[Designing a new CAD system for pulmonary nodule detection in High Resolution Computed Tomography [HRCT] images]

Lung diseases and lung cancer are among the most dangerous diseases with high mortality in both men and women. Lung nodules are abnormal pulmonary masses and are among major lung symptoms. A Computer Aided Diagnosis [CAD] system may play an important role in accurate and early detection of lung nodules. This article presents a new CAD system for lung nodule detection from chest computed tomography [CT] images. Twenty-five adult patients with lung nodules in their CT scan images presented to the National Research Institute of Tuberculosis and Lung Disease, Masih Daneshvari Hospital, Tehran, Iran in 2011-2012 were enrolled in the study. The patients were randomly assigned into two experimental [9 female, 6 male, mean age 43 +/- 5.63 yrs] and control [6 female, 4 male, mean age 39 +/- 4.91 yrs] groups. A fully-automatic method was developed for detecting lung nodules by employing medical image processing and analysis and statistical pattern recognition algorithms. Using segmentation methods, the lung parenchyma was extracted from 2-D CT images. Then, candidate regions were labeled in pseudo-color images. In the next step, some features of lung nodules were extracted. Finally, an artificial feed forward neural network was used for classification of nodules. Considering the complexity and different shapes of lung nodules and large number of CT images to evaluate, finding lung nodules are difficult and time consuming for physicians and include human error. Experimental results showed the accuracy of the proposed method to be appropriate [P<0.05] for lung nodule detection

Neuroimaging in Iran: a review

Neuroimaging allows noninvasive evaluation of the anatomy, physiology, and function of the brain. It is widely used for diagnosis, treatment planning, and treatment evaluation of neurological disorders as well as understanding functions of the brain in health and disease. Neuroimaging modalities include X-ray computed tomography [CT], magnetic resonance imaging [MRI], single photon emission computed tomography [SPECT], positron emission tomography [PET], electroencephalography [EEG], and magnetoencephalography [MEG]. This paper presents an overview of the neuroimaging research in Iran in recent years, partitioned into three categories: anatomical imaging; anatomical image analysis; and functional imaging and analysis. Published papers reflect considerable progress in development of neuroimaging infrastructure, hardware installation and software development. However, group work and research collaborations among engineers, scientists, and clinicians need significant enhancement to optimize utility of the resources and maximize productivity. This is a challenge that cannot be solved without specific plans, policies, and funding

[Assessment of the wavelet transform for noise reduction in simulated PET images]

An efficient method of tomographic imaging in unclear medicine is positron emission tomography [PET]. Compared ta SPECT, PET has the advantages of higher levels of sensitivity, spatial resolution and more accurate quantification. However, high noise levels in the image limit its diagnostic utility. Noise removal in nuclear medicine is traditionally based on Fourier decomposition of images. This method is based on frequency components, irrespective of the spatial location of the noise or signal. The wavelet transform presents a solution by providing information on the frequency content while retaining spatial information. This alleviates the shortcoming of the Fourier transform and thus, wavelet transform has been extensively used for noise reduction, edge detection and compression. In this research, we used the SimSET software to simulate PET images of the NCAT phantom. The images were acquired using 250 million counts in a 128 x 128 matrix. For the reference image, we acquired an image with high counts [6 billion]. Then, we reconstructed these images using our own software developed in MATLAB. After image reconstruction, a 250 million counts image [noisy image] and a reference image were normalized and then root-mean-square error [RMSE] was used to compare the images. Next, we wrote and applied de-noising programs. These programs were based on using 54 different wavelets and 4 methods. De-noised images were compared with the reference image using RMSE. Our results indicate that the Stationary Wavelet Transform and Global Thresholding are more efficient at noise reduction compared to the other methods that we investigated. The wavelet transform is a useful method for de-noising of simulated PET images. Noise reduction using this transform and loss of high-frequency information are simultaneous with each other. It seems that we should attend to the mutual agreement between noise reduction and the visual quality of the image