Current Medical Imaging - Volume 16, Issue 7, 2020

Various dental disorders, such as lesions, masses, carries, etc. may affect the human dental structure. Dental radiography is a technique, which passes X-rays through dental structures and records the radiographic images. These radiographic images are used to analyze the disorders present in the human teeth. Preprocessing is a primary step to enhance the radiographic images for further segmentation and classification of images. In this work, the preprocessing techniques such as unsharp masking using high pass filter, bi-level histogram equalization and hybrid metaheuristic have been utilized for dental radiographs. The performance measures of the preprocessing techniques were analyzed. Results demonstrate that a hybrid metaheuristic algorithm for dental radiographs achieves higher performance measures when compared to other enhancement methods. An average Peak Signal-to-Noise Ratio (PSNR) value of 21.6 was observed in the case of a hybrid metaheuristic technique for dental image enhancement.

Cancer is currently one of the main health issues in the world. Among different varieties of cancers, skin cancer is the most common cancer in the world and accounts for 75% of the world's cancer. Indeed, skin cancer involves abnormal changes in the outer layer of the skin. Although most people with skin cancer recover, it is one of the major concerns of people due to its high prevalence. Most types of skin cancers grow only locally and invade adjacent tissues, but some of them, especially melanoma (cancer of the pigment cells), which is the rarest type of skin cancer, may spread through the circulatory system or lymphatic system and reach the farthest points of the body. Many papers have been reviewed about the application of image processing in cancer detection. In this paper, the automatic skin cancer detection and also different steps of such a process have been discussed based on the implantation capabilities.

Malignant melanoma is considered as one of the most deadly cancers, which has broadly increased worldwide since the last decade. In 2018, around 91,270 cases of melanoma were reported and 9,320 people died in the US. However, diagnosis at the initial stage indicates a high survival rate. The conventional diagnostic methods are expensive, inconvenient and subject to the dermatologist’s expertise as well as a highly equipped environment. Recent achievements in computerized based systems are highly promising with improved accuracy and efficiency. Several measures such as irregularity, contrast stretching, change in origin, feature extraction and feature selection are considered for accurate melanoma detection and classification. Typically, digital dermoscopy comprises four fundamental image processing steps including preprocessing, segmentation, feature extraction and reduction, and lesion classification. Our survey is compared with the existing surveys in terms of preprocessing techniques (hair removal, contrast stretching) and their challenges, lesion segmentation methods, feature extraction methods with their challenges, features selection techniques, datasets for the validation of the digital system, classification methods and performance measure. Also, a brief summary of each step is presented in the tables. The challenges for each step are also described in detail, which clearly indicate why the digital systems are not performing well. Future directions are also given in this survey.

Recently, medical imaging and machine learning gained significant attention in the early detection of brain tumor. Compound structure and tumor variations, such as change of size, make brain tumor segmentation and classification a challenging task. In this review, we survey existing work on brain tumor, their stages, survival rate of patients after each stage, and computerized diagnosis methods. We discuss existing image processing techniques with a special focus on preprocessing techniques and their importance for tumor enhancement, tumor segmentation, feature extraction and features reduction techniques. We also provide the corresponding mathematical modeling, classification, performance matrices, and finally important datasets. Last but not least, a detailed analysis of existing techniques is provided which is followed by future directions in this domain.

Computer-aided diagnosis (CAD) systems are the best alternative for immediate disclosure and diagnosis of skin diseases. Such systems comprise several image processing procedures, including segmentation, feature extraction and artificial intelligence (AI) based methods. This survey highlights different CAD methodologies for diagnosing Melanoma and related skin diseases. It has also discussed types, stages, treatments and various imaging techniques of skin cancer. Currently, researchers developed new techniques to detect each stage. Extensive studies on melanoma cancer detection were performed by incorporating advanced machine learning. Still, there is a high need for an accurate, faster, affordable, portable methodology for a CAD system. This will strengthen the work in related fields and address the future direction of a similar kind of research.

Breast cancer is the second leading cause of cancer death among women preceded by cervix cancer. It has been reported that at the early stage of detection there is 85% chance of getting cured, whereas only 10% chance at later stage diagnosis. The current screening modalities are expensive, they have intricate imaging measures and they are unhealthy due to radiation exposure. Therefore, a screening tool that is non-invasive, has no connection with the body, free from radiation, such as Medical Thermography is necessary. It is reported that the sensitivity and specificity of medical thermography are high largely in dense breast tissues. The clinical interpretation primarily depends on the asymmetrical analysis of these thermograms subjectively. The appearance of an asymmetric thermal image may indicate the pathological conditions. For earlier detection of breast cancer, it is essential to identify the advanced methods in image processing techniques which enhance the significance of diagnostics. In that analysis, the required breast region is unglued from the background image. The segmented image is separated into symmetrical left and right breast tissues. The statistical and histogram features extracted from both regions are used to identify the abnormal thermograms using machine learning algorithms. From literature, it is reported that the thermal images are inherently low contrast images and have low single to noise ratio. Moreover, they are amorphous in nature and no clear edges are seen. The difficulty lies in the detection of lower breast boundaries and inframammary folds. So, in general, the first attempt is made in improving the signal to noise ratio and contrast of the image which helps to extract the true regions of breast tissues. Then, asymmetry analysis of the normal and abnormal breast tissues is performed using different techniques. This work demonstrates the review of a few image processing methods or the development which are elaborated in the detection of breast cancer from thermal images.

The Deep Neural Networks have gained prominence in the biomedical domain, becoming the most commonly used networks after machine learning technology. Mammograms can be used to detect breast cancers with high precision with the help of Convolutional Neural Network (CNN) which is deep learning technology. An exhaustive labeled data is required to train the CNN from scratch. This can be overcome by deploying Generative Adversarial Network (GAN) which comparatively needs lesser training data during a mammogram screening. In the proposed study, the application of GANs in estimating breast density, high-resolution mammogram synthesis for clustered microcalcification analysis, effective segmentation of breast tumor, analysis of the shape of breast tumor, extraction of features and augmentation of the image during mammogram classification have been extensively reviewed.

Background: Black hole sign represents a novel imaging marker for predicting hematoma expansion (HE) in patients with intracerebral hemorrhage (ICH). Several previous studies have reported the accuracy of black hole sign in predicting HE, but the accuracy was variable. We performed a meta-analysis to systematically assess the accuracy of black hole sign in predicting HE in patients with ICH. Methods: A systematic search was performed to identify relevant English and Chinese articles (from inception to January 2019). All studies on the accuracy of black hole sign in predicting HE in patients with ICH were included. Pooled sensitivity, specificity, and positive and negative likelihood ratios were calculated. Pooling was conducted using the bivariate generalized linear mixed model. Forest plots and a summary receiver operator characteristic plot were generated. We used I² to test heterogeneity and investigated the source of heterogeneity by meta-regression. Publication bias was assessed by Deeks’ funnel plot asymmetry test. Results: A total of 6 studies with 1876 patients were included in this meta-analysis. The pooled sensitivity, specificity, and positive and negative likelihood ratios of black hole sign for predicting HE were 0.30, 0.93, 4.00 and 0.75, respectively. The area under the curve (AUC) was 0.83. The studies had substantial heterogeneity (I²=89.00%, 95% CI 78.00-100.00). Low risk of publication bias was detected. Conclusion: Black hole sign is a useful imaging marker with high specificity in predicting hematoma expansion in patients with intracerebral hemorrhage.

Background and Introduction: Collateral circulation is very crucial for the prognosis of stroke patients. Transcranial color-coded duplexsonography (TCCD) is used widely to evaluate the intracranial arterial blood flow. However, approximately 20% - 30% of the patients with cerebral infarction cannot be detected via TCCD due to the interruption of thickened temporal bones. We assessed the diagnostic efficacy of contrast-enhanced transcranial color-coded duplexsonography (CE-TCCD) in stroke patients with limited bone windows. Methods: CE-TCCD was applied to 70 patients (51 males and 19 females) who presented with ischemic symptoms, to detect the openness of the anterior communicating artery (ACoA) and posterior communicating artery (PCoA) of the Willis ring before Computed Tomography angiography (CTA) or Magnetic Resonance Angiography (MRA) examination. The results from CETCCD is used to compare with CTA/MRA result to verify the diagnostic efficacy. Results: Forty-one communicating artery openings were detected by CE-TCCD, among which 37 were PCoA and 4 were ACoA. Among the 70 patients, 23 of 70 patients indicated severe stenosis within intracranial and/or extracranial arteries. Eighteen out of the 23 patients showed collateral circulation, accounting for 78.3% (18/23). Moderate stenosis were 23 cases in total, in which 7 cases showed collateral circulation, accounting for 30.4% (7/23). Slight stenosis were 24 cases in total, none of which showed collateral circulation. Conclusion: In the stroke patients with limited bone windows, CE-TCCD can evaluate intracranial collateral circulation.

Background: Quality Assurance (QA) of Magnetic Resonance Imaging (MRI) system is an essential step to avoid problems in diagnosis when image quality is low. It is considered a patient safety issue. The accreditation program of the American College of Radiology (ACR) includes a standardized image quality measurement protocol. However, it has been shown that human testing by visual inspection is not objective and not reproducible. Methods: The overall goal of the present paper was to develop and implement a fully automated method for accurate image analysis to increase its objectivity. It can positively impact the QA process by decreasing the reaction time, improving repeatability, and by reducing operator dependency. The proposed QA procedures were applied to ten clinical MRI scanners. The performance of the automated procedure was assessed by comparing the test results with the decisions made by trained MRI technologists according to ACR guidelines. The p-value, correlation coefficient of the manual and automatic measurements were also computed using the Pearson test. Results and Conclusion: Compared to the manual process, the use of the proposed approach can significantly reduce the time requirements while maintaining consistency with manual measurements and furthermore, decrease the subjectivity of the results. Accordingly, a strong correlation was found and the corresponding p-value was much lower than the significance level of 0.05 indicating a good agreement between the two measurements.

Background: One of the early screening methods of breast cancer that is still used today is mammogram due to its low cost. Unfortunately, this low cost accompanied with low performance rate also. Methods: The low performance rate in mammograms is associated with low capability in determining the best region from which the features are extracted. Therefore, we offer an automatic method to detect the Region of Interest in the mammograms based on maximizing the area under receiver operating characteristic curve utilizing Genetic Algorithms. The proposed method had been applied to the MIAS mammographic database, which is widely used in literature. Its performance had been evaluated using four different classifiers; Support Vector Machine, Naïve Bayes, K-Nearest Neighbor and Logistic Regression classifiers. Results & Conclusion: The results showed good classification performances for all the classifiers used due to the rich information contained in the features extracted from the automatically selected Region of Interest.

Background: Alveolar bone height in the posterior maxillary region is very important and critical for dental implant planning and placement. Objectives: This study aimed to evaluate the anatomy of the maxillary sinus floor in relation to the alveolar crest and to determine variations in the vertical measurements between the maxillary sinus floor and the alveolar bone crest tip in the posterior edentulous maxilla with the use of cone beam computerized tomography. Methods: This analysis enrolled 234 retrospectively selected patients (123 males with mean age 52.95±11.74 (range 32-76 years) and 111 females with mean age 58.14±11.92 (range 32-75 years)) with edentulous posterior maxillary regions. The maxillary sinus floor was divided into three anatomical segments (anterior, median and posterior) in relation to the transverse palatine suture. The measurements were performed on 3D surface rendered volumetric images by using rotation and translation of the views. Landmarks for measurement were specified by using a cursor driven pointer. Vertical lines were marked on the cross-sectional images between the alveolar ridge and the deepest point of the maxillary sinus floor for each of the three regions. P < 0.05 was regarded as statistically significant. Results: The mean distance values between the sinus floor and the alveolar crest in the anterior, median and posterior regions were 8.74±3.97 mm, 5.37±3.23 mm and 7.06±3.28 mm, respectively. Measurements in the anterior region were found to be high in both total and gender groups compared to other regions. Also, subsinus alveolar bone heights decreased with increasing age in both genders in all three regions. Conclusion: This study emphasizes that the mean subsinus alveolar bone height is highest in the anterior segment of the edentulous posterior maxilla. These results may guide clinicians to make the decision of implant placement area and lead to less invasive alternative surgery methods for edentulous posterior segments.

Objectives: The brain functional network of autism spectrum disorders (ASDs) in the earlier stages of life has been almost unknown due to difficulties in obtaining a resting-state functional magnetic resonance imaging (rs-fMRI). This study aimed to perform rs-MRI under a sedated sleep state and reveal possible alterations in the brain functional network. Methods: Rs-fMRI was performed in a group of preschool children (aged 2–6 years, 53 with ASD, 63 as controls) under a sedated sleeping state. Based on graph theoretical analysis, global and local topological metrics were calculated to investigate alterations in brain functional networks. Besides, correlation analyses were conducted between the abnormal attribute values and the Childhood Autism Rating Scale (CARS) scores. Results: The graph theoretical analysis showed that the nodal degree of the right medial frontal gyrus and the nodal efficiency of the right lingual gyrus in the ASD group were higher than those in the control group (P<0.05). There was a statistically significant positive correlation (R=0.318, P<0.05) between the right midfrontal gyrus nodal degree values and CARS scores in the ASD patients. Conclusion: Alterations of some nodal attributes in the brain network occurred in preschool autistic children which could serve as potential imaging biomarkers for evaluating ASD in earlier stages.

Background: Developmental venous anomalies (DVAs) can be determined on magnetic resonance imaging (MRI), and they may be associated with multiple sclerosis (MS) lesions. Purpose: The objective was to evaluate the MRI findings of DVAs in the brain, to compare the prevalence of them between MS patients and control subjects, and to investigate the correlation of DVA-associated fluid-attenuated inversion recovery (FLAIR) hyperintensities and MRI-derived parameters between MS patients and control subjects having DVA. Methods: Total 160 patients with a mean age of 45 ± 16 years who underwent multiparametric MRI including susceptibility-weighted imaging (SWI), diffusion-weighted imaging, 3D FLAIR, and contrast-enhanced imaging were included in this retrospective study. First, the presence of DVA was compared between the MS and control groups using the Chi-square test. Then, among the subjects having DVA, age, gender, and MRI-derived parameters such as the signal increase of DVA on FLAIR, location, and drainage of DVA were compared between the MS and control groups using Chi-square test. Results: The presence of DVA did not differ between the MS and control groups (P = 0.828). Signal increase around DVA on FLAIR (P = 0.03) and the age of less than 45 years demonstrated a significant correlation with MS group (P = 0.022). Conclusion: In our study, DVAs were effectively detected using SWI and 3D contrast-enhanced T1-weighted imaging on MRI. The signal increase of DVA was better revealed on 3D FLAIR on MRI, and it was the only significant MRI-derived parameter in patients with MS.