Current Medical Imaging - Volume 11, Issue 4, 2015

The purpose of this study was to evaluate the clinical value of PET (Positron Emission Tomography) for early prediction of tumor response to platinum-based therapy in patients with nonsmall cell lung cancer (NSCLC). The evaluation was carried out comparing the standard treatment response using RECIST (Response Evaluation Criteria in Solid Tumors) with metabolic treatment response according to European Organization for Research and Treatment of Cancer (EORTC) recommendations, PET Response Criteria in Solid Tumors (PERCIST), Total Lesion Glycolysis (TLG) and Metabolic Tumor Volume (MTV). Seventeen inoperable patients with stage IV NSCLC were enrolled between October 2011 and June 2013: PET studies were carried out before the initiation of platinum-based therapy and after the first cycle of chemotherapy for an early therapy monitoring. The lesions with the highest uptake in each patient were evaluated according to EORTC recommendations considering a cut-off of 15% (EORTC15%) and 25% (EORTC25%) to discriminate between patients who respond from those who do not respond to treatment. Moreover, PERCIST and RECIST classifications were evaluated too. Receiver operating characteristic curves were used to obtain cut-off points for therapy evaluation based on variations of TLG and MTV in sequential scans. Overall Survival (OS) time was calculated by the Kaplan-Meier test. The Kaplan-Meier analysis showed that RECIST, EORTC15%, ΔTLG, and ΔMTV proved to be a significant prognostic factor for predicting OS (p-value ≤ 0.0251). For responder patients, median OS was 595 days for RECIST, 423 for EORTC15% and ΔMTV, 492 for ΔTLG. For nonresponder patients, median OS was 238 days for RECIST and ΔTLG, 194.5 for EORTC15%, and 188 for ΔMTV. No statistically significant difference was recognized between responder and non-responder patients according to EORTC25% and PERCIST classifications (p-value ≥ 0.13). In addition, a new threshold of 17% for PERCIST classification was proposed for an early therapy monitoring rather than the conventional cut-off of 30%. Even if this is a preliminary analysis, the results suggest that PET examinations could provide an early identification of patients who benefit from platinum-based treatment.

Aim: Breast ultrasonography (US) is a widely used and effective screening method, particularly in dense breasts. However, the fact that breast US examination takes a long time is one of the most important limiting factors. In this study, we investigated how to perform this important modality in the shortest time and in the most effective way, in terms of diagnostic usefulness. Materials and Methods: Breast ultrasonography was performed on 141 patients with the aim of screening and diagnosis. US was performed on both breasts in radial and transverse plus parasagittal planes, in which we compared the time spent during the examination and the efficiency of lesion detection. Additionally, we investigated whether the efficiency of lesion detection was related to age, weight, BI-RADS (Breast Imaging Reporting and Data System), breast parenchymal pattern, and lesion size. Wilcoxon, two-proportions and chi-square tests were used to compare variables. A p-value < 0.05 was accepted to be significant. Results: A total of 141 patients (mean age of 45) were included in the study. The mean time spent during the radial plane was ten minutes, while the transverse plus parasagittal plane took seven minutes (p<0.05). Moreover, parenchymal pattern and lesion size had a significant relationship with lesion detection (p<0.05). Lesion detection was more common in patients with type 3 pattern, especially in the radial plane. Conclusion: Screening US is important, especially in dense breasts. Despite the increased examination time, the radial plane must be the preferred method in order not to underestimate small size lesions.

Parasitical diseases are among the first cause of hospitalization and mortality. The diagnostics of intestinal parasitical diseases is based on stools specimens test through the microscopic image. The manual evaluation of microscopic images is time consuming and depends on the human expert. Digital images are extensively used in medicine for diagnostic and also for survey guidance. We present in this work a method of parasite extraction from microscopic images. The extraction scheme has four main steps: The first step is the edge detection using the multi-scale wavelet transform. The second step uses edges to locate the region of interest on the image by looking for the round or approximately round objects through the Hough transform. The circles detected by the Hough transform are used as the initial contour for the active contour. The third step applied an active contour model to locate the contour of the parasite. The last step uses this contour to extract the parasite through the logic operation with the original image and the mask corresponding to the interior of the contour. Experimental results show that the proposed scheme is very efficient for the extraction of parasite from the stools images. It has accurate segmentation ability despite of the poor quality or complex background of microscopic stools images.

Organ segmentation and identification from computer tomography images is a time consuming process. Organ identifying is one of the crucial steps to give effective radiotherapy. Due to partial volume effects, gray-level similarities of adjacent organs, contrast media affect, and the relatively high variations of organ position and shape, repeatable feature values gives mismatched organs and gives wrong decision making. To overcome these difficulties, Feature based Organ Identification using Signature Quadratic Form Distance (SQFD) method is proposed to recognize the organs such as Lung, Liver, Heart, Abdominal Aorta, Spinal Cord, and Bones from CT image series. In our Analysis several features such as GLCM features, Region features, Wavelet features, Tamura Features, Histogram Intensity features were analyzed and found certain features are used to identify the organs to give effective radiation therapy. The proposed method of identifying organs has been successfully tested among 100 patients and comparing results with manually contouring by experts. Result gives 99% accuracy.

In present day societies, cancer is a widely spread disease that affects a large proportion of the human population, many research teams are developing algorithms to help medics to understand this disease. In particular, tumor growth has been studied from different viewpoints and different mathematical models have been proposed. Our aim is to make predictions about shape growth, where shapes are given as domains bounded by a closed curve in R2. These predictions are based on geometric properties of plane curves and vectors. We propose two methods of prediction and a comparison between them is shared. Both methods can be used to study the evolution in time of any 2D and 3D geometrical forms such as cancer skin and other types of cancer boundary. The first method is based on observations in the normal direction to the plane curve (boundary) at each point (normal method). The second method is based on observations at the growing boundaries in radial directions from the "center" of the shape (radius method). The real data consist of at least two input curves that bind a plane domain.

Aim: The subject of our study is to investigate the accuracy of a typical PET/CT system applying 18F-FDG to determine activity of small, hot volumes (HV) by phantom measurements under the impact of reconstruction techniques (RT) with and without Time-of-flight-Mode (TOF), activity concentration ratios into the phantom (ACR), density variations and relevant analysis parameters. Methods: We investigated a body phantom containing up to 7 spheres (d = 6-37mm) on a topical PET/CT system (SIEMENS BIOGRAPH mCT40). Recovery coefficients (RC) of small HVs dependent on the ACR and several RTs (iterative default (IT), iterative high-definition including the system Point-Spread-Function (HD), HD+TOF (UHD), filtered backprojection (FBP) and FBP+TOF (FBPT) were analysed. In addition, we determined RCs with respect to the matrix size, number of iterations, filter settings and a phantom insert of lung and bone density equivalent material. Results: The application of UHD in particular reproduces true activities of small volumes best, followed by HD, IT, FBPT and FBP. Lung and bone equivalent density inserts reduce the RCs by up to 19%. Especially for small HVs the RC can be raised by increasing the matrix size and iterations. Conclusions: Novel technology and analysis have a strong impact on the quantification. Hence, new guidelines are required enabling comparisons to other, prior systems. In order to get a precise activity determination of small structures, we recommend the UHD algorithm, high resolution matrix and high number of iterations with respect to acquisition time and administered activity.

Glaucoma is a neurodegenerative disease and is considered as one of the most common reasons of blindness. Degeneration of nerves is an irreversible process so early diagnosis of this disease is inevitable to avoid permanent loss of vision. Diagnosis of glaucoma usually consists of inspection of the structural deterioration of nerve in conjunction with the examination of visual function. This article presents a detailed overview of glaucoma, its symptoms and the potential it has to affect the persons. The research work related to its diagnosis and treatment is also reviewed in this article.

Background and objectives: It is important to obtain comparable results from different densitometry scanners in especially experimental studies. The aim of this experimental study in rat femurs is to compare bone densitometry results determined by two different DXA scanners of the same manufacturer performed at two different institutes. Materials and methods: Sixty female Wistar rats were the subject of this study. The rats were divided into four group and the groups were as follows: group 1 (n=15) (control), group 2 (n=15) (sham operated), group 3 (n=15) (ovariectomized) and group 4 (n=15) (ovariectomized and treated with losartan for eight weeks). Femurs of rats were collected and analyzed by Norland XR-36 and Norland XR-46 model DXA devices at different institutes by the same staff. In order to compare the two scanners results Paired samples T test, ANOVA and post hoc test was performed and p<0.05 considered significant. Results: The (bone mineral concentration) BMC and (bone mineral density) BMD results of two different scanners were not significantly different according to statistical analysis (p>0.05) in the all 4 groups. Additionally BMD values were significantly decreased in group 3 compared to the control group (p<0,005). Although losartan treatment caused a significant increase in the BMD values of group 4 compared to the ovariectomy group (p<0.005) this increase could not reach the BMD values in the control group. Conclusions: Although it is important to use the same scanner for follow up DXA measurements in human, this experimental rat study shows that different scanners of the same manufacturer provides comparable results.