C. Monserrat

Universitat Politècnica de València | UPV · Department of Computer Systems and Computation

One of challenges of artificial intelligence as a whole is robustness. Many issues such as adversarial examples, out of distribution performance, Clever Hans phenomena, and the wider areas of AI evaluation and explainable AI, have to do with the following question: Did the system fail because it is a hard instance or because something else? In this...

Identifying patients with high risk of hip fracture is a great challenge in osteoporosis clinical assessment. Bone Mineral Density (BMD) measured by Dual-Energy X-Ray Absorptiometry (DXA) is the current gold standard in osteoporosis clinical assessment. However, its classification accuracy is only around 65%. In order to improve this accuracy, this...

A common way of learning to perform a task is to observe how it is carried out by experts. However, it is well known that for most tasks there is no unique way to perform them. This is especially noticeable the more complex the task is because factors such as the skill or the know-how of the expert may well affect the way she solves the task. In ad...

Background and objective: A great challenge in osteoporosis clinical assessment is identifying patients at higher risk of hip fracture. Bone Mineral Density (BMD) measured by Dual-Energy X-Ray Absorptiometry (DXA) is the current gold-standard, but its classification accuracy is limited to 65%. DXA-based Finite Element (FE) models have been develope...

In this paper, a method to in vivo estimate the relative stiffness between a hepatic lesion and the liver parenchyma is presented. This method is based on the finite element simulation of the deformation that the liver undergoes during the breathing process. Boundary conditions are obtained through a registration algorithm known as Coherent Point D...

This paper presents a novel non-invasive methodology to obtain the patient-specific relative stiffness between a hepatic lesion and the liver parenchyma in vivo. This relative stiffness can be used as a biomarker about the type of lesion. This biomarker together with the rest of pathological information can be used to plan a biopsy, an image-guide...

This work presents a data-driven model to simulate the mechanical behavior of the breast tissues in real time. The aim of this model is to speed up some multimodal registration algorithms, as well as some image-guided interventions. Ten virtual breast phantoms were used in this work. Their deformation during a mammography was performed off-line usi...

We present an event detection system in a laparoscopic surgery domain, as part of a more ambitious supervision by observation project. The system, which only requires the incorporation of two cameras in a laparoscopic training box, integrates several computer vision and machine learning techniques to detect the states and movements of the elements...

One of the major difficulties in activity recognition stems from the lack of a model of the world where activities and events are to be recognised. When the domain is fixed and repetitive we can manually include this information using some kind of ontology or set of constraints. On many occasions, however, there are many new situations for which on...

This work presents a data-driven method to simulate, in real-time, the biomechanical behavior of the breast tissues in some image-guided interventions such as biopsies or radiotherapy dose delivery as well as to speed up multimodal registration algorithms. Ten real breasts were used for this work. Their deformation due to the displacement of two co...

En la actualidad la simulación numérica del comportamiento mecánico de tejidos humanos en el campo de la medicina es un ámbito de estudio que ha despertado gran interés en la comunidad científica. El estudio del comportamiento de dichos tejidos conlleva una gran dificultad, en parte, atribuida al hecho de que el comportamiento de dichos tejidos cam...

Progress in biomechanical modelling of human soft tissue is the basis for the development of new clinical applications capable of improving the diagnosis and treatment of some diseases (e.g. cancer), as well as the surgical planning and guidance of some interventions. The finite element method (FEM) is one of the most popular techniques used to pre...

Abstract This paper presents a novel methodology to in-vivo estimate the elastic constants of a constitutive model proposed to characterize the mechanical behavior of the breast tissues. An iterative search algorithm based on genetic heuristics was constructed to in-vivo estimate these parameters using only medical images, thus avoiding invasive me...

Mediante la aplicación de métodos de aprendizaje automático (redes neuronales), se pretende realizar un modelador para pacientes con queratocono a los que se les han implantado anillos intracorneales para tratar su enfermedad.

This work presents a methodology for the in vivo characterization of the complete biomechanical behavior of the human cornea of each patient. Specifically, the elastic constants of a hyperelastic, second-order Ogden model were estimated for 24 corneas corresponding to 12 patients. The finite element method was applied to simulate the deformation of...

Purpose: This work presents a complete and automatic software application to aid radiologists in breast cancer diagnosis. The application is a fully automated method that performs a complete registration of magnetic resonance (MR) images and x-ray (XR) images in both directions (from MR to XR and from XR to MR) and for both x-ray mammograms, crani...

This paper presents a novel markerless monocular tracking system aimed at guiding ophthalmologists during external eye surgery. This new tracking system performs a very accurate tracking of the eye by detecting invariant points using only textures that are present in the sclera, i.e., without using traditional features like the pupil and/or cornea...

The accuracy of the patient-specific biomechanical models of the breast is a major concern for applications related with simulation, surgical guidance or cancer diagnosis. Being able to predict the localization of a lesion depends on the realism of the selected model. However, obtaining a realistic model of the breast tissues is not straightforward...

This work proposes a new approach based on Machine Learning to predict astigmatism in patients with kera-toconus (KC) after ring implantation. KC is a non-inflamatory, progressive thinning disorder of the cornea, resulting in a protusion, myopia and irregular astigmatism. The intracorneal ring implantation surgery has become a suitable technique to...

Keratoconus (KC) is the most common type of corneal ectasia. A corneal transplantation was the treatment of choice until the last decade. However, intra-corneal ring implantation has become more and more common, and it is commonly used to treat KC thus avoiding a corneal transplantation. This work proposes a new approach based on Machine Learning t...

Virtual surgery simulators enable surgeons to learn by themselves, shortening their learning curves. Virtual simulators offer an objective evaluation of the surgeon's skills at the end of each training session. The considered evaluation parameters are based on the analysis of the surgeon's gestures performed throughout the training session. Current...

The aim of this work is to propound a mechanical behavior model for simulating the deformation of the shoe upper material in gait for footwear CAD applications. The chosen material is calfskin. The proposed material behavior for the working range is a linear elastic orthotropic model which considers large deformation and membrane and bending loadin...

The surgical interventions of the breast are a delicate process where the surgeon must be very accurate. The patient is in supine position during the intervention, however the only resource that can be used is a Magnetic Resonance Image (MRI) taken in a different time and in prone position. The work presented in this paper develops a tool that tran...

A novel method of skin segmentation is presented aimed to obtain as many pixels belonging to the real skin as possible. This method is validated by experts in radiology. In addition, a biomechanical model of the breast, which considers the skin segmented in this way, is constructed to study the influence of considering real skin in the simulation o...

In this work, artificial neural networks (ANNs) are proposed to predict the dorsal pressure over the foot surface exerted by the shoe upper while walking. A model that is based on the multilayer perceptron (MLP) is used since it can provide a single equation to model the exerted pressure for all the materials used as shoe uppers. Five different mod...

In this study, six biomechanical models for simulating lamb liver behaviour are presented. They are validated using similarity coefficients from Medical Image on reconstructed volumes from computerised tomography images. In particular, the Jaccard and Hausdorff coefficients are used. Loads of 20 and 40 g are applied to the livers and their deformat...

In this paper, a set of coefficients commonly used in Medical Image to estimate the committed error comparing two images is presented, which, combined together, allow to determine the similarity between volumes. Furthermore, an analysis of the behavior of these coefficients is performed to determine those coefficients that better discriminate the f...

Breast Magnetic Resonance Image skin has similar intensity levels than dense tissue, and may produce segmentation errors if not managed correctly. In this work a novel skin segmentation method is presented and validated by experts, aimed to obtain as many pixels belonging to the real skin as possible. Segmented skin will be used to build a breast b...

In this paper, a procedure to experimentally simulate the behavior of the liver when the pneumoperitoneum technique is applied in laparoscopic surgery is presented, as well as methodology to make the comparison of each sample before and after insufflating the gas. This comparison is carried out using the 3D reconstruction of the volume from the CT...

This paper describes the new advances carried out for Simucal. Simucal was introduced in [13] and it is a footwear virtual simulator designed to perform studies of comfort and functionality in CAD footwear design. In this paper, a new finite element model for the deformation of shoe upper materials in gait is presented. This model provides a physic...

Recently, important advances in virtual reality have made possible real improvements in computer aided design, CAD. These advances are being applied to all the fields and they have reached to the footwear design. The majority of the interaction foot-shoe simulation processes have been focused on the interaction between the foot and the sole. Howeve...

Establishing the appropriate pressure exerted by the shoe upper over the foot surface is fundamental for the design of specific footwear, although measuring the dorsal pressures can also provide important additional information. In previous works, a virtual simulator to perform studies of comfort and functionality in CAD footwear design was present...

Parkinson's disease is a degenerative disease of the central nervous system. One of the most effective treatments is deep brain stimulation. This technique requires the localization of an objective structure: the subthalamic nucleus. Unfortunately this structure is difficult to locate. In this work the creation of a deformable brain atlas that enab...

This paper presents a computational model that simulates the deformation of materials in shoe uppers and that quantifies the distribution of forces on the foot surface in a complete step. A preliminary linear elastic model based on the finite element method was developed for the upper, using a simple triangular element of three nodes. This model ca...

VR laparoscopic simulators have demonstrated its validity in recent studies, and research should be directed towards a high training effectiveness and efficacy. In this direction, an insight into simulators' didactic design and technical development is provided, by describing the methodology followed in the building of the SINERGIA simulator. It de...

In this paper we present an automatic system for segmentation and recognition of relevant tissues in maxillofacial CT images. The system allows to dynamically validate anatomical information of these structures. Our procedure differs from previous attempts in its use of advanced low level segmentation operators and specific knowledge bases that emb...

This paper describes an augmented reality (AR) system for the treatment of acrophobia. First, the technical characteristics of the original prototype are described. Second, the capacity of the immersive photography used in the AR system to provoke sense of presence in users is tested. Forty-one participants without fear of heights walked around a s...

This paper describes a new method for 3D orthodontic treatment simulation developed for an orthodontic planning system (MAGA-LLANES). We develop an original system for three-dimensional reconstruction of dental anatomy. Data are acquired directly from the patient with low cost 3D digitizers avoiding use of dental casts in orthodontic treatments. We...

Virtual reality (VR) is useful for treating several psychological problems, including phobias such as fear of flying, agoraphobia, claustrophobia, and phobia to insects and small animals. We believe that augmented reality (AR) could also be used to treat some psychological disorders. AR and VR share some advantages over traditional treatments. Howe...

Simulating the behaviour of elastic objects in real time is one of the current objectives of computer graphics. One of its fields of application lies in virtual reality, mainly in surgery simulation systems. In computer graphics, the models used for the construction of objects with deformable behaviour are known as deformable models. These have two...

The incorporation of physical behaviour to synthetic objects constructed by computer is one of the main objectives of computer graphics research. In this paper we present a new interactive deformable model based on linked volumes. As such, it is straightforward and implicitly preserves the volume of the modelled objects. The proposed model resorts...

This paper presents a method to obtain a polygonal mesh representing the ramified structure of a tree. We use a method based on L-systems to model the tree. A refining process is applied once the model of the tree is obtained. Soft transitions are achieved in the areas of the tree where there are divisions. In this way, we avoid the possible superi...

A new general image segmentation system is presented, based on the calculation of a tree representation of the original image in which image regions are assigned to tree nodes, followed by a correspondence process with a model tree, which embeds the a priori knowledge about the images. For this correspondence, an original algorithm is proposed, whi...

In this article a matching system of the Schaltenbrand–Bailey's brain atlas is presented. The system matches the Schaltenbrand–Bailey's brain atlas onto a patient's brain. It uses two patient's brain and the Schaltenbrand's brain atlas. The first patient's brain is used to match the Schaltenbrand's brain atlas with it in a semiautomatic way. Then,...

Microscopes and neuronavigators are often used in neurosurgical procedures. Our Mixed Reality prototype proposes to show neuronavigator imaging overlayed onto the surgical field of view. The proposal is to replace the traditional optical microscope with a digital one. The system is presented in this article. Firsts trials have been performed at the...

Minimally invasive surgery is a technique that permits interventions through very small incisions. This minimises the patients' trauma and permits a faster recovery in comparison with classical surgery. The disadvantage of this surgery technique, though, is its complexity, requiring a high training effort of the surgeon. In this paper, we present a...

The construction of surgery simulators will be a key tool in the development and diffusion of minimally invasive surgery. Nowadays, most simulators are oriented to training surgeons in only one surgery technique. Most of them only permit the modelling of tissues with only one kind of deformable model. In this paper, we present our generic surgery s...

A general simulator for minimally invasive surgery procedures is presented in this paper. Minimally invasive surgical procedures are complex, requiring the surgeons a high training effort. The objective of the simulator is to reduce the casualty due to training. The simulator introduces tools for creating surgical environments and provides interfac...

Most deformable models for surgery simulation are quite straightforward to achieve the computational speed required for real time simulation. However, they typically are more adjusted to merely graphical representation needs, i.e. surface-oriented (e.g. mass-spring type models), and neglect mechanical realism, although the simulation of cutting in...

A system for automatic detection of cephalometric landmarks is presented. Landmark detection is carried out in two steps: a line detection module searches for significant, well-contrasted lines of the image, such as the jaw line or the nasal spine. The landmark detection module uses the lines located in the first module to determine the search area...

Most neurosurgical procedures make use of a neuronavigator showing the surgeon's tool guidance through a reconstructed patient's anatomy model from MRI and/or CT and, if wanted, a previously planned modus operandi. It would be preferable for the surgeon not to look away from the operative scene for further information. Image guidance given by a neu...

Minimally invasive surgery is a technique that permits interventions through very small incisions. This minimises the patients' trauma and permits a faster recovery in comparison with classical surgery. The disadvantage of this surgery technique, though, is its complexity, requiring a high training effort of the surgeon. In this paper, we present a...

Simulation of the behaviour of elastic objects in real time is one of the present objectives of computer graphics. One of its fields of application lies in virtual reality, mainly in surgery simulation systems. Models used for the construction of objects with deformable behaviour in computer graphics are known as deformable models. These have two c...

A model-driven, multiscale medical image segmentation system is presented. A tree representation is calculated for the image, using a modification of the immersion algorithm used for watersheds calculation. Segmentation is carried out by a matching process between the obtained tree and a tree model, which embeds the prior knowledge about the images...

This paper presents a method to obtain a single continuos polygonal mesh representing a branched structure taken from a tree and modelled by a L-systems-based method. A refinement process is made once the model of the tree is obtained. This process produces smooth transitions in the tree areas joining the branches. That is, the overlapping problems...

A semi-automatic method for three-dimensional segmentation of medical images is proposed. A multiresolution representation is achieved through the application of morphological filters, which assures causality for image extrema. This allows for a compact scale space representation, in which each extremum is assigned a scale value. Interactive select...

In this article a new efficient method in deformable brain atlases is proposed. To achieve this goal we use the composition of two applications, an affine transformation and other one. In this second application we have used three different approaches. One approach that has been used and two new approaches. We use a set of points to be identified b...

96 3877518 Fax.; 96 3879510 1. Resumen En este artículo se describe un nuevo sistema de cirugía asistida por ordenador el cual ha sido validado en varias intervenciones de neurocirugía funcional. Dicho sistema utiliza un digitalizador tridimensional óptico para el seguimiento en tiempo real de la posición espacial del instrumental quirúrgico. Para...

This paper considers the problem of deformable contour initialization and modeling for segmentation of the human prostate in medical images. We propose a new technique for elastic deformation restriction to particular object shapes of any closed planar curve using localized multiscale contour parameterization based on the 1D dyadic wavelet transfor...