Dimensional MRI Research Articles

REMNANT-PRESERVED ACUTE ANTERIOR CRUCIATE LIGAMENT RECONSTRUCTION WITH CONCOMITANT MEDIAL AND LATERAL MENISCAL TEAR REPAIR: RADIOLOGICAL, FUNCTIONAL, AND PATIENT-REPORTED OUTCOMES AFTER 12-MONTH FOLLOW-UP

Although remnant-preserved ACL reconstruction (ACLR) restores knee joint stability and dampens the problem of acute ACL rupture-induced knee pain, an increasing number of patients still develop post-traumatic osteoarthritis (PTOA) after 10 to 15 years of ACLR. We previously found that remnant-preserved ACLR with concomitant medial and lateral meniscus repair may not prevent cartilage degeneration and weaken muscle strength, while the clinical features of PTOA are not clear. We hypothesized that remnant-preserved ACLR with concomitant medial and lateral meniscus tears is related to early cartilage damage, worse function recovery, patient-reported outcomes (PROs) and delayed duration to return to sports. The aim is to evaluate the remnant-preserved ACLR with complicated meniscal injuries in predicting which patients are at higher risk of osteoarthritic changes, worse function and limited activities after ACLR for 12 months.Human ethical issue was approved by a committee from Xi'an Jiaotong University. 26 young and active patients (24 male, 2 female) with ACL injuries (Sherman type I and II) with concomitant medial and lateral meniscus within 2 months were included from January 2014 to March 2022. The average age of the ACLR+ meniscus repair was 26.77±1.52 (8 right, 5 left) and isolated ACLR control was 31.92±2.61 years old (7 left, 6 right). Remnant-preserved ACLR with a 5- to 6-strand hamstring tendon graft was operated on by the same sports medicine specialists. MRI CUBE-T2 scanning with 48 channels was conducted by a professional radiologist. The volume of the ACL graft was created through 3 dimensional MRI model (Mimics 19, Ann Arbor). Anterior Cruciate Ligament OsteoArthritis Score (ACLOAS) was applied to score visible cartilage damage. IKDC 2000 score and VAS were assessed by two blinded researchers. Results were presented as mean± SEM of each group.The cross-sectional area and 3D volume of the ACL graft were greater in the remnant-preserved ACLR+meniscus group compared with isolated ACLR (p=0.01). It showed that ACLR+ meniscus group had early signs of joint damage and delayed meniscus healing regarding ACLOAS compared to control group (p=0.045). MRI CUBE-T2 prediction of radiographic cartilage degeneration was not obvious in both groups post remnant-preserved ACLR over 12 months (p>0.05). However, higher VAS scores, lower IKDC scores, and long-last joint swelling were reported in the ACLR+ meniscus repair group at the end of 12 months follow-up. Although remnant-preserved ACLR+ meniscus was able to maintain the restore the knee function, it showed delayed timing (>12 months) to return to play at the pre-injury stage, while no difference between the timing of returning to the normal daily routine of their ACLR knee compared to control (p=0.30). The cost of ACLR+ meniscus (average 10,520.76$) was higher than the control group (6,452.92$, p=0.018).Remnants-preserved ACLR with concomitant injured medial and lateral meniscus repair shows a higher risk of cartilage damage, greater cost, worse functional performance, and longer time for young male patients to return to sports after 12-month follow-up compared to isolated ACLR. Further evidence and long-term follow-up are needed to better understand the association between these results and the risk of development of PTOA in this patient cohort.

ASCENDING AORTIC ANEURYSMS: SURGERY, RMODELLING & AI

ASCENDING AORTIC ANEURYSMS: SURGERY, RMODELLING & AI

Delineation of Subarachnoid Cisterns Using CT Cisternography, CT Brain Positive and Negative Contrast, and a Three Dimensional MRI Sequence: A Pictorial Review.

The basic anatomy and morphology of subarachnoid cisterns of the brain are interesting and challenging topics with high clinical significance. These enlarged CSF-filled expansions are important as they transmit various neurovascular structures. The cisterns can be classified based on their location as supratentorial, at the level of the tentorium, and infratentorial. They are also classified as paired and unpaired cisterns. The anatomical and radiological information about the cisterns is clinically and surgically relevant in diagnosing and managing many neurological disorders. It is also essential in medical teaching. This pictorial essay reviews the radiological images where the subarachnoid cisterns are delineated in four unique circumstances.

Geometric deep learning on brain shape predicts sex and age.

The complex relationship between the shape and function of the human brain remains elusive despite extensive studies of cortical folding over many decades. The analysis of cortical gyrification presents an opportunity to advance our knowledge about this relationship, and better understand the etiology of a variety of pathologies involving diverse degrees of cortical folding abnormalities. Hypothesis-driven surface-based approaches have been shown to be particularly efficient in their ability to accurately describe unique features of the folded sheet topology of the cortical ribbon. However, the utility of these approaches has been blunted by their reliance on manually defined features aiming to capture the relevant geometric properties of cortical folding. In this paper, we propose an entirely novel, data-driven deep-learning based method to analyze the brain's shape that eliminates this reliance on manual feature definition. This method builds on the emerging field of geometric deep-learning and uses traditional convolutional neural network architecture uniquely adapted to the surface representation of the cortical ribbon. This method is a complete departure from prior brain MRI CNN investigations, all of which have relied on three dimensional MRI data and interpreted features of the MRI signal for prediction. MRI data from 6410 healthy subjects obtained from 11 publicly available data repositories were used for analysis. Ages ranged from 6 to 89 years. Both inner and outer cortical surfaces were extracted using Freesurfer and then registered into MNI space. For purposes of method development, both a classification and regression challenge were introduced for network learning including sex and age prediction, respectively. Two independent graph convolutional neural networks (gCNNs) were trained, the first of which to predict subject's self-identified sex, the second of which to predict subject's age. Class Activation Maps (CAM) and Regression Activation Maps (RAM) were constructed respectively to map the topographic distribution of the most influential brain regions involved in the decision process for each gCNN. Using this approach, the gCNN was able to predict a subject's sex with an average accuracy of 87.99 % and achieved a Person's coefficient of correlation of 0.93 with an average absolute error 4.58 years when predicting a subject's age. We believe this shape-based convolutional classifier offers a novel, data-driven approach to define biomedically relevant features from the brain at both the population and single subject levels and therefore lays a critical foundation for future precision medicine applications.

Extremity bone sarcoma: Evolution from primary amputation to limb salvage surgery (LSS).

e22012 Background: Primary Extremity Bone Sarcomas (PEBS) were historically managed by upfront amputation. Early success rates utilizing neoadjuvant chemotherapy followed by LSS showed promise in halting disease progression and achieving local control. This prompted surgical oncologists to consider LSS in more patients. In anatomic locations for which no functional reconstruction exists, amputation is still considered the best option. Oncologic control, patient satisfaction and improved function are the primary goals of LSS. Three Dimensional CT and MRI have made for more accurate surgical planning, ensuring adequate margins and improved local control. Methods: Between January 1983 and December 2019 the senior author (BNR) has performed, or mentored over 590 LSS worldwide.Improvement in imaging modalities has been the most important advancement contributing to decreasing tumor margins. Initially bone scans determined the margins at 7cm. In the 1970s CT became the modality used to determine local tumor extent. The resection margins were reduced to 5 cm. The early 90’s saw MRI as the main imaging modality. Margins are now set at 1-2cm.LSS was initiated in the mid 80’s using the strict criteria of age > 13, no metastasis, intramedullary extension less than 50%, and small extraosseus extent. Results: Of the 134 cases enrolled through 1990, the local control rate was 90%. Between July 1991 through December 2019, 388 patients enrolled in bone sarcoma protocols, 354 patients underwent LSS ( > 95%). The primary site was the femur (171 cases, 43%) Histological Diagnosis was OS in 93%. Ages ranged from 3-25 years (median 10.8 years) The average length of tumor resection was 7cms(range 1-38cms). 5 patients presented with pathological fracture. All successfully underwent LSS without an increase in local recurrence. Intraoperative complications included, excessive hemorrhage defined as > 10%of blood volume in 88 patients, 25%, Vascular injuries requiring repair in 9 patients, post operative neuropraxia in 28 patients. Mid and late term complications included, Superficial wound infection in 45 patients (19 requiring some surgical intervention), Deep wound problems were identified in 11 patients ( 3 required subsequent amputation), 9 patients had their prosthesis revised due to refractory infection, Local relapse in 5 patients , < 1%. Long term implant related complications, loosening of the prosthesis in 16 , Stem fracture in 13, nonunion of the allograft in 3 patients. Conclusions: Disadvantages are the need for repeat surgeries consequent to either growth discrepancy, loosening, or implant fracture.LSS is feasible in over 95% of patients, complications are generally manageable. Successful oncologic results, satisfaction and function are readily obtained. Early rehabilitation is key to functional success.

An Analysis of Deep Learning Techniques in Neuroimaging

Deep learning is a machine learning technique that has demonstrated better results and performance when compared to standard machine learning algorithms in relation to higher dimensional MRI brain imaging data. The applications of deep learning in the clinical domain are discussed in this study. A detailed analysis of several deep learning algorithms for the Alzheimer's disease diagnosis is analyzed, in which this disorder of brain that gradually spreads and destroys memory of the brain, and it is a typical disorder in elderly individuals due to dementia. When it comes to brain image processing, the most commonly used and represented method, according to most research publications, is Convolutional Neural Networks (CNN). Following a review of many relevant studies for the Alzheimer's disease diagnosis, it was shown that utilizing advanced deep learning algorithms in different datasets (OASIS and ADNI) combined to one can improve AD prediction at earlier stages.

Neuroimaging and neurodevelopmental outcome after early fetal growth restriction: NEUROPROJECT-FGR.

Adverse neurodevelopmental outcomes and MRI alterations are reported in infants born after fetal growth restriction (FGR). This study evaluates the additional role of FGR over prematurity in determining brain impairment. Retrospective observational study comparing 48 FGR and 36 appropriate for gestational age infants born between 26 and 32 weeks' gestation who underwent a cerebral MRI at term equivalent age. Exclusion criteria were twins, congenital anomalies, and findings of overt brain lesions. Main outcomes were total maturation score (TMS) and cerebral areas independently measured by two neuro-radiologists and Griffiths or Bayley scale III scores at median age of 2 years. TMS was not significantly different between the groups. Inner calvarium and parenchyma's areas were significantly smaller in FGR cases. There were no significant differences in the average quotient scores. A positive correlation between parenchyma area and cognitive score was found (r = 0.372, p = 0.0078) and confirmed after adjusting for sex, gestational age, and birth weight (p = 0.0014). Among FGR, the subgroup with umbilical arterial Doppler velocimetry alterations had significantly worse gross motor scores (p = 0.005). FGR plays additional role over prematurity in determining brain impairment. An early structural dimensional MRI evaluation may identify infants who are at higher risk. Fetal growth-restricted infants showed smaller cerebral parenchymal areas than preterm controls. There is a positive correlation between the parenchyma area and the cognitive score. These results highlight the already known link between structure and function and add importance to the role of a structural dimensional MRI evaluation even in the absence of overt brain lesions.

Systematic review of radiological cervical foraminal grading systems.

The study design of this paper is systematic review. The purpose of this review is to evaluate the existing radiological grading systems that are used to assess cervical foraminal stenosis. The importance of imaging the cervical spine using CT or MRI in evaluating cervical foraminal stenosis is widely accepted; however, there is no consensus for standardized methodology to assess the compression of the cervical nerve roots. A systematic search of Ovid Medline databases, Embase 1947 to present, Cinahl, Web of Science, Cochrane Library, ISRCTN and WHO international clinical trials was performed for reports of cervical foraminal stenosis published before 01 February 2020. In collaboration with the University of Leeds, a search strategy was developed. A total of 6952 articles were identified with 59 included. Most of the reports involved multiple imaging modalities with standard axial and sagittal imaging used most. The grading themes that came from this systematic review show that the most mature for cervical foraminal stenosis is described by (Kim et al. Korean J Radiol 16:1294, 2015) and (Park et al. Br J Radiol 86:20120515, 2013). Imaging of the cervical nerve root canals is mostly performed using MRI and is reported using subjective terminology. The Park, Kim and Modified Kim systems for classifying the degree of stenosis of the nerve root canal have been described. Clinical application of these scoring systems is limited by their reliance on nonstandard imaging (Park), limited validation against clinical symptoms and surgical outcome data. Oblique fine cut images derived from three dimensional MRI datasets may yield more consistency, better clinical correlation, enhanced surgical decision-making and outcomes.

High-Resolution Three Dimensional MRI Findings After Plugging Surgery for Superior Semicircular Canal Dehiscence: A Case Report

: Superior semicircular canal dehiscence (SSCD) is known as abnormal communication of the superior semicircular canal (SCC) to the intracranial space secondary to a bony defect in the canal. Patients who are subjected to surgical repair usually have intractable symptoms, and recently, plugging of SCC using a transmastoid approach has been widely recommended. In this report, we describe a case of incomplete plugging for SSCD in a 37-year-old woman, along with the high-resolution three dimensional magnetic resonance imaging (3D MRI) findings using Pöschl view reconstruction. Postoperative MRI of 3D T2-wieghted sampling perfection with application optimized contrasts using different flip angle evolution (SPACE) Pöschl plane demonstrated an incomplete plugging of the SCC with partially visible perilymphatic fluid in the posterior limb above the common crus. A 3D fluid-attenuated inversion recovery (FLAIR) sequence showed an enhancement involving the vestibule and SCC, suggesting labyrinthitis. Although there are few reports about incomplete plugging for SSCD, this case could demonstrate postoperative status and complication after plugging of SSCD using a high-resolution 3D MRI sequences with Pöschl view reconstruction.

A novel approach for pre-surgical evaluation of facial morphometry in children using magnetic resonance imaging scans of the brain: A feasibility analysis

Pre-surgical evaluation of facial morphometry is frequently warranted for children with facial dysmorphism. Though many methods utilized previously for such purposes, data is scarce on using magnetic resonance (MRI) brain images for such purposes. The purpose of this study was to appraise the feasibility of utilizing MRI brain scans done in epilepsy imaging protocol to assess facial morphometry. Measurements of the face; orbit, mouth, and nose of children aged 1 to 7 years were obtained using T1 sagittal, T2 axial and three dimensional (3D) MRI images of the brain (n=20). Ability to obtain facial measurements, inter and intra-observer variability calculated. The mean age of the studied children was 4±2 years, of which 40% (n=8) were boys, and 60% (n=12) were girls. Obtaining facial measurements were reliable with high intra-observer (α=0.757 to 0.999) and inter-observer agreements (α=0.823 to 0.997). The landmarks of the cranium, upper face, and upper nose could be identified (100%) in both two dimensional (2D) and 3D images when such landmarks were contained in the imaging field of view (FOV). Landmarks of lower nose, (subalar width = 0%) or mouth (0%) were not contained in the FOV of 2D images, but contained in 3D images (100%). Both 2D and 3D images did not allow assessment of lower face or the mandible as such landmarks were not contained in the FOV.We conclude thatBrain MRIs performed to evaluate cerebral pathology can be executed to assess facial measurements, provided the FOV of the scan is adjusted to include all significant landmarks.

Delayed contrast dynamics as marker of regional impairment in pulmonary fibrosis using 5D MRI - a pilot study

To analyse delayed contrast dynamics of fibrotic lesions in interstitial lung disease (ILD) using five dimensional (5D) MRI and to correlate contrast dynamics with disease severity. 20 patients (mean age: 71 years; M:F, 13:7), with chronic fibrosing ILD: n = 12 idiopathic pulmonary fibrosis (IPF) and n = 8 non-IPF, underwent thin-section multislice CT as part of the standard diagnostic workup and additionally MRI of the lung. 2 min after contrast injection, a radial gradient echo sequence with golden-angle spacing was acquired during 5 min of free-breathing, followed by 5D image reconstruction. Disease was categorized as severe or non-severe according to CT morphological regional severity. For each patient, 10 lesions were analysed. IPF lesions showed later peak enhancement compared to non-IPF (severe: p = 0.01, non-severe: p = 0.003). Severe lesions showed later peak enhancement compared to non-severe lesions, in non-IPF (p = 0.04), but not in IPF (p = 0.35). There was a tendency towards higher accumulation and washout rates in IPF compared to non-IPF in non-severe disease. Severe lesions had lower washout rate than non-severe ones in both IPF (p = 0.003) and non-IPF (p = 0.005). Continuous contrast agent accumulation, without washout, was found only in IPF lesions. Contrast agent dynamics are influenced by type and severity of pulmonary fibrosis, which might enable a more thorough characterisation of disease burden. The regional impairment is of particular interest in the context of antifibrotic treatments and was characterised using a non-invasive, non-irradiating, free-breathing method. Delayed contrast enhancement patterns allow the assessment of regional lung impairment which could represent different disease stages or phenotypes in ILD.

Influence of the hypothalamus–pituitary–gonadal axis reactivation and pubertal hormones on gray matter volume in early pubertal girls

Objective Puberty is a sensitive period of brain development accompany with pubertal hormones fluctuation. However, the underlying mechanisms of the impact of hypothalamus–pituitary–gonadal (HPG) axis reactivation and associated elevated pubertal hormones on brain structure are still unclear. Here, we investigated the brain structure differences between girls with and without HPG axis reactivation and the influence of pubertal hormones on these brain regions. Methods 126 girls aged 8–9.5 years underwent a gonadotropin-releasing hormone (GnRH) stimulation test to identify the HPG axis status and categorized into HPG+ group (n = 80) and HPG- group (n = 46). T1-weighted gradient echo three dimensional MRI was performed using a 3.0-Tesla scanner to assess the difference in GMV between the two groups. Correlation analyses were conducted to explore the relations between the brain regions showing significant GMV differences and serum hormone concentrations. Result The HPG+ group showed significantly higher GMV in the bilateral lingual gyrus and lower GMV within the right orbital inferior frontal gyrus compare to the HPG − group. Furthermore, GMV in the right orbital inferior frontal gyrus was positively associated with plasma concentrations of follicle stimulating hormone (FSH) in HPG+ group. Conclusion The present study suggests that the reactivated HPG axis could affects regional structural brain changes in early pubertal girls. FSH production play an important role in bilateral lingual gyrus, which are involved in vision processing, semantic processing and emotional expression.

Statistical analysis of spinal cord injury severity detection on high dimensional MRI data

Staggered Segmenting on the programmed spinal rope form is a vital advance for evaluating spinal line decay in different infections. Outlining dark issue (GM) and white issue (WM) is additionally helpful for measuring GM decay or for extricating multiparametric MRI measurements into WMs tracts. Spinal line division in clinical research isn't as created as cerebrum division, anyway with the considerable change of MR groupings adjusted to spinal line MR examinations, the field of spinal rope MR division has progressed extraordinarily inside the most recent decade. Division strategies with variable exactness and level of multifaceted nature have been produced. In this paper, we talked about a portion of the current strategies for line and WM/GM division, including power based, surface-based, and picture based and staggered based techniques. We likewise give suggestions to approving spinal rope division systems, as it is essential to comprehend the inborn qualities of the strategies and to assess their execution and constraints. In conclusion, we represent a few applications in the solid and neurotic spinal string. In this task, an Automatic Spinal Cord Injury (SCI) is identified utilizing a staggered division technique.

Complete 3 dimensional reconstruction of parturient pelvic floor

IntroductionThe women pelvic floor is a complex system, which seems to endure several modifications during pregnancy and childbirth. Our primary purpose was to build an extensive 3 dimensional (3D) numerical anatomical model of the women pelvic floor. MethodsFirst, the role and the location of each organ, muscle, or ligament, were identified through an extensive literature review. Then, different entities were selected because of their visibility and importance in the pelvic floor. Each entity was identified using anatomical knowledge, and outlined on 2 dimensional (2D) MRI images, that were carried out on 4 pregnant women, using sequences T1, T2 and proton density weighted, through AVIZO program. The overlay of these 2D outlines produced a 3D geometrical reconstruction, which was then reworked with the program CATIA to obtain a usable geometric model. ResultsWe identified and integrated 15 anatomical structures to the geometrical model, including organs, ligament and muscles from the pelvis and perineum. This geometrical model allowed us to obtain a visual interactive representation with 3D images. These different steps resulted in the creation of a complete numerical model of the female pelvic floor, which might be used in Finite Element simulation. ConclusionA new complete and accurate 3D numerical anatomical model of the women pelvic floor was elaborated. It presents simultaneously analytical prospects, through the observation of the strains and deformations that are imposed on the different structures, and educational prospects, through the detailed visual representation of several situations.

Clinical 2 years results after implantation of patient-specific mini-metal implants in middle-aged patients alfter previous frustrane cartilage surgery

Aims and Objectives: The treatment of focal (osteo-)chondral lesions of the knee in middle-aged patients with previous frustane cartilage regeneration is challenging. This prospective study presents clinical 2 years results after implantation of patient-specific mini-metal implants. Materials and Methods: Based on a three dimensional MRI planing 7 patient-specific mini-metal implants (Episealer, Episurf, Stockholm, Sweden) were produced for 4 patients (2 men, 2 women). In one patient a single implant was implanted to the trochlea, the other 3 patients received each a trochlea and a medial condyle implant. Corresponding cartilage lesions in the tibia plateau or retropatellar surface were excluded by a detailed specific MRI resulting in a so called “damage report”. MRI data was used to manufacture patient-specific implants and guide instruments, to fit the unique anatomy of each individual knee. Demographic, operative and clinical scores (VAS of pain and patient satisfaction, KOOS and Forgotten Joint Score) were collected preop and at 6 months, 1 and 2 years postoperatively. Results: Mean age was 51.5 (47-57) years. All patients had failed previous cartilage surgery. All mean KOOS domain scores were improved at 1 and 2 years. Mean preoperative aggregated KOOS (35) improved to 60 at 12 months and remained at 60 also at 24 months. Mean VAS score (pain) improved from 7.4 preoperatively to 2.1 at 24 months, mean VAS score (patient satisfaction) improved from 3.1 to 8.5 postoperatively at 24 months. The forgotten joint score reached a mean of 24 points. No revisions or complications were detected. Conclusion: Despite a very small patient number and leck of statistical power the new patient-spedific mini-metal implant seems to be a good options fort he treatment of middle-aged patients who already underwent frustane cartilage surgery. In these relatively rare caes early unicompartimental arthroplasty can be prevented. Further clinical studies with larger patient collectives have to prove the effectiveness of this new technique.

Clinical Results of an indivisualised mini-metal implant for Cartilage lesions in the knee

Aims and Objectives: We report prospective detailed results of patients undergoing treatment for chondral defects usingEpisealer, a patient-specific 2nd generation individualized mini-metal implant, based on a three dimensional MRI planing. Materials and Methods: Analysis of consecutive patients from 13 surgeons. Total of 98 patients, of which 59 patients (25 men,34 women) with focal cartilage lesions ICRS grade 3 or 4 underwent partial resurfacing, 48 on medial condyle, 4 lateral condyle and 7 on trochlea. Mean age 48 (27-67) years, mean BMI 28 (21-41). 64% had failed previous cartilage surgery. Detailed specific MRI data was used to manufacture patient-specific implants and guide instruments, to fit the unique anatomy of each individual knee. Implants were uncemented and made of cobalt-chrome, double coated with hydroxyapatite on top of titanium. Demographic, operative and clinical scores (VAS and KOOS) were collected preop and at 6 months, 1 and 2 years postoperatively. Results: All mean KOOS domain scores were significantly improved at 1 and 2 years (p<0.05). Mean preoperative aggregated KOOS (37) improved to 62 at 12 months and remained at 62 also at 24 months (n=32), (p<0.05). Mean VAS score improved from 65 preoperatively to 37 at 24 months. All mean KOOS domain scores were significantly improved at 1 and 2 years (p<0.05). 2 patients (3.4%) underwent revision (at 9 months for infection and at 30 months for progression of arthritis). Conclusion: The study shows excellent early clinical results in the treatment of focal symptomatic cartilage lesions on the femoral condyles or trochlea with a second-generation patient specific metal implant and instruments. Adherence to strict indications has allowed for high patient reported scores and low early revision rate.

Review on 2D and 3D MRI Image Segmentation Techniques

Magnetic Resonance Imaging is most widely used for early diagnosis of abnormalities in human organs. Due to the technical advancement in digital image processing, automatic computer aided medical image segmentation has been widely used in medical diagnostics. Image segmentation is an image processing technique which is used for extracting image features, searching and mining the medical image records for better and accurate medical diagnostics. Commonly used segmentation techniques are threshold based image segmentation, clustering based image segmentation, edge based image segmentation, region based image segmentation, atlas based image segmentation, and artificial neural network based image segmentation. This survey aims at providing an insight about different 2-Dimensional and 3- Dimensional MRI image segmentation techniques and to facilitate better understanding to the people who are new in this field. This comparative study summarizes the benefits and limitations of various segmentation techniques.

Brain Tumor Levels Detection in Three Dimensional MRI using Machine Learning and MapReduce

Brain Tumor Levels Detection in Three Dimensional MRI using Machine Learning and MapReduce

Diagnostic accuracy of 3.0T high-resolution MRI for assessment mesorectal lymph node metastases in patients with rectal cancer

To evaluate the diagnostic value of 3.0T high-resolution MRI in mesorectal lymph node metastasis of rectal cancer. The images and postoperative pathological data of patients with pathologically diagnosed rectal cancer who underwent prospective 3.0T two dimensional high-resolution MRI rectal examinations and surgery within two weeks after MRI examination at the First Affiliated Hospital, Sun Yat-sen University from November 2015 to November 2016 were retrospectively collected. Patients who received preoperative neoadjuvant therapy and those who did not undergo operation after MRI examination were excluded. The MRI sequences included high-resolution sagittal, coronal and oblique axial T2 weighted image (T2WI) (repetition time/echo time, 3000-4000 ms/77-87 ms; slice thickness/gap, 3 mm/0 mm; field of view, 18-22 cm). Two abdominal MRI radiologists independently assessed the morphology, margin, signal of all visible mesorectal nodes, measured their minor axes (three times for each radiologist) and gave estimation of the malignancy. The criteria of metastatic nodes on high-resolution MRI T2WI were nodes with irregular shape, ill-defined border and/or heterogeneous signal. The results of MRI diagnosis were compared with postoperative pathology. The sensitivity, specificity, accuracy, positive predictive value(PPV) and negative predictive value(NPV) of mesorectal nodes and nodes with different short-axis diameter ranges were calculated to evaluate the diagnostic efficiency of high-resolution MRI. Kappa statistics was used to evaluate the agreement for per node and for per patient between high-resolution MRI and pathological results. A Kappa value of 0-0.20 indicated poor agreement; 0.21-0.40 fair agreement; 0.41-0.60 moderate agreement; 0.61-0.80 good agreement; and 0.81-1.00 excellent agreement. A total of 81 patients were enrolled in the retrospective cohort study, including 50 males and 31 females with age of (59.3±11.1) years. Histopathology showed 1 case of well differentiated adenocarcinoma, 63 of moderately differentiated adenocarcinoma, 9 of moderately to poorly differentiated adenocarcinoma, 2 of poorly differentiated adenocarcinoma, 3 of mucinous adenocarcinoma and 3 of tubulovillous adenocarcinoma. Histopathological staging showed 2 cases in T1 stage, 20 in T2 stage, 45 in T3 stage and 14 in T4 stage; 34 in N0 stage, 40 in N1 stage and 7 in N2 stage; 76 in M0 stage and 5 in M1 stage. A total of 377 nodes were included in the node-by-node evaluation, of which 168 (44.6%) nodes were metastatic from 58.0% (47/81) patients. The median short-axis diameter was 5.4(2.4-18.6) mm in metastatic nodes, which was significantly larger than 3.8 (2.0-8.7) mm in non-metastatic nodes[Z=10.586, P=0.000]. The sensitivity, specificity, accuracy, PPV and NPV were 74.4% (125/168), 94.7% (198/209), 85.7% (323/377), 91.9% (125/136) and 82.2% (198/241), respectively. The Kappa values between high-resolution MRI and histopathological diagnosis for node-by-node and patient-by-patient were 0.71 and 0.70 respectively, indicating good agreements. Fourteen nodes >10 mm were all metastatic. The results of high-resolution MRI for nodal status were consistent with the results of histopathological diagnosis, and the sensitivity, accuracy and PPV were all 100.0%. Among 124 nodes with short-axis diameter of 5-10 mm, 95 (76.6%) were metastatic, and the sensitivity, specificity, accuracy, PPV and NPV were 78.9% (75/95), 86.2% (25/29), 80.6% (100/124), 94.9% (75/79) and 55.6% (25/45), respectively. The agreement was fair (Kappa value 0.55) between high-resolution MRI and histopathological diagnosis. Among 239 nodes with short-axis diameter ≤5 mm, 59(24.7%) were metastatic, and the sensitivity, specificity, accuracy, PPV and NPV were 61.0% (36/59), 96.1%(173/180), 87.4%(209/239), 83.7%(36/43) and 88.3%(173/196), respectively. The agreement was good (Kappa value 0.63) between high-resolution MRI and histopathological diagnosis. Rectal high-resolution MRI has good diagnostic value for estimating metastatic mesorectal nodes by evaluating the morphology, margin and signal of nodes.

Development of 3D Printed Brain Tumour for Surgical Biopsy

The traditional treatment for brain cancer includes surgery, radiotherapy and chemotherapy. The neurosurgeon usually plans and performs the surgery with the help of two dimensional MRI images. If the size and the exact location of tumour is not clear it is difficult to perform a safe surgery and sometimes may cause permanent damage to the normal tissues. Such problems could be minimized by having first hand information regarding the tumour and hence an attempt is made in the present research to create 3D printed models of brain tumours using a medical image processing software based on MRI datasets. The MRI data saved as DICOM (Digital Imaging and Communications in Medicine) would be integrated with 3D slicer. 3D printer produces a physical 3D models using STL file generated from 3D slicer. Thus the 3D printed model of Brain Tumour could be served as a tool for preoperative surgical planning.