Biomedical Engineering: Applications, Basis and Communications

Human induced pluripotent stem (iPS) cells represent an ideal source for patient specific cell-based regenerative medicine. For practical uses of iPS cells, large-scale, cost- and time-effective production of fully reprogrammed iPS cells from a number of patients should be achieved. To achieve this goal, culture protocols for inducing iPS cells as well as methods for selecting fully reprogrammed iPS cells in a mixture of cells which are still in reprogramming and non-iPS differentiated cells, should be improved. This paper proposes a convolutional neural network (CNN) structure to classify a bright-field microscopy image as respective probability images. Each probability image represents regions of differentiated cells, fully reprogrammed iPS cells or cells still in reprogramming, respectively. The CNN classifier was trained by multiple types of image patches which represent differentiated, reprogramming and reprogrammed iPS cells, etc. Classification of an image containing the confirmed iPS cells by the trained CNN classifier shows that high classification accuracy can be achieved. Classifications of sets of time-lapse microscopy images show that growth and transition from CD34$+$ human cord blood cells through reprogramming to reprogrammed iPS cells can be visualized and quantitatively analyzed by the output time-series probability images. These experiment results show our CNN structure yields a potential tool to detect the differentiated cells that possibly undergo reprogramming to iPS cells for screening reagents or culture conditions in human iPS induction, and ultimately further understand the ideal culturing conditions for practical use in regenerative medicine.

Electrical impedance tomography is a recently established technique by which impedance of an object (medical or nonmedical applications) is measured data from the surface of the object, and a numerically simulated reconstruction of the object internal shape of the image can be obtained. This imaging technique based on boundary or surface voltage is measured when the different current pattern is injected into it. For current pulse, we are creating a voltage controlled current source, which is based on the different RC circuits, according to current amplitude and frequency values. The current source used in inject the current pulse of the various phantoms. The current position and measuring voltage is controlled by the created control unit or programmable system on chip (PSOC) of the proposed EIT system. After that image reconstruction of the cross-sectional image of resistivity requires sufficient data collection from used phantoms, which is based on finite element method (FEM) method and Tikhonov regularization method with helps of graphical user interface (GUI) on MatLab. The objective of the GUI was to produce an image (2D/3D), impedance distribution graph, and the FEM mesh model according to used electrode combinations from the various phantoms. EIT system has a great potential for imaging modality, is non-invasive, radiation-free, and inexpensive for medical applications.

Medical image segmentation is a vital process in medical diagnosis and evaluation of tumor response to therapy. Current segmentation methods works only on single modality image like positron emission tomography has low resolution and gives only functional information; Computed Tomography has low contrast and provides structural information. This paper focus on segmentation of multimodality PET-CT image. In recent days PET-CT is advanced multimodal imaging equipment, which gives both functional and anatomical information in a single image. Probability random index is a new methodology adopted to segment the portion of an image, which is most essential for determining the actual intricacies involved in the portion of a body. The clustering is another methodology used to group similar pixel locations into a single group based on unpredictable random values of an image. The probability based clustering has been incorporated to overcome the drawbacks of existing methods of segmentation like over segmentation and under segmentation. The over segmentation has been eliminated by incorporating random values generated from the features of dataset of images. Similarly under segmentation has been eliminated by removing barriers of lack of collecting similarly values from clustering. Thus, the proposed method eliminates both over segmentation and under segmentation drawbacks of the existing methods. The proposed probability random index based clustering has yielded good results in comparision with other contemporary methods, which shall be observed from the section, results and analysis. The proposed probability random indexed clustering has yielded a good result of 88.41% on benchmark dataset.

Close range photogrammetry is an image-based method of measurement that can be used to create a three-dimensional model of objects. This method is very popular due to its high speed, low cost, and non-invasiveness as a measurement tool in medicine. The main weakness of these systems is the lack of topology between points, which is especially important in the medical field. The neural gas network can learn the topology of the points associated with the 3D model of objects. Considering the capabilities mentioned regarding the photogrammetry and the neural gas network, a combination of these can be used as a diagnostic tool in such a way that in addition to considering the points of the model, neighboring points are also considered in the diagnostic process. Accordingly, in this study, we want to design a system by combining these two tools that can recognize diseases by their apparent symptoms. Moreover, the use of the neural gas network and the possibility of local and general examination of the organs increase the accuracy of diagnosis. Diagnosis of foot disease has been used as a case study in this system. The results showed that the neural gas network has a high degree of flexibility for modeling the human body compared to previous methods, and provides a better approximation. Also, the accuracy of reconstruction of the 3D model of the object is effective in the process of diagnosis and influences the level of intelligence of the system as well. Finally, in the system implemented, the results showed that the disease was correctly diagnosed in all 5 feet of the patient. Also, in 4 cases, the location of the disease was correctly detected and in one case the location of the disease was misdiagnosed.

Back Ground: Liver tumors are a type of growth found in the liver which can be categorized as malignant or benign. It is also called as hepatic tumors. Early stage detection of tumor could be treated at a faster phase; if it is left undiagnosed it may lead to several complications. Traditional method adopted for diagnosis can be time consuming, error-prone and also requires an experts study. Hence a non invasive diagnostic method is required which overcomes the flaws of conventional method. Liver segmentation from CT images in post processing techniques not only is an essential prerequisite, but, by playing an important role in confirming liver function, pathological, and anatomical studies, is also a key technique for diagnosis of liver disease. Hence in the proposed study Fast greedy snakes algorithm in abdominal CT images were used for segmenting tumor portion.

Aim & Objectives: The aim and objectives of study is: (i) to segment tumor region in the liver image using Fast Greedy Snakes Algorithm (FGSA); (ii) to extract the GLCM features from the segmented region; (iii) to classify the normal and abnormal liver image using neural network classifier.

Methodology: The study involved a total of 30 normal and 30 abnormal Images from database. In the proposed study automated segmentation was performed using Fast Greedy Snakes (FGS) Algorithm and the features were extracted using GLCM method. Classification of normal and abnormal images was carried out using Back propagation Neural Network classifier.

Result: The proposed FGS algorithm provides accurate segmentation in liver images. Statistical features like mean, kurtosis, correlation and Entropy showed a higher value for the normal image than liver tumor image. On the other hand, features like Skewness, Homogeneity, contrast, Energy and standard deviation showed a comparatively higher value for a liver tumor image than the normal. Statistical features such as Mean, Contrast, Homogeneity and standard deviation are statistically significant at $p<0.01$. Features like correlation, entropy and energy exhibits significance at $p<0.05$. The feature extracted values provided significant difference between the normal and abnormal liver images. The neural network classifier yields the sensitivity of 95.8%, sensitivity of 81.4% and achieved the overall accuracy of 92%.

Conclusion: A most accurate, reliable and fast automated method was implemented to segment the liver tumor image using Fast Greedy snakes algorithm. Hence the proposed algorithm resulted in effective segmentation and the classifier could classify the normal and abnormal images with greater accuracy.

The dynamic threshold algorithm (DTA) presented by Pan Tompkins is a popular QRS detection method, and it has high sensitivity and specificity. However, the accuracy of this algorithm would be compromised if its sensitivity is increased. In this study, an enhanced dynamic threshold algorithm (EDTA) based on dynamic threshold rules is proposed, which add a compensation scheme to reduce the rate of misdetection and missed detection of R wave in low signal-to-noise ratio condition, sensitivity and detection error rate are calculated on simulated and clinical data to compare the performance between EDTA and DTA, and EDTA yields a competitive results. For the clinical data, the average accuracy rate of EDTA is 99.24%, which is higher than that of DTA at 95.98%. Further compared experiments among EDTA and the two other popular algorithms are conducted and the results of their validation over a public database are given and discussed, which prove our superiority.

In medicine, data collection plays an important role in the diagnosis of diseases and treatment of patients. Physicians have to cope with a large amount of patient-related data, and they often have to review the patient’s whole history or other similar cases. Data is mainly collected to find out if there are related patterns and results that can shed light on the nature of the investigated disease. Mechanized data mining has tremendous value in the diagnosis and treatment of diseases, and can be especially helpful in the diagnosis and treatment of diabetes, a disease that inflicts a large portion of the population. Now, diabetes is the fourth cause of mortality among the general population in developing countries. Retinopathy is a chronic complication of diabetes that has serious consequences including blindness if not diagnosed as early as possible. This study uses a sample of 310 Diabetic patients, half of them have the diagnosis of retinopathy ($n=155$) and investigates 29 variables including age, gender, HbA1c, treatment type and etc. Our results indicate that Decorate algorithm in Weka software is the most vigorous algorithm for the purpose of the study with an accuracy rate of 0.86. The study also investigates efficacy criteria related to databases and risk factors related to this disease including age, duration of the disease, BMI, HDL level, HbA1c, FBS, 2HPPG, blood pressure, and treatment method.

Active implantable medical devices (AIMDs) like implantable cardiac pacemakers play very important role in extending lives of patients with some cardiovascular diseases. The life of implantable device depends on life of battery. If this device can be charged from outside with power transfer device, then the cost of surgical procedures for patient can be saved. One must ensure, while transferring this power there should not be any abnormal effect on human body tissues. Wireless recharging of such devices through magnetic resonant coupling is of concern and hence the topic of more research to have uninterrupted supply from battery. The technique of wireless power transfer, primary or transmitting coil is assumed to be on body and receiver coil is assumed to be inside the human body. Several critical aspects need to be studied while designing coil for wireless power transfer (WPT). One of which is choice of operational frequency. In this research experiment, designed circuit is tested for checking power transfer was studied. Effect of the distance between primary and secondary coil affects the efficiency of power transfer. Authors also tied to test this for using different medium like air, placing 80 GSM paper and cloth. It is found that the medium between the primary and secondary affects the transfer of power. Careful thought needs to be given while designing power transfer system.