Omid Haji Maghsoudi

A computer aided method to detect bleeding, tumor, and disease regions in Wireless Capsule Endoscopy

Wireless Capsule Endoscopy (WCE) is a relatively new technology to record the entire gastrointestinal (GI) tract, in vivo. A large amount of images (frames) are captured during the WCE examination. Reviewing this number of images by a gastroenterologist would be time consuming and prone to human error. Therefore, a diagnostic computer-aided technique is essential to detect and segment regions of abnormalities. In this study, a novel method based on textural features (such as Gabor filters, local binary pattern, and Haralick) in HSV color space, Fisher score test, and neural networks is presented to detect and differentiate regions such as bleeding, tumor, and other types of gastric diseases including Crohns, Lymphangectasia, Stenosis, Lymphoid Hyperslasia and Xanathoma. The experimental results indicate that this method is able to classify a lesion from a normal region in every single frame and group them into normal and abnormal frames to be considered for surgery/treatment planning by an expert.

Detection of abnormalities in wireless capsule endoscopy frames using local fuzzy patterns

Wireless capsule endoscopy (WCE) is a relatively new technology that captures images of the entire gastrointestinal tract non-invasively and painlessly. A large number of frames are captured during an examination. The main difficulty for physicians is to review the frames visually. In recent years, some methods have been devised to help physicians. Here, a novel method is devised to distinguish among frames whether the frames contain abnormal region or not. This goal is achieved using local fuzzy patterns (LFP) that is compared with rotation invariant local binary patterns (LBP). In addition to this comparison, four color channels (red, hue, green, and gray scale) are examined to find the best one for using LFP. Experimental results indicate that LFP is a powerful method for texture detection, and can develop the LBP results in the WCE frames.

Investigating the effect of poly-l-lactic acid nanoparticles carrying hypericin on the flow-biased diffusive motion of HeLa cell organelles

In this study, we investigate in human cervical epithelial HeLa cells the intracellular dynamics and the mutual interaction with the organelles of the poly‐l‐lactic acid nanoparticles (PLLA NPs) carrying the naturally occurring hydrophobic photosensitizer hypericin.

Detection of small bowel tumor in wireless capsule endoscopy images using an adaptive neuro-fuzzy inference system

Automatic diagnosis tool helps physicians to evaluate capsule endoscopic examinations faster and more accurate. The purpose of this study was to evaluate the validity and reliability of an automatic post-processing method for identifying and classifying wireless capsule endoscopic images, and investigate statistical measures to differentiate normal and abnormal images. The proposed technique consists of two main stages, namely, feature extraction and classification. Primarily, 32 features incorporating four statistical measures (contrast, correlation, homogeneity and energy) calculated from co-occurrence metrics were computed. Then, mutual information was used to select features with maximal dependence on the target class and with minimal redundancy between features. Finally, a trained classifier, adaptive neuro-fuzzy interface system was implemented to classify endoscopic images into tumor, healthy and unhealthy classes. Classification accuracy of 94.2% was obtained using the proposed pipeline. Such techniques are valuable for accurate detection characterization and interpretation of endoscopic images.Automatic diagnosis tool helps physicians to evaluate capsule endoscopic examinations faster and more accurate. The purpose of this study was to evaluate the validity and reliability of an automatic post-processing method for identifying and classifying wireless capsule endoscopic images, and investigate statistical measures to differentiate normal and abnormal images. The proposed technique consists of two main stages, namely, feature extraction and classification. Primarily, 32 features incorporating four statistical measures (contrast, correlation, homogeneity and energy) calculated from co-occurrence metrics were computed. Then, mutual information was used to select features with maximal dependence on the target class and with minimal redundancy between features. Finally, a trained classifier, adaptive neuro-fuzzy interface system was implemented to classify endoscopic images into tumor, healthy and unhealthy classes. Classification accuracy of 94.2% was obtained using the proposed pipeline. Such techniques are valuable for accurate detection characterization and interpretation of endoscopic images.

Detection of uninformative regions in wireless capsule endoscopy images

Wireless capsule endoscopy (WCE) is able to investigate the entire gastrointestinal tract including the small bowel. To reduce the reviewing time of captured images by gastroenterologists and increasing the accuracy rate for automatic detection of abnormalities, it is beneficial to remove regions which have less or no clinical information of small bowel texture (i.e., uninformative regions). In this research study, a multi-stage method including Chan-Vese, color range ratio, adaptive gamma correction method (AGCM), canny color edge detection operator, and morphological processing is proposed to detect these uninformative regions. The results support the effectiveness of the proposed method. In conclusion, the proposed method is a simple method to implement and performed well in removing the uninformative regions of small bowel images.

Matlab software for impedance spectroscopy designed for neuroscience applications

BACKGROUND Metal electrodes are a mainstay of neuroscience. Characterization of the electrical impedance properties of these cuffs is important to ensure successful and repeatable fabrication, achieve a target impedance, revise novel designs, and quantify the success or failure of implantation and any potential subsequent damage or encapsulation by scar tissue. NEW METHODS Impedances are frequently characterized using lumped-parameter circuit models of the electrode-electrolyte interface. Open-source tools to gather and analyze these frequency sweep data are lacking. Here, we present such software, in the form of Matlab code, which includes a GUI. It automatically acquires frequency sweep data and subsequently fits a simplified Randles model to these data, over a user specified frequency range, providing the user with the model parameter estimates. Also, it can measure an unknown impedance of an element over a range of frequencies, as long as an external resistor can be added for the measurements. RESULTS The tool was tested on five bright platinum nerve cuffs in vitro. The average charge transfer resistance, solution resistance, CPE value, and impedance magnitude were estimated. COMPARISON TO EXISTING METHODS The measured values of the impedance of cuffs were in agreement with the literature (Wei and Grill, 2009). Variation between cuffs fabricated as consistently as possible amounted to 10% for impedance magnitude and 4° for impedance phase. CONCLUSION The results show that this low-cost tool can be used to characterize a cuff across different conditions including after implantation. The latter makes it useful for a longer-term study of electrode viability.