Improving gender classification accuracy in the wild

[EN]In this paper, we focus on gender recognition in challenging large scale scenarios. Firstly, we review the literature results achieved for the problem in large datasets, and select the currently hardest dataset: The Images of Groups. Secondly, we study the extraction of features from the face and its local context to improve the recognition accuracy. Diff erent descriptors, resolutions and classfii ers are studied, overcoming previous literature results, reaching an accuracy of 89.8%.

Simulation Guided Navigation in cranio-maxillo-facial surgery: a new approach to improve intraoperative three-dimensional accuracy and reproducibility during surgery.

The aim of this PhD thesis " Simulation Guided Navigation in cranio- maxillo- facial surgery : a new approach to Improve intraoperative three-dimensional accuracy and reproducibility during surgery ." was at the center of its attention the various applications of a method introduced by our School in 2010 and has as its theme the increase of interest of reproducibility of surgical programs through methods that in whole or in part are using intraoperative navigation. It was introduced in Orthognathic Surgery Validation a new method for the interventions carried out according to the method Simulation Guided Navigation in facial deformities ; was then analyzed the method of three-dimensional control of the osteotomies through the use of templates and cutting of plates using the method precontoured CAD -CAM and laser sintering . It was finally proceeded to introduce the method of piezonavigated surgery in the various branches of maxillofacial surgery . These studies have been subjected to validation processes and the results are presented .

Accuracy of liver stiffness measurement using fibroscan ® to predict the response to antiviral therapy in patients with chronic hepatitis c viral infection

Introduction: Antiviral therapy can prevent disease progression in patients with chronic hepatitis C . Transient Elastografy (TE; Fibroscan) is an accurate surrogate marker to liver fibrosis, by measuring liver stiffness (LS). LS decrease has been associated with sustained virologic response (SVR). Aim: to assess the changes of LS measurments in CHC patients during and one year after Interferon (IFN)-based antiviral therapy (IFN/ribavirin) or (telaprevir+IFN/ribavirin). Methods: consecutive 69 CHC patients (53.6% females, mean age 57.9 ± 11.4) who underwent antiviral therapy for at least 20 weeks were enrolled. LS was measured using FibroScan at baseline, after three months, at the end of treatment and one year after treatment discontinuation. Fibrosis was graded using METAVIR score. Results: twenty patients treated with triple therapy and 49 with IFN/ribavirin. Fifty patients had SVR and 19 were non-responders. SVR patients: F0-F1, F2 and F3 patients (39.1%, 7.2% and 17.4%; respectively) showed no significant LS decrease (P= 0.186, 0.068 and 0.075; respectively). Conversely, in F4 patients (36.2%) LS was significantly decreased (P=0.015) after one year of treatment completion. In all patients with no SVR, no significant decrease in LS was observed. Interestingly, all Patients with F4 fibrosis (even non-responders) showed an initial significant decrease in LS (P=0.024) at 3 months after the start of treatment. However, this decrease was not predictive of SVR; area under the ROC curve 0.369 (CI %: 0.145-0.592) P= 0.265. Conclusion: Our study showed that initial decrease in LSM, especially in patients with higher baseline fibrosis score is unlikely to predict an SVR. In addition no significant association was found between clinical or virological parameters and fibrosis improvement. Further studies are needed to delineate the most appropriate clinical scenarios for the LSM by Fibroscan in chronic hepatitis C and its role in monitoring the response to antiviral treatment.

Influence of examiner's clinical experience on the reproducibility and accuracy of radiographic examination in detecting occlusal caries

The aim of this in vitro study was to assess the influence of varying examiner's clinical experience on the reproducibility and accuracy of radiographic examination for occlusal caries detection. Standardized bitewing radiographs were obtained from 166 permanent molars. Radiographic examination was performed by final-year dental students from two universities (A, n = 5; B, n = 5) and by dentists with 5 to 7 years of experience who work in two different countries (C, n = 5; D, n = 5). All examinations were repeated after 1-week interval. The teeth were histologically prepared and assessed for caries extension. For intraexaminer reproducibility, the unweighted kappa values were: A (0.11-0.40), B (0.12-0.33), C (0.47-0.58), and D (0.42-0.71). Interexaminer reproducibility statistics were computed based on means ± SD of unweighted kappa values: A (0.07 ± 0.05), B (0.12 ± 0.09), C (0.24 ± 0.08), and D (0.33 ± 0.10). Sensitivity, specificity, and accuracy were calculated at D(1) and D(3) thresholds and compared by performing McNemar test (p = 0.05). D(1) sensitivity ranged between 0.29 and 0.75 and specificity between 0.24 and 0.85. D(3) specificity was moderate to high (between 0.62 and 0.95) for all groups, with statistically significant difference between the dentists groups (C and D). Sensitivity was low to moderate (between 0.21 and 0.57) with statistically significant difference for groups B and D. Accuracy was similar for all groups (0.55). Spearman's correlations were: A (0.12), B (0.24), C (0.30), and D (0.38). In conclusion, the reproducibility of radiographic examination was influenced by the examiner's clinical experience, training, and dental education as well as the accuracy in detecting occlusal caries.

An accuracy approach to robotic microsurgery in the ear

Concerns of rising healthcare costs and the ever increasing desire to improve surgical outcome have motivated the development of a new robotic assisted surgical procedure for the implantation of artificial hearing devices (AHDs). This paper describes our efforts to enable minimally invasive, cost effective surgery for the implantation of AHDs. We approach this problem with a fundamental goal to reduce errors from every component of the surgical workflow from imaging and trajectory planning to patient tracking and robot development. These efforts were successful in reducing overall system error to a previously unattained level.

Diagnostic performance and accuracy of 3-D spoiled gradient-dual-echo MRI with water- and fat-signal separation in liver-fat quantification: comparison to liver biopsy

To prospectively evaluate a 3-dimensional spoiled gradient-dual-echo (3D SPGR-DE) magnetic resonance imaging (MRI) sequence for the qualitative and quantitative analysis of liver fat content (LFC) in patients with the suspicion of fatty liver disease using histopathology as the standard of reference.

Iterative Reconstruction Algorithm for Abdominal Multidetector CT at Different Tube Voltages: Assessment of Diagnostic Accuracy, Image Quality, and Radiation Dose in a Phantom Study

To assess the diagnostic accuracy, image quality, and radiation dose of an iterative reconstruction algorithm compared with a filtered back projection (FBP) algorithm for abdominal computed tomography (CT) at different tube voltages.

Accuracy of diagnostic tests for clinically suspected upper extremity deep vein thrombosis: a systematic review

The best available test for the diagnosis of upper extremity deep venous thrombosis (UEDVT) is contrast venography. The aim of this systematic review was to assess whether the diagnostic accuracy of other tests for clinically suspected UEDVT is high enough to justify their use in clinical practise and to evaluate if any test can replace venography.

Influence of Smoothing on Voxel-Based Mesh Accuracy in Micro-Finite Element

The interest in automatic volume meshing for finite element analysis (FEA) has grown more since the appearance of microfocus CT (μCT), due to its high resolution, which allows for the assessment of mechanical behaviour at a high precision. Nevertheless, the basic meshing approach of generating one hexahedron per voxel produces jagged edges. To prevent this effect, smoothing algorithms have been introduced to enhance the topology of the mesh. However, whether smoothing also improves the accuracy of voxel-based meshes in clinical applications is still under question. There is a trade-off between smoothing and quality of elements in the mesh. Distorted elements may be produced by excessive smoothing and reduce accuracy of the mesh. In the present work, influence of smoothing on the accuracy of voxel-based meshes in micro-FE was assessed. An accurate 3D model of a trabecular structure with known apparent mechanical properties was used as a reference model. Virtual CT scans of this reference model (with resolutions of 16, 32 and 64 μm) were then created and used to build voxel-based meshes of the microarchitecture. Effects of smoothing on the apparent mechanical properties of the voxel-based meshes as compared to the reference model were evaluated. Apparent Young’s moduli of the smooth voxel-based mesh were significantly closer to those of the reference model for the 16 and 32 μm resolutions. Improvements were not significant for the 64 μm, due to loss of trabecular connectivity in the model. This study shows that smoothing offers a real benefit to voxel-based meshes used in micro-FE. It might also broaden voxel-based meshing to other biomechanical domains where it was not used previously due to lack of accuracy. As an example, this work will be used in the framework of the European project ContraCancrum, which aims at providing a patient-specific simulation of tumour development in brain and lungs for oncologists. For this type of clinical application, such a fast, automatic, and accurate generation of the mesh is of great benefit.