Sagittal Slices Research Articles

Abstract PD9-07: The rate of parenchymal enhancement during DCE-MRI reflects response to neoadjuvant therapy

Abstract Background: Dynamic contrast-enhanced MRI (DCE-MRI) plays a crucial role in breast cancer detection and monitoring of therapeutic response by assessing the tumor vascular network. The MRI contrast agent enhances not only tumor tissue, but also normal breast tissue, a phenomena known as background parenchymal enhancement (BPE). The degree of BPE can be variable between different individuals, is influenced by the hormonal milieu, and displays variable anatomical and kinetic patterns. Changes in BPE intensity over the course of neoadjuvant therapy (NAT) have been found to correlate with response to therapy in both the ipsilateral and contralateral breast. The present study assesses changes in the rate of BPE over the course of NAT to predict eventual response to NAT. Importantly, this study was conducted in collaboration with community-care centers (i.e., not research-oriented medical centers). Methods: Women with locally advanced breast cancer (N = 19) were imaged four times during the course of NAT: 1) prior to the start of NAT, 2) after 1 cycle of NAT, 3) after 2-4 cycles of NAT, and 4) 1 cycle after MRI #3. Imaging data was acquired on 3T Siemens Skyra scanners equipped with a 8- or 16-channel breast coils. Gadolinium-based contrast agent (0.1 mmol/kg of Multihance or 10 mL of Gadovist) was administered intravenously at 2 mL/sec after the acquisition of baseline scans. DCE-MRI data was collected in 10 sagittal slices with a temporal resolution of 7.27 sec for a total acquisition time of eight minutes. The tumor was semi-automatically segmented using a manually drawn region of interest (ROI) followed by fuzzy c-means clustering of a post-contrast high-resolution anatomical scan. Fibroglandular parenchyma was segmented from adipose tissue using fuzzy c-means clustering and further segmented into three concentric rings at prescribed radial distances from the tumor boundary. Enhancement time courses were temporally smoothed using a moving average filter of width 5 and normalized to baseline intensity to yield relative enhancement time courses. The enhancement slope, defined as the slope of the curve at half maximal intensity, was calculated from the kinetic uptake curve for the tumor, adipose, and concentric rings of fibroglandular tissue surrounding the tumor. Results: Prior to commencing NAT, women with faster tumoral enhancement had faster fibroglandular enhancement (r = 0.62, p < 0.05). There was no correlation between tumor and adipose enhancement rate or fibroglandular and adipose enhancement rate. Among concentric rings of fibroglandular tissue, the tissue most distal to the tumor enhanced faster than tissue more proximal to the tumor (p < 0.05). Over the course of NAT, the rate of fibroglandular enhancement tended to slow (p < 0.01). Women who ultimately achieved pathological complete response (pCR; n = 6) displayed a faster rate of fibroglandular enhancement at the 4th MRI than women with residual tumor at the time of resection (n = 13). This correlation was preserved (p < 0.01) in a subset of patients who received AC◊T therapy, the most common treatment arm in this study (n = 4 pCR, n = 8 non-pCR). Furthermore, this relationship persisted when the fibroglandular enhancement rate was normalized by the tumoral enhancement rate in both the full patient cohort and in the 12 patients who received AC◊T (p < 0.05, both cohorts). Conclusions: These results demonstrate that the rate of fibroglandular tissue enhancement in patients receiving NAT for breast cancer reflects response to therapy. The increased rate of fibroglandular enhancement in women achieving pCR may reflect treatment-mediated vascular remodeling in healthy breast tissue. Incorporating quantitative characterization of parenchymal enhancement kinetics into treatment response models warrants further investigation. Citation Format: John Virostko, Erin M Higgins, Chengyue Wu, Anna G Sorace, Debra Patt, Boone Goodgame, Thomas E Yankeelov. The rate of parenchymal enhancement during DCE-MRI reflects response to neoadjuvant therapy [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr PD9-07.

Assessment of knee pain from MR imaging using a convolutional Siamese network.

It remains difficult to characterize the source of pain in knee joints either using radiographs or magnetic resonance imaging (MRI). We sought to determine if advanced machine learning methods such as deep neural networks could distinguish knees with pain from those without it and identify the structural features that are associated with knee pain. We constructed a convolutional Siamese network to associate MRI scans obtained on subjects from the Osteoarthritis Initiative (OAI) with frequent unilateral knee pain comparing the knee with frequent pain to the contralateral knee without pain. The Siamese network architecture enabled pairwise learning of information from two-dimensional (2D) sagittal intermediate-weighted turbo spin echo slices obtained from similar locations on both knees. Class activation mapping (CAM) was utilized to create saliency maps, which highlighted the regions most associated with knee pain. The MRI scans and the CAMs of each subject were reviewed by an expert radiologist to identify the presence of abnormalities within the model-predicted regions of high association. Using 10-fold cross-validation, our model achieved an area under curve (AUC) value of 0.808. When individuals whose knee WOMAC pain scores were not discordant were excluded, model performance increased to 0.853. The radiologist review revealed that about 86% of the cases that were predicted correctly had effusion-synovitis within the regions that were most associated with pain. This study demonstrates a proof of principle that deep learning can be applied to assess knee pain from MRI scans. • Our article is the first to leverage a deep learning framework to associate MR images of the knee with knee pain. • We developed a convolutional Siamese network that had the ability to fuse information from multiple two-dimensional (2D) MRI slices from the knee with pain and the contralateral knee of the same individual without pain to predict unilateral knee pain. • Our model achieved an area under curve (AUC) value of 0.808. When individuals who had WOMAC pain scores that were not discordant for knees (pain discordance < 3) were excluded, model performance increased to 0.853.

The evaluation of lateral pterygoid signal intensity changes related to temporomandibular joint anterior disc displacement.

The aim of this study was to evaluate lateral pterygoid muscle (LPM) signal intensity changes related to different degrees of disc status groups using magnetic resonance imaging (MRI). Four groups were established according to disc displacement of examined temporomandibular joint (TMJ) and opposing TMJ. MRI scans were obtained with a 1.5T device. The signal intensity was measured from the sagittal slice where midportion of the muscle belly was observed. The regions of interest were placed over the superior and inferior heads of LPM. To normalize differences in signal intensity, signal intensity of the grey matter and white matter of the brain was also measured. Signal intensity ratio (SIR) was calculated. Measurements were performed in both T1- and T2-weighted images. In the intergroup comparisons, a statistically significant difference was found for SIR1 sup., SIR1 inf., SIR2 sup. and SIR2 inf. variables (p < 0.01). According to multiple comparison test (Tukey), no significant difference was found between group 0 and group 1 for SIR1 sup., SIR1 inf., SIR2 sup. and SIR2 inf. variables (p > 0.05). For SIR1 sup., SIR1 inf., SIR2 sup. and SIR2 inf. variables, significant differences were found between five binary comparison groups (1-2, 1-3, 2-3, 0-2, 0-3) (p < 0.05). In conclusion, SIR is correlated with the degree of disc displacement. The internal derangement of the opposite joint did not affect SIR of superior and inferior heads of examined LPM in both T1- and T2-weighted images. The inferior head of LPM was also found affected by disc displacement.

Fully automated segmentation of left ventricular scar from 3D late gadolinium enhancement magnetic resonance imaging using a cascaded multi-planar U-Net (CMPU-Net).

Three-dimensional (3D) late gadolinium enhancement magnetic resonance (LGE-MR) imaging enables the quantification of myocardial scar at high resolution with unprecedented volumetric visualization. Automated segmentation of myocardial scar is critical for the potential clinical translation of this technique given the number of tomographic images acquired. In this paper, we describe the development of cascaded multi-planar U-Net (CMPU-Net) to efficiently segment the boundary of the left ventricle (LV) myocardium and scar from 3D LGE-MR images. In this approach, two subnets, each containing three U-Nets, were cascaded to first segment the LV myocardium and then segment the scar within the presegmented LV myocardium. The U-Nets were trained separately using two-dimensional (2D) slices extracted from axial, sagittal, and coronal slices of 3D LGE-MR images. We used 3D LGE-MR images from 34 subjects with chronic ischemic cardiomyopathy. The U-Nets were trained using 8430 slices, extracted in three orthogonal directions from 18 images. In the testing phase, the outputs of U-Nets of each subnet were combined using the majority voting system for final label prediction of each voxel in the image. The developed method was tested for accuracy by comparing its results to manual segmentations of LV myocardium and LV scar from 7250 slices extracted from 16 3D LGE-MR images. Our method was also compared to numerous alternative methods based on machine learning, energy minimization, and intensity-thresholds. Our algorithm reported a mean dice similarity coefficient (DSC), absolute volume difference (AVD), and Hausdorff distance (HD) of 85.14%±3.36%, 43.72±27.18cm3 , and 19.21±4.74mm for determining the boundaries of LV myocardium from LGE-MR images. Our method also yielded a mean DSC, AVD, and HD of 88.61%±2.54%, 9.33±7.24cm3 , and 17.04±9.93mm for LV scar segmentation on the unobserved test dataset. Our method significantly outperformed the alternative techniques in segmentation accuracy (P<0.05). The CMPU-Net method provided fully automated segmentation of LV scar from 3D LGE-MR images and outperformed the alternative techniques.

PERBEDAAN KUALITAS CITRA MRI LUMBAL SEKUEN TIWI TSE POTONGAN SAGITAL DENGAN DAN TANPA PENGGUNAAN SENSITIVITY ENCODING (SENSE) PADA KASUS HERNIA NUCLEUS PULPOSUS

Background: Parallel imaging is a time-reduction technique that uses phased-array coils. Phased array coils measure and process signals from a piece, then combine pieces to form an image of a larger area of anatomy. Commonly used imaging parallel technique is one of them is sensitivity encoding or SENSE. This research is the difference of image quality MRI Lumbal sequence T1WI TSE of sagittal pieces with and without the use of SENSE in the case of hernia nucleus pulposus (HNP). The aim of this research is to know the difference between SNR, anatomy information and better image quality and anatomical information. MRI Lumbal sequence T1WI TSE sagittal pieces with and without the SENSE in HNP.Methods: The type of this research is quantitative with an experimental approach, The research was conducted in RS dr. Saiful Anwar Malang. The research was taken from Lumbal MRI examination in 9 patients with predetermined exclusion and inclusion criteria of T1WI TSE sectional pieces with and without the use of SENSE in HNP. Image assessment was done by two radiologists. Analysis of data used paired T-test and Wilcoxon test with alpha 5%.Results: The results showed that there was a difference in image quality of MRI Lumbal T1WI TSE sequence of sagittal pieces with and without the use of SENSE in HNP with p-value = 0.001, and anatomic information with p-value = 0.001. Sensitivity encoding resulted in better image quality and anatomical information, with mean values at SNR 181,4333, and rank information anatomy 11,00.Conclusion: There is a difference in SNR and anatomical information between the use of SENSE andwithout the use of SENSE on sagittal lumbar MRI. The use of SENSE is capable of producing quality MRIimagery (SNR) and Anatomical Information on lumbar MRI lumbar sequences of sagittal slices.

Pathological examination of CT findings of tumor infiltration to the periarterial plexus in pancreatic cancer.

644 Background: The radiographic diagnosis of tumor infiltration into the periarterial plexus in pancreatic cancer is important because it is related to the classification, however, it is difficult to distinguish the abnormal shadow along the artery caused by inflammation or cancer infiltration. The aim of this study is to investigate CT values of the abnormal shadow along artery could distinguish between inflammation and tumor invasion. Methods: Study 1: Of 26 patients who underwent DP-CAR between 2009 and 2018, we analyzed 19 patients who had dynamic CT and obtained sagittal slice taken 120 seconds after injection with less than 2.5 mm slice thickness. At first, we measured CT values at upper and lower point of CeA and CHA each sagittal slice using CT. Next, we evaluated tumor invasion at the upper and lower plexus of CeA and CHA in each section of the pathological specimen, and evaluated the relationship between the tumor invasion and the CT value. Study 2: Using these 19 patients and 40 patients who underwent DP for PDAC between 2010 and 2014, we analyzed the relation between CT value and long-term states. Results: Study 1: CT value was totally measured at the 606 points using 19 patients who underwent DP-CAR. At the 490 points, we did not observe cancer infiltration and fibrosis. At the 70 points, we observed fibrosis without cancer cells. At the 46 points, we observed cancer infiltration. CT value was significantly higher in the tumor infiltration group than that in the without cancer infiltration and fibrosis group (P &lt; 0.01). Study 2: The best cut-off of CT value of the presence of cancer infiltration was 44.9 HU using ROC curve (AUC = 0.861). The median survival time of patients who had the points of CT value &gt; 44.9 HU around arteries was significantly shorter than that of patients who did not have the points of CT value &gt; 44.9 HU (2.17 vs. 4.55 years, p = 0.03). Conclusions: The CT value around the arteries was significantly higher in the points of pathological tumor infiltration than that in the points of fibrosis without cancer cells. The best cut-off CT value of the presence of cancer infiltration around arteries was 44.9 HU, and the presence of the point of CT value &gt; 44.9 HU around arteries was associated with poor survival.

Reliability of cephalograms derived of cone beam computed tomography versus lateral cephalograms to estimate cervical vertebrae maturity in a Peruvian population: A retrospective study

Cone beam computed tomography (CBCT) images can be useful for estimating cervical vertebrae maturity (CVM). The aim of this study was to evaluate the reliability of cephalograms derived from CBCT versus lateral cephalograms (LC) to estimate the CVM in a Peruvian population. The sample evaluated consisted of 40 cephalograms derived from CBCT and 40 LC images from individuals aged 10-19 years. One trained and calibrated observer (Kappa scores≥0.90) interpreted the CBCT and LC images twice. Intra-observer reliability of each maturation stage on CBCT sagittal slices and LC images were analysed using the weighted kappa statistics (α=0.05). Comparison of CVM stages between CBCT slices and LC images were analysed by the Spearman rank correlation coefficient, p<0.05. The weighted kappa test showed almost perfect intra-observer agreement for the CVM stages using the CBCT sagittal slices (0.873). Considering the LC images, the weighted kappa test showed almost perfect intra-observer agreement too (0.937). In both intra-observer agreement, the difference was limited to one maturation stage of the CVM method. The first and second intra-observer agreement for the CVM stages between the CBCT sagittal slices and LC images were almost perfect (0.937 and 0.874). High correlation values at the first (0.975) and second (0.976) intra-observer agreement for the CVM stages between CBCT sagittal slices and LC images were also found. CBCT is a reliable method for CVM assessment and can be used as an alternative method for this purpose. The orthodontists might use the CBCT scans as a valuable tool for CVM method estimation.

Three-dimensional architecture of the neurovascular and adipose zones of the upper and lower lumbar intervertebral foramina: an epoxy sheet plastination study.

Kambin's triangle and the safe triangle are common posterolateral approaches for lumbar transforaminal endoscopic surgery and epidural injection. To date, no consensus has been reached on the optimal transforaminal approach, in particular its underlying anatomical mechanism. The aim of this study was to investigate the 3D architecture of the neurovascular and adipose zones in the upper and lower lumbar intervertebral foramina (IVFs). Using the epoxy sheet plastination technology, 22 cadaveric lumbar spines (12 female and 10 male, age range 46-89 years) were prepared as a series of transverse (11 sets), sagittal (8 sets), and coronal (3 sets) slices with a thickness of 0.25 mm (6 sets) or 2.5 mm (16 sets). The high-resolution images of the slices were scanned and analyzed. The height, area, and volume of 30 IVFs from T12-L1 to L4-5 were estimated and compared. This study was performed in accord with the authors' institutional ethical guidelines and approved by the institutional ethics committees. The findings were as follows. 1) The 3D boundaries of the lumbar IVF and its subdivisions were precisely defined. 2) The 3D configuration of the neurovascular and adipose zones was different between the upper and lower lumbar IVFs; zoning in the upper lumbar IVFs was much more complex than that in the lower lumbar IVFs. 3) In general, the infraneural adipose zone gradually tapered and rotated from the inferoposterolateral aspect to the superoanteromedial aspect. 4) The average height, area, and volume of the IVF gradually increased from the upper to the lower lumbar spine. Within a lumbar IVF, the volumes below and above the inferior border of the dorsal root ganglia were similar. This study highlights differences of fine 3D architecture of neurovascular and adipose tissues between the upper and lower lumbar IVFs, with related effects on the transforaminal approaches. The findings may contribute to optimization of the surgical approaches to and through the IVF at different lumbar spinal levels and also may help to shorten the learning curve for the transforminal techniques.

The Stacked-Ellipse Algorithm: An Ultrasound-Based 3-D Uterine Segmentation Tool for Enabling Adaptive Radiotherapy for Uterine Cervix Cancer

The stacked-ellipse (SE) algorithm was developed to rapidly segment the uterus on 3-D ultrasound (US) for the purpose of enabling US-guided adaptive radiotherapy (RT) for uterine cervix cancer patients. The algorithm was initialised manually on a single sagittal slice to provide a series of elliptical initialisation contours in semi-axial planes along the uterus. The elliptical initialisation contours were deformed according to US features such that they conformed to the uterine boundary. The uterus of 15 patients was scanned with 3-D US using the Clarity System (Elekta Ltd.) at multiple days during RT and manually contoured (n = 49 images and corresponding contours). The median (interquartile range) Dice similarity coefficient and mean surface-to-surface-distance between the SE algorithm and manual contours were 0.80 (0.03) and 3.3 (0.2) mm, respectively, which are within the ranges of reported inter-observer contouring variabilities. The SE algorithm could be implemented in adaptive RT to precisely segment the uterus on 3-D US.

Normal anatomic reference of pastern and coffin joints in Egyptian buffalo (Bubalus bubalis): A compared atlas of cross-sectional anatomy, magnetic resonance imaging and computed tomography.

The purpose of this study was to describe normal magnetic resonance imaging and computed tomographic anatomy of pastern and coffin joints in Egyptian buffalo using cadaveric distal limbs. This study was achieved using twelve fresh cadaveric distal limbs from adult healthy buffaloes of both sexes. These cadaveric limbs were scanned using a 1Tesla MRI scanner and CT scanner, injected with red latex, frozen at -20°C for 1week, and then sectioned into sagittal, dorsal and transverse slices. The obtained MR and CT images were selected to be matched with their corresponding anatomical cross-sections for identification and evaluation of the clinically correlated anatomical structures of the pastern and coffin joints. The difference in signal intensities on CT and MRI scans amongst the tissues allowed clear differentiation of major bone and soft tissue structures of the pastern and coffin joints. CT provided a high spatial resolution of bone and soft tissue structures, however, MRI allowed a better and higher resolution and definition between soft tissues. The current study provided a normal CT and MRI anatomic reference which could help veterinary clinicians for interpretation and diagnosis of the clinically affected pastern and coffin joints in buffalo.

Comparison of morphological measures of nasal bone between genders using computed tomography

Facial anthropometry plays an important role in forensic investigations especially in identification of decomposed bodies from skeletal remains. Dimensions of nasal bone is one such parameter that can be used to determine sex. The main aim of the study is to compare the morphological measures of nasal bone between genders using computed tomography. This is a retrospective study on the patients visiting the Dept. of Radiodiagnosis and Imaging, Kasturba Hospital, Manipal. Based on the inclusion and exclusion criteria, 240 subjects (120 males and 120 females) who were referred for routine CT OMC were included in the study. The axial sections were converted into coronal and sagittal slices using the multiplanar reconstruction tool. Three anthropometric parameters were measured. A total of 240 (120 males and 120 females) data was obtained with age group ranging from 20–50 years. Mean and Standard deviation was calculated. The data was normally distributed hence independent sample T test was performed to find the difference between males and females. Null hypothesis stated that there was no significant difference based on the anthropometric measurements between the males and females since the (p>0.05). Alternate hypothesis stated that there is significant difference noted between the males and females (p<0.001). In present study, three measurements of the nasal bone were taken such as the nasofrontal angle, pyramidal angle and the linear distance out of which two variables i.e. the pyramidal angle and the linear distance were good discriminators which contributed high accuracy in comparison to the other parameter.

Deep Learning-Based HCNN and CRF-RRNN Model for Brain Tumor Segmentation

This paper proposes a strategy where a structure is developed to recognize and order the tumor type. Over a time of years, numerous specialists have been examined and proposed a technique in this space. A brain tumor segmentation approach is developed based on efficient, deep learning techniques implemented in a unified system to achieve the appearance and spatial accuracy outcomes through Conditional Radom Fields (CRF) and Heterogeneous Convolution Neural Networks (HCNN). In these steps the 2D image patching and picture slices of the deep-learning model is developed. The Proposed method has following steps as follows: 1) train HCNN by image patches; 2) train CRF with CRF-Recurrent Regression based Neural Network (RRNN) by means of image slices with fixed variables of HCNN; 3) fine tune with HCNN and CRF-RRNN image slices. In general, 3 segmentation models have been trained using axial-, coronary-and sagittal image patches and slices, Further assembled into brain tumor segments using a voting fusion technique and it can be examined with Internet of Medical Things (IoMT) Platform. The experimental results proved that our approach has been capable of developing a Flair, T1c and T2 segmenting model and of achieving good performance as with Flair, T1, T1c, and T2 scans.

Cross-sectional anatomy, magnetic resonance imaging and computed tomography of fetlock joint in camel

The current study aimed to describe the normal cross-sectional anatomy, magnetic resonance imaging, and computed tomography of the fetlock joint in adult healthy dromedary camel. This study was carried out on twelve fetlock joints of the fresh cadavers from three camels. The gross examination of these camels revealed that they were normal with no orthopedic abnormalities. The cadaveric fetlock joints (n=12) were scanned using a computed tomographic (CT) scanner and a 1 Tesla MRI scanner. Then the joints were injected with a colored latex and sectioned into transverse, dorsal, and sagittal slices. Cross-anatomical sections were correlated with their corresponding CT and MR images for evaluation of the normal relevant anatomical structures that appeared with different signal intensities on the CT and MRI scans. The current investigation revealed that all the major soft tissue structures in the fetlock joint of the dromedary camel were clearly visualized on both CT and MR scans, except the short and the cruciate sesamoidean ligaments that could not be identified on both the CT and MR images. The anatomical sections with their corresponding CT and MR images obtained in this study could be used as a reference for subsequent clinical diagnosis and the interpretation of fetlock joint pathologies in dromedary camel.

Normal cross-sectional anatomy and magnetic resonance imaging of pastern and coffin joints in camel

The present study aimed to describe the normal cross sectional anatomy and magnetic resonance imaging of pastern and coffin joints in dromedary camel. This study was conducted on twelve distal limbs (fore and hind) of fresh cadavers from three adult camels of both sexes. The specimens appeared normal without orthopedic disorders. Twelve distal limbs were scanned using a 1 Tesla MRI scanner and then injected with colored latex to be sectioned into sagittal, dorsal and transverse slices. Cross anatomical sections were matched with their corresponding MR images for identification and evaluation of the clinically relevant anatomical structures that appeared with different signal intensities on MRI scans. The present study showed that all major soft tissues in pastern and coffin joints of camel were clearly depicted on MR images, however, the palmar/plantar ligaments of pastern joint and ligaments of navicular cartilage could not be identified on MR images. The annotated cross anatomical sections with the corresponding MR images could be used as a normal reference for interpretation of some clinical diseases in pastern and coffin joints of camel.

The sacral screw placement depending on morphological and anatomical peculiarities.

Various pathologies of the lumbosacral junction require fusion of the L5/S1 segment. However, pseudarthroses, which often come along with sacral screw loosening, are problematic. The aim of the present investigation was to elaborate the morphological features of the L5/S1 segment to define a so-called "safe zone" for bi- or tricortical screw placement without risking a damage of the iliac vessels. A total of one hundred computed tomographies of the pelvis were included in this investigation. On axial and sagittal slices, pedicle morphologies, the prevertebral position of the iliac vessels, the spinal canal and the area with the largest bone density were analyzed. Beginning from the entry point of S1-srews iliac vessels were located at an average angle of 7° convergence, the spinal canal at 38°. Bone density was significantly higher centrally with a mean value of 276 Hounsfield Units compared to the area of the Ala ossis sacri. The largest intraosseous screw length could be achieved at an angle of 25°. The average pedicle width was 20mm and the pedicle height 13mm. A "safe-zone" for bicortical screw placement at S1 with regard to the course of the iliac vessels could be defined between 7° and 38° convergence. Regarding the area offering the largest bone density and the maximal possible screw length, a convergence of 25° is recommended at S1 to reduce the incidence of screw loosening. Screw diameter, as a further influence factor on screw holding, is limited by pedicle height not pedicle width.

Detecting liver fibrosis using a machine learning-based approach to the quantification of the heart-induced deformation in tagged MR images.

Liver disease causes millions of deaths per year worldwide, and approximately half of these cases are due to cirrhosis, which is an advanced stage of liver fibrosis that can be accompanied by liver failure and portal hypertension. Early detection of liver fibrosis helps in improving its treatment and prevents its progression to cirrhosis. In this work, we present a novel noninvasive method to detect liver fibrosis from tagged MRI images using a machine learning-based approach. Specifically, coronal and sagittal tagged MRI imaging are analyzed separately to capture cardiac-induced deformation of the liver. The liver is manually delineated and a novel image feature, namely, the histogram of the peak strain (HPS) value, is computed from the segmented liver region and is used to classify the liver as being either normal or fibrotic. Classification is achieved using a support vector machine algorithm. The in vivo study included 15 healthy volunteers (10 males; age range 30-45years) and 22 patients (15 males; age range 25-50years) with liver fibrosis verified and graded by transient elastography, and 10 patients only had a liver biopsy and were diagnosed with a score of F3-F4. The proposed method demonstrates the usefulness and efficiency of extracting the HPS features from the sagittal slices for patients with moderate fibrosis. Cross-validation of the method showed an accuracy of 83.7% (specificity=86.6%, sensitivity=81.8%).

The accuracy of standard multiple sclerosis MRI brain sequences for the diagnosis of optic neuropathy

BackgroundDetection of optic neuropathy on MRI has potential implications for the diagnosis and management of Multiple Sclerosis (MS). ObjectiveThis study assessed the accuracy of T2 sagittal MRI brain for detection of optic neuropathy, compared to coronal STIR orbit. Methods and materialsRetrospective single-center blinded diagnostic accuracy study of 100 consecutive patients who underwent both T2 sagittal brain and coronal STIR orbit MRI. All were performed on 1.5T scanners. T2 sagittal slice thickness was 4 mm for the first 50 patients (group1) and 3 mm for the second 50 (group2). The MRIs were reviewed in a blinded fashion to determine the presence of optic neuropathy. Coronal STIR orbit sequences were considered the diagnostic reference standard. ResultsThe sensitivity of T2 sagittal brain imaging for ON was 44% in group 1 and 85% in group 2 (p = 0.007). The specificities were 98% and 97% respectively (p = 0.9). Sensitivity was poorest for evaluation of the intraorbital nerve segment (56% grp1, 69% grp2, p = 0.4). ConclusionT2 sagittal MRI brain has high specificity for the detection of optic neuropathy when compared to coronal STIR orbit. Sensitivity is increased when slice thickness is reduced, but remains poor for evaluation of the intraorbital segment.

Design and fabrication of a new multi-loop saddle coil for 1.5 T MRI.

Radiofrequency coils provide high-resolution magnetic resonance (MR) imaging of human tissues. A small RF coil produces MR images with a higher resolution compared to the commercial volume MR coils from mass limited samples. Signal to noise ratio (SNR) plays a key role in the optimal design of receiver radiofrequency coils. In this work, we present a three-loop saddle coil suitable for MR imaging of digits of the human body. The geometry of the introduced coil is optimized to achieve the highest SNR. The coil performance is evaluated through comparing the measured SNR maps of the optimal coil derived from MR images of a saline phantom with the corresponding measured SNR maps of a commercial head coil in axial and sagittal slices. Results verify that the image SNR of the introduced coil is 3.4 times higher than that of the head coil and 2 times higher than that of the similar saddle coils represented in the literature recently. To validate the measured results, SNR maps of the introduced saddle and head coils were simulated and their SNR difference was compared with the corresponding measured data of the two coils. Results show that the simulated and measured data are in agreement with less than 11.8% error.

Quantitative evaluation for cervical foraminal bony stenosis based on angled sagittal slices along a nerve root on computed tomography

Cervical foraminal canal stenosis is a common disease, but any relationships between the measurement values of cervical foraminal canals and clinical symptoms have yet to be explored. We aim to determine a numerical cutoff point of cervical foraminal bony canal size that does not lead to radiculopathy so as to establish criteria for the surgical indication. We reconstructed angled sagittal slices along a nerve root on computed tomography (CT) on a workstation from pre-operative CT data and measured 1152 cervical foraminal canals (144 patients) from Cervical (C) 4/5 to C7/Thoracic (Th) 1. We evaluated the relationship between the size of foraminal canals and clinical manifestations. Receiver operating characteristic (ROC) analysis was used to calculate cutoff points of each foraminal canal size with positive neurologic manifestations. Of the 144 patients’ 1152 nerve roots, 286 nerve roots (24.8%) were diagnosed as radiculopathy by neurological examinations. The mean measured value of all foraminal canals on angled sagittal CT imagery was 3.39 ± 1.37 mm. The cutoff point of foraminal canal sizes without radiculopathy was 2.7 mm (sensitivity 0.680, specificity 0.591) overall. A cutoff point ascertained by quantitative evaluation of cervical foraminal canal size is useful for making diagnosis of cervical foraminal canal bony stenosis.

Inter-rater reliability of TT-TG distance is good and does not vary based on preselected versus independent slice selection on MRI

ObjectivesPatellar instability is a frequent cause of knee dysfunction in young, active patients. Tibial tubercle–trochlear groove (TT-TG) distance, trochlear morphology (trochlear depth and sulcus angle) and patellar height are felt to contribute to patellar instability and may influence treatment. These measurements are frequently performed on MRI images. We hypothesised that inter-rater reliability of measures would be good and that inter-rater variation is driven primarily by slice selection.MethodsTwenty-six patients with at least one documented episode of patellar instability confirmed by MRI were identified. Six raters reviewed MRI images from each patient. Each rater measured and recorded TT-TG distance, trochlear depth and sulcus angle, and patellotrochlear index (PTI) for each patient and the slices used for the measurements. Each rater repeated the measurement using preselected slices. Inter-rater reliability was calculated by intraclass correlations (ICCs) for slice selection and for TT-TG distance, trochlear morphology measures and PTI with both independently selected and preselected slices. Statistically significant differences (p<0.05) in ICC based on slice selection were defined as values without overlap of their 95% CIs.ResultsInter-rater reliability was excellent for tibial (ICC=0.93) axial slice selection and sagittal slice selection (ICC=0.94), and good for femoral (ICC=0.88) axial slice selection. Using independent slice selection, inter-rater reliability was good for TT-TG distance (ICC=0.79) and fair for trochlear depth (ICC=0.57), sulcus angle (ICC=0.57) and PTI (ICC=0.71). When preselected slices were used, inter-rater reliability was good for TT-TG distance (ICC=0.85), sulcus angle (ICC=0.83) and PTI (ICC=0.77) and fair for trochlear depth (ICC=0.68). Only sulcus angle demonstrated a significant (p<0.05) improvement in inter-rater reliability with the use of preselected slices.Discussion and conclusionInter-rater reliability of TT-TG distance is good and does not vary based on preselected versus independent slice selection on MRI. Inter-rater reliability of trochlear morphology measures based on axial MRI slices and PTI is fair. Inter-rater variation can be reduced (particularly in the case of sulcus angle) through agreement on slice selection.Level of evidenceLevel III, diagnostic.