VF) which is characterized by fast and irregular heart rhythm. Since difference between VT and VF is diagnosed by specialist in a critical and stressful situation, the possibility of wrong decision is not low. Here, various set of ECG features belonged to different domains are implemented to investigate the predictability and discriminability of VT and VF episodes. Informative features from different domains such as correlation dimension (phase space) and power spectrum (frequency domain) were elicited from electrocardiogram (ECG) signals to describe the amount of irregularity/variation through the attack. In addition raw signal samples were used to assess the classification task based on the time domain features. Applying correlation dimension, power spectrum and the raw samples of ECGs to artificial neural network (ANN) classifier provides 91%, 92% and 71% classification accuracy between VT and VF signals, respectively. However, to enrich the time domain features, surrogate data was generated and the results of time domain is increased up to 87% which represents that ANNs are able to learn the dynamic nature of chaotic signals.