Current Medical Imaging - Volume 14, Issue 4, 2018

Background: This paper reviews the recent techniques employed to process brain tumor images from Magnetic Resonance (MR) images. Automation in tumor and its sub-compartment segmentation from MR images is rising and necessary in rapid diagnosis and planning of human brain treatment. Discussion: The success of fine tumor segmentation depends on an entire pipeline of the processes such as preprocessing, segmentation, feature extraction and classification tools. Conclusion: Hence, the review aims to present a collective overview of existing techniques and explore the metrics which are used for various performance analysis for image processing techniques.

Background: Tuberculosis (TB) has become a global pandemic, and its eradication requires efficient screening methods, diagnostic tests, and effective drugs. Artificial intelligencebased Computer-aided Diagnostic (CADx) systems are purported to play a significant role in the mass screening of TB. Discussion: The research on the development of CADx systems started four decades ago, and a large number of CADx systems have been developed till date. However, no independent survey focussing on the advancements in these systems has been presented. This paper fills this gap by consolidating the advancements and presents a comprehensive survey of CADx systems for TB detection developed till date with a focus on their underlying principles. It also discusses a practical model using which CADx systems can be used for screening TB in places where medical facilities and experts are not adequately available. Conclusion: The paper also presents an overview of the current state of deep learning-based CADx systems. The development of these systems will remain in focus in the near future and will improve state-of-the-art performance in various medical domains.

Background: The paper proposes a new method for improving the quality of ultrasound medical images using a two stage wavelet based denoising approach. The first stage utilizes an adaptive wavelet shrinkage function with a novel interscale measure based variance estimate and a new subband adaptive threshold for reducing the speckle present in ultrasound images. Methods: The performance of the first stage is further improved in the second by employing an error measure estimate. Thus the proposed denoising approach is optimized in the mean square error sense and results in improved noise removal as well as in the edge preservation of the reconstructed image. It also improves the visual quality of the ultrasound images. Next, quantitative evaluation is also done with performance metrics like Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), Structural Similarity Index Measure (SSIM), Equivalent Number of Looks (ENL) and Edge preservation Index (EPI). Results: The performance of the proposed approach is measured and compared with the existing approaches and found to be better in improving the quality of ultrasound images.

Objective: Positron Emission Tomography (PET) with F-18-Fluoro-deoxy-glucose (FDG) emerged as a prognostic tool to predict treatment outcome in Hodgkin Lymphoma (HL). Moreover, a FDG-PET adapted strategy is currently assessed in clinical trial to minimize the toxic effect while maintaining the efficacy of treatment in HL. Purpose was to analyze the quantitative parameters to support the prognostic role of FDG-PET today based on the semi-quantitative Deauville 5-point Scale (D5-PS). Methods: This retrospective study included 53 patients diagnosed with advanced-stage HL between 2009 and 2014, enrolled in the PET response-adapted clinical trial HD 0607. FDG-PET was performed at baseline (PET0) and after two cycles of chemotherapy (PET2). Analysis was based on two main approaches: on the single lesion presenting the higher FDG uptake and on the five hottest lesions. Different metabolic parameters were analyzed. Patients were classified into responders and nonresponders. Optimal cut-offs were derived from Receiver Operating Characteristic (ROC) curves. Results were correlated with Progression Free Survival (PFS) using Kaplan-Meier. Results: A 71% threshold in SUVmax reduction (ΔSUVmax) was found to be the best cutoff quantitative parameter able to identify responders vs. non-responders, also with a multivariate analysis, joining clinical data with SUVmax. After a mean follow-up of 34.2 months (95% CI, 26.2 to 39.1), the median PFS for non-responders was 8 months vs. not reached for responders. These results were superimposable to that obtained by an independent group of reviewers using the D5-PS. Conclusion: Semi-quantitative analysis by ΔSUVmax outperforms qualitative assessment by D5-PS in predicting treatment outcome in ABVD-treated advanced-stage HL.

Aim: The aim of our study was to investigate the etiology of patients presenting with double vision due to sixth cranial nerve palsy, and evaluate the relationships between the sixth cranial nerve and vascular structures, with Constructive Interference in a Steady-State (CISS) sequence. Materials and Methods: For this study, we evaluated the brain and orbital Magnetic Resonance Imaging (MRI) of 26 patients retrospectively, who were diagnosed with binocular diplopia associated with sixth nerve palsy between 2011 and 2016. The MRI images were assessed for those pathologies that can cause diplopia. Additionally, the cisternal segment of the sixth cranial nerve was assessed with CISS sequences for possible vascular contact or compression. Results: Nine of the 26 patients were over 50 years old, with medical histories of diabetes mellitus, hypertension, and hyperlipidemia/hypercholesterolemia. Microvascular angiopathy was considered for diagnosis in those patients with sixth cranial nerve palsy, while the medical history was normal in 17 of these 26 patients. Brain and orbital MRIs were used to detect any cavernous sinus pathology and/or dural sinus vein thrombosis; however, the MRIs were normal in 9 of 17 patients with sixth cranial nerve palsy. Therefore, vascular compression and/or contact were detected in these patients using the CISS sequence, because diplopia can be caused by vascular contact or compression of the sixth cranial nerve. Conclusion: Patients with complaints of double vision and normal brain and orbital MRIs should be evaluated using CISS sequences to show the relationships between the sixth cranial nerve and vascular structures.

Background: Segmentation of nuclear cardiac images is used as an influential tool in the characterization of heart ailments. Methods: The processing techniques confirmed in this study were contrast enhancement, noise removal, skeletonization and segmentation. The images were preserved by means of TIFF to maintain their quality. Then, diverse segmentation techniques were used to segment the heart tissues. Result: The computation rate tested used matching rate and corresponding percentage. The results were 99.1+10.39 (p > 0.05). Conclusion: This method offered precise identification of the heart tissue with significantly high exactness percentage.

Objective: The drive of this research activity is to report a colorization method to enhance the visualization, cell characterization and interpretation of brain Magnetic Resonance (MR) images, indicative of the underlying physical density of bio tissues. Background: In daily health care environments, medical scanners are being used to generate gray scale images of anatomical structures. Method: Severity of diseases is determined from this luminance component only. Irrefutably, if this high dimensional gray scale, medical data is visualized in colored versions, the definitive and more accurate the pathological assessment process will be. A number of methods have been reported to represent brain Magnetic Resonance Images (MRI) data in color with the cost of computational complexity. In this research work, an efficient method of colorization using frequencies from visible range of color spectrum, has been proposed to embody the variations and sensitivity of the brain MRI images. Conclusion: Side by side, visual comparison based on multiple MRI sequences of identical subjects, by domain experts, proved adequate success and fruitfulness of the story.

Background: This paper proposes an improved Tetrolet algorithm of image denoising. Tetrolets, as the newest members of the post wavelets, represent the most concentrated image energy and provide the best sparseness. However, Tetrolet denoising is usually accompanied with occurrence of adverse blocking artifacts. Methods: To resolve this problem, this paper introduces a recursive translation method, which can effectively eliminate the fuzzy blocks in the image and use the matching window threshold function, thus improving the original algorithm. In the simulation experiment, a medical image, in particular, nuclear magnetic resonance image of patient's head, is denoised by the proposed algorithm and compared with those provided by alternative denoising methods, such as wavelet, improved wavelet, improved Contourlet, and traditional Tetrolet methods. Results: The comparative analysis results proved the superiority of the proposed improved method by the peak signal-to-noise ratios. Discussion: It can effectively eliminate the noise, retain the local characteristics of the original image, and keep the most clear texture of the processed medical images.

Purpose: Breast Imaging Reporting and Data System (BI-RADS) has many descriptors whose combined evaluation leads to the differentiation of benignancy and malignancy of breast lesions. The purpose of this study was to determine the best combination. Materials and Methods: Institutional Review Board approval was waived for this retrospective analysis. Five radiologists evaluated BI-RADS descriptors of 109 breast lesions of 71 consecutive patients (mean age 45.3 years, range 22-70 years) and 5 descriptors (shape, orientation, margin, echo pattern, posterior features) were collected as binary type data. Using SPSS package, for each radiologist, cross tabulation analysis was conducted between each of the descriptors and pathologic diagnosis. Subsequently, logistic regression analysis was conducted for variable combinations of descriptors, and the best model was selected. Finally, the same step was used for the dataset made by 5 radiologists together. Results: Sensitivity and specificity of 5 descriptors were evaluated by 5 radiologists. Shape showed 76% of sensitivity (56-92%) and 52% of specificity (39-64%). Orientation showed 51% of sensitivity (31-72%) and 83% of specificity (73-94%). Margin showed 83% of sensitivity (72-90%) and 53% of specificity (36-71%). Echo pattern showed 87% of sensitivity (54-100%) and 22% of specificity (3-60%). Posterior features showed 34% of sensitivity (21-46%) and 86% of specificity (71- 97%). As a result, relatively high sensitive descriptor was echo pattern and relatively high specific descriptor was posterior feature. Subsequently, regression model for the large dataset output by 5 radiologists together were made by ≥ 2 descriptors and the best model that showed the highest PCC (Percentage Correctly Classified) was shape + orientation (72.1, range 67.9-72.5) in combination of two. In combination of three, the best model was shape + orientation + posterior feature (72.5, range 71.6-76.1). Shape + orientation+ posterior feature + echo pattern (73.8, range 71.6-77.1) was the best model for combination of four and in combination of five, the PCC was 72.7(range 72.5-76.1). Conclusion: The highest PCC in each of the combinations showed no significant difference. Therefore, combination of two descriptors was sufficient for regression model and the combination of shape and orientation was the most significant descriptors in distinction of benign and malignant breast masses.

Aim: The aim of this study was to evaluate changes in the subtrochanteric femoral cortical thickness and the medullary canal diameter according to age and gender. Materials and Methods: In total, 840 radiographs of 420 hips of patients aged between 10 and 80 (210 males and 210 females) were included in this study. The two groups (males and females) were divided into three subgroups according to age, forming six groups in total. Results: Males showed significant differences in the thickness of the lateral, anterior, and posterior cortex with age, but no age-related differences were noted in the diameters of the mediolateral medullary canal and anteroposterior medullary canal. Females showed significant differences in the thickness of medial, lateral, and anterior cortex with age and significant age-related differences in the diameters of mediolateral medullary canal and anteroposterior medullary canal. Conclusion: The anterior, posterior, medial, and lateral cortex thicknesses of the femur and the anteroposterior and mediolateral widths of the medullary canal independently show different developmental characteristics according to age and gender.

Purpose: to assess the reproducibility of the volumetric assessment of the inner ear using three dimensional (3D) reconstruction of the high resolution MR-Imaging (HR-MRI) sequences. Materials and Methods: The study was retrospectively performed on 50 patients referred for MRI due to different disorders of the inner ear. HR-MRI was performed using a T2-weighted SPACE sequence. 3D-reconstructions were performed using Advantage Workstation. The volumes of both inner ears together, each side alone, cochlea & vestibular system were assessed. Volume measurements were performed twice with 1 year between both measurements. The agreement between both results was tested using intra-class correlation (ICC) test. Results: For the first assessment the mean volume of both sides together was 1.01 cm3 +/-0.13, of each side alone 0.5 cm3 +/-0.07, of the cochlea 0.17 cm3 +/-0.024 & of the vestibular system 0.33 cm3 +/-0.06. For the second assessment the mean volume of both sides together was 0.99 cm3 +/- 0.11, of each side alone 0.5 cm3 +/-0.06, of the cochlea 0.17 cm3 +/-0.025 & of the vestibular system 0.33 cm3 +/-0.05. A statistically significant (p<0.0001) excellent agreement for volumetric assessment of both sides together (ICC-Coefficient=0.82) and of the cochlea (ICC-Coefficient=0.88) and substantial agreement for each side alone (ICC-Coefficient=0.79) and of the vestibular system (ICC-Coefficient= 0.7) were noted. Conclusion: Volumetric assessment of the inner ear using 3D reconstruction of HR-MRI is a reproducible method with statistically significant excellent agreement for volumetric assessment of both sides together and cochlea and substantial agreement for each side alone and vestibular system.

Background: More data needed about the variations of the skull base and relationships between the structures which form the skull base and paranasal sinuses to minimize the complications during endoscopic sinus surgery. The aim of this study was to investigate the relationships between Cribriform Plate (CP) dimensions, frontal sinus ostium sizes and frontal sinus volumes. Materials and Methods: We measured the depth and width of the CP, the larger anteroposterior and side-to-side diameter of the frontal sinus ostia and the frontal sinus volumes on Computerized Tomography (CT) images of 120 adult subjects. Results: The mean depth of CP was 8.43 ± 1.62 mm at the right side and 8.48 ± 1.66 at the left side. The mean frontal sinus volumes were 4.83 ± 3.58 cm3 at the right side and 5.09 ± 3.74 cm3 at the left side. There were significantly larger frontal sinus volumes in males (male/female median values: 4.50/3.54 cm3 ; p=0.03). There were significant positive correlations between CP depth, larger side-to-side and anteroposterior diameter of the frontal sinus ostium and frontal sinus volume (p<0.01). There was a significant negative correlation between larger side-to-side diameter of the frontal sinus ostium and CP width (p:0.041). The presence of ipsilateral hyperpneumatized frontal sinus and deep CP should be taken into consideration to avoid iatrogenic injury. In light of our results, surgeons should remember that the frontal recess may be narrow if the CP is wider than 4.5 mm in preoperative evaluation.

Background: Perfusion Weighted Imaging (PWI) has been used to differentiate the solitary metastases from high-grade glioma. However, the CBV value obtained from classical algorithm can be affected by a number of factors. Tissue Similarity Map (TSM) is a new algorithm depending solely on the signal intensity time course but does not require the use of Concentration Time Curve (CTC), which is different from the classical rCBV algorithm. Objective: The purpose of this study is to investigate the diagnostic utility of TSM based rCBV values in differentiating between high-grade gliomas and solitary metastases. Methods: The preoperative MR PWI studies of 30 patients with a solitary cerebral neoplasm were retrospectively analyzed. All PWI data were calculated for TSM algorithm with MATLAB 7.6 (Math Works, Natick, MA). The TSM based rCBV (rCBVTSM) were measured in the parenchyma of tumor and peritumoral regions. All specimens were histopathologically categorized as high-grade gliomas or metastases and were correlated with corresponding rCBVTSM values. Results: All PWI data could be post-processed by using the TSM post-processing software. The differences in the rCBVTSM-T between high-grade gliomas and metastases were not statistically significant (p=0.299). While the rCBVTSM-P of high-grade gliomas was higher than that of metastases (p=0.040). ROC curve analysis showed significant AUC of rCBVTSM-P for differentiating highgrade gliomas from metastases (p=0.020). The rCBVTSM-P threshold of 1.725 was found to be a significant cutoff value for high-grade glioma/metastases prediction with 0.619 sensitivity and 0.889 specificity. Conclusion: The rCBVTSM-P derived from MR PWI may be helpful in differentiating high-grade gliomas from solitary metastases.

Objective: the purpose of this study was to assess residual neck nodes after Chemo- Radiotherapy (CRT) using diffusion and T2 weighted imaging. Methods: Apparent Diffusion Coefficient (ADC) and T2 signal intensity were retrospectively analyzed in 45 residual nodes 8 weeks after the end of CRT. Sensitivity, specificity, Positive and Negative Predictive Values (PPV, NPV) and accuracy of diffusion and T2 weighted sequences were calculated. In a subgroup of 25 patients ADC and T2 signal intensity were compared to Positron Emission Tomography - Computed Tomography (PET-CT) results. Histopathology and a 24 months follow- up after CRT were used as reference standard. Results: T2 signal intensity showed an accuracy of 63%, a sensitivity of 73.3%, a specificity of 55.6%, a PPV of 53.8% and a NPV of 75%. ADC showed an accuracy of 71.1%, using a threshold of 1.25 10-3 mm2/s, a sensitivity of 52.6%, a specificity of 84.6%, a PPV of 71.4% and a NPV of 71%. The combination of T2 signal intensity and ADC showed an accuracy of 73.3%, a sensitivity of 50%, a specificity of 84.6%, a PPV of 60% and a NPV of 78.7%. In the subset of 25 patients we correlated ADC and T2 data with PET-CT data: the latter showed higher accuracy, sensitivity and NPV and lower specificity; the PPV was similar between the two methods. Conclusion: Our results demonstrated that MRI, using diffusion and T2 weighted imaging, has high specificity in detecting post-CRT residual neck nodes but it has a lower accuracy, sensitivity and NPV than PET-CT.

Background: To simulate the anisotropic solid texture in 3D space, a kind of solid texture synthesis algorithm based on the samples that extends the 2D texture patch pasting technology to 3D hepatic artery solids is proposed. Methods: Firstly, the surface model is tetrahedral with Delaunay tetrahedral method. Secondly, the color attribute is used to generate the depth field and the depth field gradient direction is the first vector field, then the second vector field is defined by the arrow icon. The third vector is the product of the first two vectors. Finally, the vector field of the whole model is generated by Laplasse interpolation. Results & Conclusion: The experimental results showed that the proposed method can be used to synthesize the desired texture effect on any three-dimensional region, and it can simulate the internal structure of the hepatic artery and realize the real-time visualization of the cutting.

Objective: The Vestibulocerebellar Tract (VCT) connects the vestibular nuclear complex mainly with the uvula-nodulus of the cerebellum. Many animal studies have revealed the anatomy of the VCT, but very little is known about the VCT in the human brain. In this study, we attempted to identify the VCT in the normal human brain using diffusion tensor tractography. Methods: Twenty healthy volunteers were recruited for this study. Diffusion tensor imaging data was acquired using 1.5-T, and the VCT was obtained using FMRIB software. The VCT was determined by selection of fibers passing through two Regions of Interest (ROIs): seed ROI - vestibular nuclei areas, target ROI - the uvula of the cerebellum. Results: VCTs that originated from the vestibular nuclei passed posteriorly via the inferior cerebellar peduncle, then terminated at the uvula of cerebellum. No significant differences were observed in terms of fractional anisotropy, mean diffusivity, and tract volume of the VCT according to the side of the hemisphere (p > 0.05). Conclusion: We identified the VCT in the human brain using diffusion tensor tractography and believe that the methodology and results reported here will be helpful to researchers and clinicians in this field.

Background: PET imaging involves several parameters. However, it is not well known which parameters in each measurable site of tumors may have a greater effect on oncologic outcomes. Method: This study included a total of 42 patients diagnosed with histology-proven NSCLC. Baseline and consecutive FDG-PET scans were carried out 3 weeks before and at least 3 months after treatment, respectively. Results: According to the PERCIST, 9 (21%) patients had Complete Response (CR), and 3 of them underwent surgical resection. Sixteen patients (38%) had Partial Response (PR), and 4 (9.5%) had Stable Disease (SD), the remaining 15 (35%) had Progressive Disease (PD). Patients who responded to the treatment had significantly higher Overall Survival (OS) and Progression Free Survival (PFS), 15 versus 9.5 months and 10 versus 6 months, respectively. The SUVmax values decreased from 12.2 to 3.7 (P=<0.001), and the mediastinal lymph nodes SUVmax decreased from 6 to 0 (P=<0.001) in the responder group. The median TLG of the primary lesion showed a statistically significant decrease from 279 to 13.6 (P=0.0001), the median TLG of the mediastinal lymph nodes decreased from 13.2 to 0 (P= 0.04), and the median Whole Body (WB) TLG decreased from 338 to 17 (P=0.002) in the responder group. The posttreatment WB TLG was significantly correlated with OS and PFS. Conclusion: This study showed that the posttreatment WB TLG is significantly associated with oncologic outcomes. The metabolic changes in the mediastinal lymph nodes, as shown by the SUVmax and TLG, have a prognostic value in patients who develop new metastases during treatment. Further study in a homogenous NSCLC population can help increase our understanding of the prognostic value of PET/CT parameters.

Background: Endoscopic methods are well accepted as important treatments to clear the bile duct in patients with choledocholithiasis. However, little is known about bile dynamics after endoscopic treatment. Objective: The aim of this study was to assess the hepatic clearance and gallbladder function using fatty meal Cholescintigraphy (CS) in patients after endoscopic treatment for choledocholithiasis. Methods: Fatty meal CS was performed after intravenous injection of 8 mCi Tc-99m DISIDA. Hepatic clearance was calculated at 45 minutes (E45) after radiotracer injection. Gallbladder Ejection Fraction (GBEF) values were obtained at 30, 45, 60 and 75 minutes after fatty meal ingestion. Results: Thirty-nine patients were recruited. E45 was 39.5 ± 12.7%, which was significantly lower than that in the cholecystectomized asymptomatic controls (66.2 ± 9%, p < 0.001) in our prior study. The gallbladder was not properly filled in 13 patients. The GBEF values at 30, 45, 60 and 75 minutes were 82.7 ± 14.6%, 90.4 ± 12.8%, 90.4 ± 11.1%, and 93.9 ± 9.2%, respectively. GBEF was lower in the presence of gallstones, which reached statistical significance at 45 minutes and 60 minutes in the comparison of patients with and without gallstones (p < 0.05). Conclusion: Hepatic clearance in patients after endoscopic treatment for choledocholithiasis was generally lower. One-third of gallbladders lose their usual function. The GBEF was significantly reduced in the patients with gallstones. We suggested that GBEF with fatty meal stimulation could be used to follow patients after endoscopic treatment for choledocholithiasis.

Objective: The aims of this study were to: a) evaluate tumor response rates using modified- Response-evaluation-criteria-in-solid-tumors (mRecist) criteria, b) evaluate safety of Degradable Starch Microspheres Trans-arterial-chemo-embolization (DSMs-TACE) for unresectable hepatocellular- carcinoma (HCC) treatment. Materials and Methods: We prospectively enrolled 24 HCC cirrhotic patients (21/3 M/F, mean age 66.3 years) to be treated with repeated DSMs-TACE procedures, performed at 4-6 week intervals on the basis of tumor response and patients tolerance. Clinical and biochemical evaluations were performed before and after each procedure. Treatment response was also assessed by Computedtomography (CT) or Magnetic-resonance-imaging (MRI)-scan 4-6 weeks following each procedure. Results: In our experience, DSMs-TACE was both safe and effective. A total of 53 DSMs-TACE procedures were performed (2.2 per patient). No procedure-related death was observed. Complete Response (CR) was observed in 5/24 (20.8%), 4/17 (23.5%) and 5/12 (41.6%) patients after the first, second and third procedure, respectively. At the end of each treatment, all patients experienced at least a partial response. At the end of the repeated procedures, no differences between mono- or bi-lobar disease were observed in patients with CR (64.2% vs 50%; p=ns). In most cases, treatment discontinuation was due to worsening liver function. Conclusion: DSMs-TACE is a valid, well-tolerated alternative treatment to Lipiodol-TACE or DEB-TACE, as it has demonstrated to achieve a relatively high percentage of complete tumor necrosis. CR rates were similar between patients with mono- or bi-lobar disease indicating the possibility of carrying-out repeated procedure in a safe and effective way in both types of patients.