Current Medical Imaging - Volume 17, Issue 3, 2021

Objective: The aim was to study image fusion-based cancer classification models used to diagnose cancer and assess medical problems in earlier stages that help doctors or health care professionals to make the treatment plan accordingly. Methods: In this work, a novel image fusion method based on Curvelet transform is developed. CT and PET scan images of benign type tumors were fused together using the proposed fusion algorithm and the same way, MRI and PET scan images of malignant type tumors were fused together to achieve the combined benefits of individual imaging techniques. Then, the marker-controlled watershed algorithm was applied on fused images to segment cancer affected area. The various color features, shape features and texture-based features were extracted from the segmented image. Following this, a data set was formed with various features, given as input to different classifiers namely neural network classifier, Random forest classifier, and K-NN classifier to determine the nature of cancer. The results of the classifier showed normal, benign or malignant category of cancer. Results: The performance of the proposed fusion algorithm was compared with the existing fusion techniques based on the parameters PSNR, SSIM, Entropy, Mean and Standard Deviation. Curvelet transform based fusion method performs better than already existing methods in terms of five parameters. The performances of the classifiers were evaluated using three parameters: accuracy, sensitivity, and specificity. The K-NN Classifier performed better compared to the other two classifiers and it provided an overall accuracy of 94%, sensitivity of 88% and specificity of 84%. Conclusion: The proposed Curvelet transform based image fusion method combined with the KNN classifier provides better results compared to other two classifiers when two input images were used individually.

Background: Early diagnosis of a brain tumor may increase life expectancy. Magnetic resonance imaging (MRI) accompanied by several segmentation algorithms is preferred as a reliable method for assessment. The availability of high-dimensional medical image data during diagnosis places a heavy computational burden and a suitable pre-processing step is required for lower- dimensional representation. The storage requirement and complexity of image data are also a concern. To address this concern, the random projection technique (RPT) is widely used as a multivariate approach for data reduction. Aim: This study mainly focuses on T1-weighted MRI image clustering for brain tumor segmentation with dimension reduction by using the conventional principal component analysis (PCA) and RPT. Methods: Two clustering algorithms, K-means and fuzzy c-means (FCM) were used for brain tumor detection. The primary study objective was to present a comparison of the two clustering methods between MRI images subjected to PCA and RPT. In addition to the original dimension of 512 x 512, three other image sizes, 256 x 256, 128 x 128, and 64 x 64, were used to determine the effect of the methods. Results: In terms of average reconstruction, Euclidean distance, and segmentation distance errors, the RPT produced better results than the PCA method for all the clustered images from clustering techniques. Conclusion: According to the values of performance metrics, RPT supported fuzzy c-means in achieving the best clustering performance and provided significant results for each new size of the MRI images.

Background: The learning-based algorithms provide an ability to automatically estimate and refine GM, WM and CSF. The ground truth manually achieved from the 3T MR image may not be accurate and reliable with poor image intensity contrast. It will seriously influence the classification performance because the supervised learning-based algorithms extremely rely on the ground truth. Recently, the 7T MR images brings about the excellent image intensity contrast, while Structured Random Forest (SRF) performs the pixel-level classification and achieves structural and contextual information in images. Materials and Methods: In this paper, a automatic segmentation algorithm is proposed based on ground truth achieved by the corresponding 7T subjects for segmenting the 3T&1.5T brain tissues using SRF classifiers. Through taking advantage of the 7T brain MR images, we can achieve the highly accuracy and reliable ground truth and then implement the training of SRF classifiers. Our proposed algorithm effectively integrates the T1-weighed images along with the probability maps to train the SRF classifiers for brain tissue segmentation. Results: Specifically, for the mean Dice ratio of all 10 subjects, the proposed method achieved 95.14%±0.9%, 90.17%±1.83%, and 81.96%±4.32% for WM, GM, and CSF. With the experiment results, the proposed algorithm can achieve better performances than other automatic segmentation methods. Further experiments are performed on the 200 3T&1.5T brain MR images of ADNI dataset and our proposed method shows promised performances. & Conclusion: The authors have developed and validated a novel fully automated method for 3T brain MR image segmentation.

Background: Cluster analysis is a robust tool for exploring the underlining structures in data and grouping them with similar objects. In the researches of Functional Magnetic Resonance Imaging (fMRI), clustering approaches attempt to classify voxels depending on their time-course signals into a similar hemodynamic response over time. Objective: In this work, a novel unsupervised learning approach is proposed that relies on using Enhanced Neural Gas (ENG) algorithm in fMRI data for comparison with Neural Gas (NG) method, which has yet to be utilized for that aim. The ENG algorithm depends on the network structure of the NG and concentrates on an efficacious prototype-based clustering approach. Methods: The comparison outcomes on real auditory fMRI data show that ENG outperforms the NG and statistical parametric mapping (SPM) methods due to its insensitivity to the ordering of input data sequence, various initializations for selecting a set of neurons, and the existence of extreme values (outliers). The findings also prove its capability to discover the exact and real values of a cluster number effectively. Results: Four validation indices are applied to evaluate the performance of the proposed ENG method with fMRI and compare it with a clustering approach (NG algorithm) and model-based data analysis (SPM). These validation indices include the Jaccard Coefficient (JC), Receiver Operating Characteristic (ROC), Minimum Description Length (MDL) value, and Minimum Square Error (MSE). Conclusion: The ENG technique can tackle all shortcomings of NG application with fMRI data, identify the active area of the human brain effectively, and determine the locations of the cluster center based on the MDL value during the process of network learning.

Background: The two of the most common indications for magnetic resonance (MR) imaging of the cervical spine include acute spine trauma and degenerative disease. Objective: We aimed to correlate the measurements of the free area of the spinal canal (FASC), a new approach to the cervical spinal canal compromise, with the Torg´s ratio quantification of the cervical spine. Methods: A cross-sectional study including 50 cervical-spine MR evaluations of patients with acute cervical trauma or degenerative disease was performed. We used multivariate analysis of covariance (MANCOVA) to identify the type of lesion, intervertebral level and gender differences between FASC and Torg´s ratio quantification of the cervical spine; age was the controlled covariate. Correlates between FASC and Torg´s ratio were obtained at each intervertebral level. Results: There was a non-significant interaction between the type of lesion, gender and intervertebral levels between FASC and Torg´s ratio measurements, F (8, 456) 0.260, p = .978; Wilks' Lambda 0.991; with a small effect size (partial η² = .005). Among the main effects, only the gender was statistically significant: F (2, 228) = 3.682, p = .027. The age (controlled covariate) was non-significantly related to FASC and Torg´s ratio quantification: F (2, 228) = .098, p = .907. The Pearson´s correlation coefficient depicted a poor, non-significant agreement between FASC and Torg´s ratio. Conclusion: FASC provides an integrative evaluation of the cervical spinal canal compromise in acute, cervical spine trauma and degenerative disease. Further observations and correlation with specific neurological symptoms, surgical findings and clinical outcomes are necessary to assess the usefulness of FASC in clinical settings.

Purpose: Both CT and PET radiomics is considered as a potential prognostic biomarker in head and neck cancer. This study investigates the value of fused pre-treatment functional imaging (18F-FDG PET/CT) radiomics for modeling of local recurrence of head and neck cancers. Materials and Methods: Firstly, 298 patients have been divided into a training set (n = 192) and verification set (n = 106). Secondly, PETs and CTs are fused based on wavelet transform. Thirdly, radiomics features are extracted from the 3D tumor area from PETCT fusion. The training set is used to select the features reduction and predict local recurrence, and the random forest prediction models combining radiomics and clinical variables are constructed. Finally, the ROC curve and KM analysis are used to evaluate the prediction efficiency of the model on the validation set. Results: Two PET/CT fusion radiomics features and three clinic parameters are extracted to construct the radiomics model. AUC value in the verification set 0.70 is better than no fused sets 0.69. The accuracy of 0.66 is not the highest value (0.67). Either consistency index CI 0.70 (from 0.67 to 0.70) or the p-value 0.025 (from 0.03 to 0.025) get the best result in all four models. Conclusion: The radiomics model based on the fusion of PETCT is better than the model based on PET or CT alone in predicting local recurrence, the inclusion of clinical parameters may result in more accurate predictions, which has certain guiding significance for the development of personalized, precise treatment scheme.

Background: Evaluation of the maxillary sinus anatomy prior to sinus lift procedures is important to avoid surgical complications due to the close anatomical relationship between the posterior maxillary region and the maxillary sinüs. Introduction: In order to avoid traumatizing the posterior superior alveolar artery and to prevent perioperative bleeding, locating the exact position of the artery is imperative before the surgical procedures. Methods: 150 CBCT scans were evaluated. The distance from the inferior border of the PSAA to the alveolar crest (the vertical line from the artery to the crest) and floor of the maxillary sinus, the distance from PSAA to medial sinus wall, nasal septum, zygomatic arch, position, the distance from the floor of maxillary sinüs to the alveolar crest and diameter of the PSAA were assessed. Locations of the artery were classified. Results: The artery diameters were mostly ≥ 1 mm. The artery was mostly intraosseous (59.7%), 21.7% was superficial and only 18.7% was intra-sinuscular. Conclusion: The location of PSAA is intraosseous in most patients. The artery diameters were mostly ≥ 1 mm and we can say that increasing the size also increases the risk of complications.

Background: Intraperitoneal fluid accumulations are a common matter in current clinical practice, being encountered by most medical and surgical fields. Objective: To assess ascites fluid with attenuation values in the form of Hounsfield units (HU) in order to determine a non-invasive differentiation criterion for the diagnosis of intraperitoneal collections. Methods: Sixty patients with known intra-peritoneal collections who underwent computed tomography (CT) for reasons such as tumor staging, post-surgical follow-up or other indications, were retrospectively included in this study. All subjects had a final pathological analysis of the fluid collections. Two radiologists measured the attenuation values for each collection. The averaged values were used for comparing benign and malignancy-related ascites (MRA), bland and hemorrhagic ascites and infected and noninfected fluid collections by consuming the Mann–Whitney U test. Also, the receiver operating characteristic analysis was performed for the statistically significant results (P<0.05), and the area under the curve (AUC) was calculated. Results: Attenuation values could differentiate between benign and MRA (P=0.04; AUC=0.656; sensitivity, 65.52%; specificity, 71.43%) but failed to distinguish between bland ascites and ascites with hemorrhagic component (P=0.85), and between infected and noninfected fluid collections (P=0.47). Conclusion: Although the results are statistically significant, the substrate of differentiation between benign and MRA ascites cannot be clearly stated. As being the first study to investigate this issue, it opens the way for other researches in the field to determine the dynamics of imaging quantitative measurements according to the fluid’s pathological features.

Background: While iodine-enhanced computed tomography has been studied, detailed information on gadolinium-enhanced magnetic resonance imaging has not been reported. Objective: We evaluated the effects of different gadolinium contrast agent (Gd-CA) factors on the enhancement of aortic magnetic resonance angiography (MRA) using computer simulation. Methods: We developed computer-simulation software that combines pharmacokinetic models and tables; it converts the blood concentration of particular Gd-CAs into the signal intensity (SI). We simulated aortic time-intensity curves (TIC) in our MRA study and compared the effect of the Gd-- CA volume, injection rate, and of different Gd-CAs on the TIC. Results: An increase in the Gd-CA volume from 14.0 to 28.0 ml increased maximal aortic intensity 1.11 times. Changing the injection rate from 1.0 to 2.8 ml/s increased it 1.10 times. The maximal SI of gadoterate-meglumine and gadobutrol was 1.03 and 1.01 times, respectively, that of gadoteridol. Conclusion: In our computer-simulated MRA study, different Gd-CA factors resulted in no significant difference in the maximal aortic SI.

Purpose: The aim of this study is to evaluate the accuracy and dosimetric effects for auto- segmentation of the CTV for GO in CT images based on FCN. Methods: An FCN-8s network architecture for auto-segmentation was built based on Caffe. CT images of 121 patients with GO who have received radiotherapy at the West China Hospital of Sichuan University were randomly selected for training and testing. Two methods were used to segment the CTV of GO: treating the two-part CTV as a whole anatomical region or considering the two parts of CTV as two independent regions. Dice Similarity Coefficient (DSC) and Hausdorff Distance (HD) were used as evaluation criteria. The auto-segmented contours were imported into the original treatment plan to analyse the dosimetric characteristics. Results: The similarity comparison between manual contours and auto-segmental contours showed an average DSC value of up to 0.83. The max HD values for segmenting two parts of CTV separately was a little bit smaller than treating CTV with one label (8.23±2.80 vs. 9.03±2.78). The dosimetric comparison between manual contours and auto-segmental contours showed there was a significant difference (p<0.05) with the lack of dose for auto-segmental CTV. Conclusion: Based on deep learning architecture, the automatic segmentation model for small target areas can carry out auto contouring tasks well. Treating separate parts of one target as different anatomic regions can help to improve the auto-contouring quality. The dosimetric evaluation can provide us with different perspectives for further exploration of automatic sketching tools.

Background: The mental foramen (MF) is an important anatomical landmark on the mandible. MFs may occur singly or with extra foramina (accessory mental foramen (AMF)). Objective: The aim of the present study was to discuss the importance of recognition and classification of AMFs. Methods: This study assessed CBCT images of 593 patients (208 male, 385 female) and proposed an AMF classification scheme based on three different characteristics: (1) location of the AMF with respect to the MF; (2) relationship of AMFs with tooth apices; and (3) origin of the AMF. The chi-squared test was used to compare the qualitative data as well as descriptive statistical methods when the study data were evaluated. Significance was assessed at the p < 0.05 level. Results: A total of 80 AMFs were found in 71 patients. Six cases involved bilateral AMFs, whereas three cases showed double AMFs. The most common AMF location was the region posterior of the MF. Notably, all the AMFs detected in line with the mesial half of the first molar were present in female patients. Based on AMF origins, 44.5%, 48.6%, and 6.7% of the AMFs were categorized as type I, type II, and a new category identified in this study, type III. Conclusion: This study revealed that some foramina locations were not considered in the previous classification. Recognition and identification of these foramina might be critical to prevent possible complications. The proposed classification scheme may facilitate this objective and can be a new focus for future studies.

Background: Obesity is a significant risk factor for cardiovascular (CV) disease. Abdominal fat is composed of abdominal subcutaneous fat and intra-abdominal (visceral) fat. Computed tomography (CT) is considered one of the most accurate and reliable methods for assessing abdominal fat. Introduction: The present study was based on evaluating abdominal fat by computed tomography and the determination of association between CT obtained abdominal fat volumes, anthropometric indices, and lipid profile. Methods: The prospective study was carried out on 120 subjects referred to the Radiology department for a CT scan. Non - contrast CT scan was performed with 5 mm slice thickness. Abdominal fat volumes were recorded by using CT attenuation values (- 250 to -50 HU). The section was selected at the level of the umbilicus (L4-L5). Intra-abdominal fat and subcutaneous fat volumes were calculated. Body Mass Index (BMI) and lipid profile were recorded for each subject. A comparative study of the CT values, BMI, and lipid profile was undertaken. Results: In the present study, by comparing the anthropometric parameters, CT findings, and lipid profile and blood parameters of the obese and non-obese groups by sex revealed significant sex differences in all the parameters under study. It was also found that the obese male and female groups showed a high prevalence of diabetes, Non-Alcoholic fatty liver disease (NAFLD), and hypertension than non-obese groups. This finding also adds to the chances of getting cardiovascular diseases, specifically in obese individuals. The results found that in obese males and females the abdominal fat-related parameters Visceral fatty acid (VFA) and subcutaneous fatty acid (SFA) showed highly significant relation to anthropometric parameters like BMI, waist circumference (WC) and waist/hip (W/H) ratio on the other hand blood parameters high-density lipoprotein (HDL), low-density lipoprotein (LDL), very-low-density lipoprotein (VLDL), total cholesterol and triglycerides to some extent have a significant relation to abdominal fat-related parameters. In non-obese groups, by studying the influence of anthropometric parameters on abdominal fat-related parameters, it was revealed that WC was strongly affected by the VFA in both sexes. In obese females, more fat was accumulated in the VFA and SFA and for obese males in SFA and for non-obese males in total fatty acid (TFA). Conclusion: Computed tomography assessed visceral fat area remains the most sensitive independent predictor of cardiovascular risk.

Background: The incidence rate of renal disease is high, which can cause end-stage renal disease. Ultrasound is a commonly used imaging method, including conventional ultrasound, color ultrasound, elastography, etc. Machine learning is a potential method which has been widely used in clinical practices. Objective: To compare the diagnostic performance of different ultrasonic image measurement parameters for kidney diseases, and to compare different machine learning methods with the human- reading method. Methods: Ninety-four patients with pathologically diagnosed renal diseases and 109 normal controls were included in this study. The patients were examined by conventional ultrasound, color ultrasound and shear wave elasticity, respectively. Ultrasonic data were analyzed by Support vector machine (SVM), random forest (RF), K-nearest neighbor (KNN) and artificial neural network (ANN), respectively, and compared with the human-reading method. Results: Only ultrasound elastography data have a diagnostic value for renal diseases. The accuracy of SVM, RF, KNN and ANN methods is 80.98%, 80.32%, 78.03% and 79.67%, respectively, while the accuracy of human-reading is 78.33%. In the data of machine learning ultrasound elastography, the elastic hardness parameters of the renal cortex are most important. Conclusion: Ultrasound elastography is of the highest diagnostic value in machine learning for nephropathy, the diagnostic efficiency of the machine learning method is slightly higher than that of the human-reading method, and the diagnostic ability of the SVM method is higher than other methods.

Background: Calcium-suppressed (CaSupp) technique involving spectral-based images has been used to observe bone marrow edema by removing calcium components from the image. Objective: This study aimed to evaluate the knee articular cartilage using the CaSupp technique in dual-layer detector computed tomography (DLCT). Methods: Twenty-eight healthy participants and two patients with osteoarthritis were enrolled, who underwent DLCT and magnetic resonance imaging (MRI) examination. CaSupp images were reconstructed from spectral-based images using a calcium suppression algorithm and were overlaid with conventional CT images for visual evaluation. The morphology of the knee cartilage was evaluated, and the thickness of the articular cartilage was measured on sagittal proton density-weighted and CaSupp images in the patellofemoral compartment. Results: No abnormal signal or density, cartilage defect, and subjacent bone ulceration were observed in the lateral and medial femorotibial compartments and the patellofemoral compartment on MRI images and CaSupp images for the 48 normal knee joints. CaSupp images could clearly identify cartilage thinning, defect, subjacent bone marrow edema, and edema of the infrapatellar fat pad in the same way as MRI images in the three knee joints with osteoarthritis. A significant difference was found in the mean thickness of the patellar cartilage between MRI images and CaSupp images, while the femoral cartilage presented no significant difference in thickness between MRI images and CaSupp images in all 48 knee joints. Conclusion: The present study demonstrated that CaSupp images could effectively be used to perform the visual and quantitative assessment of knee cartilage.

Cystic lymphangioma presenting with multiple bone lesions in an adult patient is a rare occurrence, with a limited number of reported cases in the literature. In this case report, we describe a 32-year old female patient with chronic neck and pelvic pain, and multiple lytic bone lesions on radiological imaging, which were eventually discovered to originate from cystic hygroma and widespread bone lymphangiomas that were present for more than 10 years. It should be kept in mind that there may be multiple benign causes of differential diagnosis in patients presenting with findings suggestive of diffuse bone metastasis. Misdiagnosis may cause the patient to receive unnecessary treatments, especially radiotherapy. In this case, we reached the diagnosis of benign disease, diffuse bone lymphangiomatosis. For this purpose, we also examine the long and stable medical history of the patient with the findings of BT, ultrasound, and bone scintigraphy. We think that as long as there are no stable and serious results ofclinical and radiological findings of the patient, the patient should be approached with medical follow-up without treatment.

Background: Endometriosis is a pathological state in which the endometrial glands and stroma are originated at sites other than the uterine cavity. Endometriotic cystic lesions are principally positioned in ovaries but, infrequently, can also be encountered in the myometrium. It is a rare manifestation which poses a multitude of differential diagnoses and therapeutic options. Case Study: A 27-year-old female patient walked into hospital complaining of severe abdominal pain since 2 days with a medical history of dysmenorrhoea since one and a half years. On examination, the abdomen was found to be soft with the existence of lower abdominal tenderness. An Ultrasonography revealed a myometrial posterior wall cyst of size 4 x 4 cm with thick wall and spotted internal echoes favoring a differential diagnosis of either an endometriotic cyst or rudimentary horn. Both ovaries were found normal. The patient was taken for laparoscopy in which a bulge was seen on the posterior uterine wall. It was punctured to expose a myometrial cyst with dark chocolate colored fluid collection. The intact cyst was enucleated and sent for histopathology, which confirmed the diagnosis of the endometriotic cyst. Upon the literature survey, it was found that only two such cases were available in the literature on internet which described a similar pathology. Conclusion: Endometrioma should be considered as a probable pathology whenever myometrial cysts are encountered. A focused transvaginal examination can be very accommodating as a diagnostic modality to set up proper management. This case should not only make the health careproviders rethink the etiopathogenesis of endometriosis and debate the credibility of the retrograde menstruation hypothesis but also encourage to accurately investigate any lesion anywhere in the body, which looks like an endometriotic chocolate cyst. In this case, it is evident that a myometrial spot is rare but not impossible. These rare occurrences shall direct to diversify the perception of this pathology.