Current Medical Imaging Reviews (v.11, #1)

Meet Our Editorial Board Member by David M. Goldenberg (1-1).

A Survey on Medical Image Segmentation by Saleha Masood, Muhammad Sharif, Afifa Masood, Mussarat Yasmin, Mudassar Raza (3-14).
Much work has been done in the field of Image segmentation but still there is a room for improvement. Medical image segmentation is a sub field of image segmentation in digital image processing that has many important applications in the prospect of medical image analysis and diagnostics. Here in this paper different approaches of medical image segmentation will be classified along with their sub fields and sub methods. Recent techniques proposed in each category will also be discussed followed by a comparison of these methods.

Segmentation of noisy and low-contrast images remains one of the most challenging and difficult tasks, especially in the context of medical imaging. In this work, we propose an extension of the Active Shape Models (ASM) which is based on a priori knowledge about the shape and the deformation modes of the studied Region(s) of Interest (ROI). The main contribution of the proposed extension resides in the integration of a statistical directional relationship within the ASM, which is learned during a training phase. In particular, in order to force the active contour to move towards points in space that satisfy the spatial relationship, we propose a fuzzy directional constraint that allows a more robust localization of ROI. In fact, the learned a priori knowledge has been modeled using fuzzy logic in order to model uncertainty and ambiguity of the spatial representation. Realized tests on scintigraphic and MRI images proved the performance of the proposed model for the detection of multiple objects of interest in noisy and low-contrast images, even when real contours are ill-defined.

Imaging in Drug Side Effects by Sandra Baleato-Gonzalez, Roberto Garcia-Figueiras, Maria Ageitos Casais, Amadeo Arango Diaz, Ivan Sanz-Falque, Joan C. Vilanova (23-37).
Drug side effects are common in clinical practice and its diagnosis and radiologic manifestations are not always evident or known. Adverse effects may cause medical complications and negatively affect prognosis and outcome of patients. In this setting, an early diagnosis might have relevant clinical and therapeutic implications. Different studies have shown that adverse drug reactions related hospital admissions comprise up to 10% of the total number of hospitalizations and are an important cause of morbidity and mortality in adults. Most adverse drugs reactions have no distinctive radiological features. However, certain iatrogenic disorders have distinctive imaging characteristics that allow their recognition. We illustrate the imaging findings of drugs side effects and review those complications that radiologists may diagnose.

Neonatal Atlas Templates for the Study of Brain Development Using Magnetic Resonance Images by Maryam Momeni, Hamid Abrishami Moghaddam, Reinhard Grebe, Catherine Gondry-Jouet, Fabrice Wallois (38-48).
The quantification of neonatal brain development has a significant role in understanding, prevention, diagnosis and treatment of nervous system diseases in infancy time. On the other hand, brain development and its morphological changes are very fast during the first weeks after birth. So, age-related brain atlases representing sharp anatomical features of a neonatal population are indispensable. In this paper, we constructed two neonatal brain atlases for the age ranges of 39-40 and 41-42 weeks' gestational age with 16 T1-weighted magnetic resonance images using an improved groupwise registration. Neonatal images were normalized to the newly created and previously available neonatal atlases. The similarity between these atlases and normalized images was calculated via mutual information. The mean of mutual information between normalized images and the new atlases using the proposed algorithm is larger. This result confirms the greater similarity between normalized images and the atlases created in this paper. Besides, neonatal brain development was analyzed using deformation based morphometry and brain anatomical changes could be seen in local regions.

Diabetic Retinopathy (DR) is the most common disease induced by the complication of diabetes, causing blindness. In many rural areas, the contributions of ophthalmologists are predicatively less to treat the disease. Detection of lesions in the early stage is a progressive measure to diagnose DR. Initially, a preprocessing method is performed to detect the Optic Nerve Head (ONH) in the lesion. Based on the degree of reflectance in ONH, feature extraction is computed using multi-scale Local Binary Pattern (LBP) algorithm. Here, Gabor convolution is estimated and the structure of ONH is encoded. This extends to a statistical computation in terms of the moment and standard deviation. A Support Vector Machine (SVM) classification is formulated to locate the hemorrhages and exudates and an effective probabilistic multi-label lesion classification is performed to acquire five sets of results representing the diabetic retinopathy: 1) Grade-1 Exudates, 2) Grade-2 Exudates, 3) Micro aneurysms, 4) Hemorrhages, 5) Neovascularization. Finally, the affected area of lesions is used to diagnose the disease.

Aim of Study: The lung cancer is noted at the end stages of disease, the morbidity and mortality rate related to is higher than others. To minimize this rate, early diagnosis of lung solitary nodules before spreading metastases to lymph nodes and other organs is of consideration. At the moment, Computerized Tomography is one of the most important modalities in diagnosing lung solitary nodules, however the CT exposure rates are much higher than diagnostic radiology field. Therefore it is necessary to do evaluation of CT. Scan ability in detecting chest nodules for preventing unnecessary radiation dose to patient.
Materials and Methods: In this study, a chest phantom including different nodules of sizes and types was designed. Imaging of phantom was performed by TOSHIBA spiral CT of Imam Khomaini medical complex and GE spiral CT of Fayyazbakhsh hospital with 5,3,1 mm slices and 80, 120 kVp and 50,60,80,100 mA.
Results: This study revealed, 4 mm width nodules were noted both in slices with 120 kV and 50, 100 mA and 3, 5 mm thickness and in 80 kV and 60 mA with 3 mm thickness. The calcium carbonated particles were noted in 6, 8, and 10 mm whitish but not in smaller nodules; however a low density of carbonated calcium was noted just in 10 mm width nodules.
Conclusion: CT.scan is a useful technique for detection of lung tumor with sizes more than 4mm.

In the ultrasound image, the gallbladder has a wide variety of shape, type, size, and slope, so the existing researches into contour-based gallbladder detection cannot be applied to the various types of gallbladder. Therefore, in this paper, we present a new approach to locate the gallbladder from ultrasound image using ends-in search stretching and self organizing map (SOM)-based color quantization. We use the ends-in search stretching technique to emphasize the difference in brightness among the structures, and SOM-based color quantization is used to classify pixels into the gallbladder and gallbladder wall. After SOM-based color quantization, we extract gallbladder candidates. To verify the gallbladder candidates, we define five morphologic characteristics-based rough conditions. We experiment using 30 ultrasound images. Experiment results shows the proposed method locates the gallbladder excellently despite gallbladders have a variety of shapes.