Search

Bone identification and segmentation in X-ray images are crucial in orthopedics for the automation of clinical procedures, but it often involves some manual operations. In this work, using a modified SegNet neural network, we automatically identify and segment lower limb bone structures on radiographs presenting various fields of view and different patient orientations.A wide contextual neural network architecture is proposed to perform a high-quality pixel-wise semantic segmentation on X-ray images presenting structures with a similar appearance and strong superposition. The proposed architecture is based on the premise that every output pixel on the label map has a wide receptive field. This allows the network to capture both global and local contextual information. The overlapping between structures is handled with additional labels.The proposed approach was evaluated on a test dataset composed of 70 radiographs with entire and partial bones. We obtained an average detection rate of 98.00% and an average Dice coefficient of 95.25 ± 9.02% across all classes. For the challenging subset of images with high superposition, we obtained an average detection rate of 96.36% and an average Dice coefficient of 93.81 ± 10.03% across all classes.The results show the effectiveness of the proposed approach in segmenting and identifying lower limb bone structures and overlapping structures in radiographs with strong bone superposition and highly variable configurations, as well as in radiographs containing only small pieces of bone structures.

BACKGROUND In glioblastomas (GBM) patients the first post-radiation MRI is usually difficult to interpret given the post-radiotherapy enhancement and possible pseudo-progression which is present in almost 50% of the patients. METHODS We retrospectively analyzed all patients with GBM treated between 2006 and 2017 at the CHUM (SARDO database). If the first brain MRI done within 3 months after the systemic treatment showed progression of contrast, these patients were considered pseudoprogressors (PsP) while the patients who had stable or response to treatment were the non-progressors (nP). If progression persisted in subsequent MRI with a change of treatment within 6 months, they were considered early progressors (eP). If subsequent MRI improved or was stable, they were classified as nP (or true pseudo-progression). RESULTS In our cohort of 470 patients with GBM, 57.7% were nP and 42.3% were PsP after the first post-treatment imagery. The median follow-up was 10 months. The nP had a longer mOS 15.3m vs 11.3m, p < 0.001, regardless of subsequent evolution. After the second assessment, 67.8% of PsP patients were then considered as eP and 36.4% of nP patients also progressed within 6 months. The nP either after the first or second evaluation had the same mOS (19.9m vs 18.3m), just like the eP (9.3m vs 8.6m), independently of the subsequent treatment. No demographic, molecular or clinical factor predicted PsP, except for tumor size (> 5cm, p=0.024). PsP incidence was similar between 2006–2011 (PsP 57.8%) and 2012–2017 (42.2%). The 1y OS with pseudo progression at the first MRI was 39.7% vs 54.8% with no progression (p=0.001) which has a meaningful impact for the patient. CONCLUSION Pseudo-progression is frequent (42%) in glioblastoma and predicts a poorer prognosis with 1y OS of 39,7%. In fact, PsP patients have more than two-thirds chance to progress precociously.

BACKGROUND Before 2012, Temozolomide (TMZ) was used for low-grade gliomas (LGG) to avoid Lomustine toxicity. After 2012 RTOG data, Procarbazine, Lomustine and Vincristine (PCV) given sequentially with radiotherapy became standard treatment with the side effects burden associated. METHODS We retrospectively reviewed our SARDO clinical database of patients with low-grade glioma LGG grades 2 and 3 between 2006 and 2017 at CHUM University Health Center. Molecular profile for these tumors is reflex and standard in our institution since 2016. RESULTS A total of 123 patients were identified with grade 2 and grade 3 LGG; 37 (30%) treated with PCV and 86 (70%) received TMZ. Median follow up was 11mo for PCV 11mo vs 22mo TMZ. Both groups were balanced in terms of median age, sex, neurologic symptoms and surgery rate. 53% patients had tumor untested for IDH1-2 and codeletion1p19q because of diagnostic before 2016. Disparities were noted with a predominance of grade 3 in the TMZ group (74% vs 27%, p< 0.01). TMZ was the preferred regimen before 2012 (100% vs 43%) and PCV became the standard of care after 2012 (0% vs 57%). Radiation use as first line treatment was 90%. The 4y OS was not significantly longer for PCV 50% compared with TMZ 47% with mOS between both groups (PCV NR vs TMZ 39.9mo, p=0.158). When controlled for tumor grade, the 2y OS was 80% for PCV vs 64% for TMZ,p=0.542. The 4y PFS was trendly longer for PCV group 78% than TMZ group 45%, p=0.148. CONCLUSION As we are still waiting for the prospective ongoing prospective trials comparing these 2 regimens, PCV and radiation are still standard of care regimens for grade 2 and 3 LGG. This retrospective data is not reassuring for a replacement for PCV with TMZ.

Purpose: The purpose of the study was to determine the extent of displacement of the pulmonary vein antrums resulting from the intrinsic motion of the heart using 4D cardiac dual-source computed tomography (DSCT). Methods: Ten consecutive female patients were enrolled in this prospective planning study. In breath-hold, a contrast-injected cardiac 4-dimensional (4D) computed tomography (CT) synchronized to the electrocardiogram was obtained using a prospective sequential acquisition method including the extreme phases of systole and diastole. Right and left atrial fibrillation target volumes (CTVR and CTVL) were defined, with each target volume containing the antral regions of the superior and inferior pulmonary veins. Four points of interest were used as surrogates for the right superior and inferior pulmonary vein antrum (RSPVA and RIPVA) and the left superior and inferior pulmonary vein antrum (LSPVA and LIPVA). On our 4D post-processing workstation (MIM Maestro™, MIM Software Inc.), maximum displacement of each point of interest from diastole to systole was measured in the mediolateral (ML), anteroposterior (AP), and superoinferior (SI) directions. Results: Median age of the enrolled patients was 60 years (range, 56-71 years). Within the CTVR, the mean displacements of the superior and inferior surrogates were 3 mm vs. 1 mm (p=0.002), 2 mm vs. 0 mm (p= 0.001), and 3 mm vs. 0 mm (p=0.00001), in the ML, AP, and SI directions, respectively. On the left, mean absolute displacements of the LSPVA vs. LIPVA were similar at 4 mm vs. 1 mm (p=0.0008), 2 mm vs. 0 mm (p= 0.001), and 3 mm vs. 1 mm (p=0.00001) in the ML, AP, and SI directions. Conclusion: When isolated from breathing, cardiac contraction is associated with minimal inferior pulmonary veins motion and modest (1-6 mm) motion of the superior veins. Target deformation was thus of a magnitude similar or greater than target motion, limiting the potential gains of cardiac tracking. Optimal strategies for cardiac radiosurgery should thus either incorporate the generation of an internal target or cardiac gating. In either case, cardiac 4D DSCT would allow for personalized margin definition.

La commercialisation de la, communication de masse telle que nous la connaissons au Canada et aux États-Unis superpose à la relation primitive de communication (source —» message —» récepteur) un autre système de relations dans lequel le diffuseur vend un auditoire à un commanditaire. Ce second système n'est pas sans avoir de profondes influences sur le premier. Ce sont ces conséquences que nous voudrions examiner ici brièvement.

The objective of this study was to determine the immediate postoperative effect of 2 corrective operations for cranial cruciate ligament (CCL)-deficient stifle by evaluating 3-dimensional (3-D) stifle kinematics. Ten hindlimbs from large-breed canine cadavers were used. Range of motion was induced by applying 100 N of traction on the quadriceps tendon and recorded with electromagnetic movement sensors for each situation: intact stifle (control), CCL-sectioned stifle, and surgical correction of the sectioned ligament with the modified retinacular imbrication technique (MRIT) and then with a tibial plateau leveling osteotomy — Montavon (TPLO-M). The results for the experimental situations were compared with the results for the control situation by 1-way repeated-measures analysis of variance and with each other by post-hoc analysis with the least-significant-difference method. Range of motion was significantly decreased by MRIT as compared with the other situations. Normal cranial tibial translation was restored after MRIT, whereas TPLO-M resulted in significant caudal translation. A significant increase in external rotation was observed after both MRIT and TPLO-M. A significant increase in tibial adduction throughout the range of motion was observed with TPLO-M, whereas a significant increase in tibial abduction was observed after MRIT. This study allowed us to better understand objectively the effects on 3-D canine stifle kinematics of MRIT and TPLO-M. We suggest that this type of in vitro study would be useful to evaluate established and upcoming surgical techniques and potentially improve corrective surgery.

Type 2 diabetes is a heterogeneous disease characterized by hyperglycemia and insulin resistance in peripheral tissues such as adipose tissue and skeletal muscle. This review focuses on obesity as one of the major environmental factors contributing to the development of diabetes. It has become evident that adipose tissue represents an active secretory organ capable of releasing a variety of cytokines such as TNFalpha, IL-6, adiponectin and other still unknown factors that might constitute the missing link between adipose tissue and insulin resistance. In fact, adipocyte-derived factors are significantly increased in obesity and represent good predictors of the development of type 2 diabetes. The negative crosstalk between adipocytes and skeletal muscle cells leads to disturbances in muscle cell insulin signalling and insulin resistance involving major pathways in inflammation, cellular stress and mitogenesis. Positive regulators of insulin sensitivity include the adipocyte hormone adiponectin and inhibitors of inflammatory pathways such as JNK-, IKK- and ERK-inhibitors. In summary, a better knowledge of intracellular and intercellular mechanisms by which adipose tissue affects skeletal muscle cell physiology may help to develop new strategies for diabetes treatment.

Ultrasound 3D systems are being developed to potentially replace traditional 2D equipments in the diagnosis of vascular disease. Ultrasound is non-invasive, low-cost, and nonionizing. To allow the 3D reconstruction of ultrasound images, the probe position must be known at all time. Freehand tracking devices were first considered but they lacked constant precision. Robots are now being developed to perform the ultrasound scan, allowing a constant precision, a simple manipulation, and standardization. Researchers are currently experimenting semiautomated robot scans, for facilitating further the examination. Different applications call for different robot architectures. Peripheral arterial disease has not yet been considered for automation probably because lower limb vessel examination requires long acquisitions. This paper presents a new robot for an ultrasound robot design that addresses this problem. This parallel robot will be able to perform multiple ultrasound 3D examinations on a human patient.

We propose a wavelet multi-resolution analysis to localize specific features in both lateral and frontal radiographs. This analysis allows an elegant spectral investigation that leads simultaneously to image de-noising and edge extraction. It is combined with an a priori knowledge of the spine's morphology and a 3D spline curve characterization of its global shape. Actual work deals with identifying the contours of the vertebral bodies and the localization of vertebrae's endplates. However, this information could also lead to the selection of a 3D statistical model of the spine suited for the studied deformation. Working with retro-projections of the model, we aim at creating edge models for each vertebra that will be used to geometrically match the wavelet's edges. The manual feature identification could then be replaced in the reconstruction of the 3D representation of the spine.

Idiopathic scoliosis is a 3-D deformation affecting the position of the spine in space. The regional deformity has been studied extensively but the local changes have not been widely investigated and this being mainly due to the rarity of anatomical specimens. The objective of this study was to identify a deformation pattern for idiopathic scoliosis. We thus studied one complete scoliotic specimen using a digitizing protocol developed by our research group. The anatomical specimen was selected from the Hamann-Todd Osteology Collection at the Cleveland Natural History Museum, which contains over 1,300 skeletons. We were also able to match this scoliotic specimen with one normal specimen for age, sex, race, height and weight. Each vertebra was measured by taking approximately 200 points on each surface. Parameters for each vertebra were then calculated from these sets of points. Each scoliotic vertebra was then compared with a corresponding normal vertebra of the matched specimen. We present the first findings of these measurements, which show pedicle and posterior elements changes that are thought to be secondary to the scoliotic deformation.

A peroperative measurement technique based on magnetic fields was used to document the thoracic and lumbar vertebral changes induced by derotation of the concave rod during corrective surgery of idiopathic scoliotic deformities with Cotrel-Dubousset instrumentation.Objective was to accurately document the immediate changes induced by the derotation maneuver to gain a better understanding of its effect on curve correction during a surgical procedure.Accurate peroperative documentation of these three-dimensional changes was very limited, and the exact contribution of the derotation maneuver in the entire process of correction of a scoliotic deformity was still poorly understood.A digitizer using magnetic fields was used to record the three-dimensional orientation and x, y, and z coordinates of the tip of every spinous process exposed at surgery before and after derotation of the concave rod in a group of 22 female patients with idiopathic scoliosis.A significant improvement of the scoliotic deformity was noted in the frontal plane, and improvement of thoracic hypokyphosis and lumbar hypolordosis in the sagittal plane was seen. Vertebral axial rotation remained unchanged.The derotation maneuver is effective in achieving three-dimensional correction of idiopathic scoliosis, but vertebral axial derotation is not an important component of this correction.

Le dictionnaire Robert définit la distance comme «la longueur minimum qui sépare une chose d'une autre». On dira ainsi que la ville de Québec est à 150 milles de Montréal ou à 3500 milles de San Francisco. C'est là cependant le point de vue géographique ordinaire, qui décrit la terre en parallèles et en méridiens, en mettant tous les points du globe sur le même pied. Nous proposons de parler de distance relationnelle pour introduire dans la distance un certain point de vue sociologique. Le sociologue étudie entre autres choses les relations entre les hommes, entre les diverses entités sociales. L'auteur parle de «distance relationnelle» pour tenir compte de la structure des relations sociales dans l'évaluation de la distance». Il essaie ensuite de voir comment cette notion peut s'appliquer à certaines situations., de Guise Jacques. La distance relationnelle : réflexion sur les média comme outils de relation sociale. In: Communication Information, volume 2 n°1, hiver 1977. pp. 31-49.

The objective of this paper is to compare the performance of conventional FFT-based (basic periodogram and Welch's method) and modern parametric (all-pole and pole-zero modeling) methods in estimating the spectral distribution of cardiac bioprosthetic valve sounds, and for the extraction of the two most dominant frequency peaks (DFP). These methods were tested for stability by adding random noise and truncating the bioprosthetic valve closing sounds, and for reproducibility by measuring the variance of the spectra obtained from three consecutive recordings of each patient. Results from a group of 11 patients show that the basic periodogram and Steiglitz-McBride's method with maximum entropy (pole-zero modeling) provide the most consistent (minimal variance) estimates of the DFP's of the closing sounds. However, for estimating spectral distributions, the most stable methods appear to be the basic periodogram and Steiglitz-McBride's method with extrapolation to zero. The basic periodogram appears to be the best compromise to estimate both the spectral distribution and the DFP's of the bioprosthetic closing sounds.

Les mass média prennent une part de plus en plus grande dans la production et la diffusion des œuvres culturelles et de l'information dans notre société. Les ressources financières qui supportent ces média deviennent alors, pour une part de plus en plus importante, les ressources même de la culture. Dans notre système actuel de diffusion, les média vivent d'abord et avant tout de publicité. Certes ils ne vivent pas uniquement de publicité : en radio et en télévision nous avons instauré un système où l'État intervient dans le financement des stations ; dans les journaux, le public lecteur fournit directement à peu près 35% des recettes brutes du médium qu'il achète. Il n'en demeure pas moins que ce sont les commanditaires qui financent la part la plus importante de nos média, au point que sans eux l'existence même de ces derniers serait compromise. Ce sont ces ressources publicitaires qui font l'objet de notre analyse. Nous essayerons de voir quelle est la part faite aux principaux média francophones et anglophones du Canada.

Sounds produced by prosthetic heart valves are known to contain diagnostic information regarding the structural and functional integrity of their components. Analog techniques for processing prosthetic valve phonocardiograms have met with limited success in extracting this information, because of their poor spectral resolution and lack of versatility. Numerical methods of signal processing overcome most of these limitations, but the need for a computer to implement numerical methods raises the question of cost-effectiveness in many applications. Numerical analysis of prosthetic valve signals has therefore received very little attention outside the academic and laboratory context. Cost reductions in computer hardware arising from the use of micro-processors, make it possible to envisage dedicated clinical instruments for processing prosthetic valve sounds in view of assessing overall valve performance and detecting component degradation at an early stage. Basic spectral considerations for the design of such instruments are discusses in this paper.

We tested a procedure to recover two moving dipole (TMD) parameters from bidipolar potential distributions generated over the surface of a numerical human torso model, using 120 surface sampling points. The surface distributions were computed for either a finite homogeneous torso (T1), a finite torso with lungs (T2), or a finite torso with lungs and blood masses (T3). Inverse calculations were carried out to initially recover the multipole series components (15, 24, or 35 terms) using either a finite homogeneous torso (I1) or a finite torso with lungs (I2), and a least-squares difference procedure. Next, the TMD parameters were obtained by fitting these multipole series components (15 or 24 terms) to the multipole series components estimated from the Brody shift equations, using the Levenberg-Marquardt iterative algorithm and a series of initial estimates for the TMD solution. A simulation run involved 253 different input dipole-pair combinations and 20 initial estimates. The correct TMD solution almost always coincided with that yielding the minimum residue, which was also the solution obtained by the largest fraction of initial estimates. The lowest rms position error of 2.7 mm was obtained with a T1-I1 torso combination, and with 35 recovered multipole series components and 24 Brody shift equations. Larger errors were obtained using a lower number of recovered multipole series components and Brody shift equations, or pairs of parallel or antiparallel dipoles, or dipoles of unequal amplitude.