Long-range multicore optical fiber displacement sensor.

In this Letter, a long-range optical fiber displacement sensor based on an extrinsic Fabry-Perot interferometer (EFPI) built with a strongly coupled multicore fiber (SCMCF) is proposed and demonstrated. To fabricate the device, 9.2 mm of SCMCF was spliced to a conventional single-mode fiber (SMF). The sensor reflection spectrum is affected by super-mode interference in the SCMCF and the interference produced by the EFPI. Displacement of the SMF-SCMCF tip with respect to a reflecting surface produces quantifiable changes in the amplitude and period of the interference pattern in the reflection spectrum. Since the multicore fiber is an efficient light collecting area, sufficient signal intensity can be obtained for displacements of several centimeters. By analyzing the interference pattern in the Fourier domain, it was possible to measure displacements up to 50 mm with a resolution of approximately 500 nm. To our knowledge, this is the first time that a multicore fiber has been used to build a displacement sensor. The dynamic measurement range is at least seven times larger than that achieved with an EFPI built with a conventional SMF. Moreover, the SMF-SCMCF tip is robust and easy to fabricate and replicate.

Prediction of osteoporotic hip fracture in postmenopausal women through patient-specific FE analyses and machine learning.

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 developed to predict the mechanical failure of the bone. Yet, their contribution has been modest. In this study, supervised machine learning (ML) is applied in conjunction with clinical and computationally driven mechanical attributes. Through this multi-technique approach, we aimed to obtain a predictive model that outperforms BMD and other clinical data alone, as well as to identify the best-learned ML classifier within a group of suitable algorithms. A total number of 137 postmenopausal women (81.4 ± 6.95 years) were included in the study and separated into a fracture group (n = 89) and a control group (n = 48). A semi-automatic and patient-specific DXA-based FE model was used to generate mechanical attributes, describing the geometry, the impact force, bone structure and mechanical response of the bone after a sideways-fall. After preprocessing the whole dataset, 19 attributes were selected as predictors. Support Vector Machine (SVM) with radial basis function (RBF), Logistic Regression, Shallow Neural Networks and Random Forest were tested through a comprehensive validation procedure to compare their predictive performance. Clinical attributes were used alone in another experimental setup for the sake of comparison. SVM was confirmed to generate the best-learned algorithm for both experimental setups, including 19 attributes and only clinical attributes. The first, generated the best-learned model and outperformed BMD by 14pp. The results suggests that this approach could be easily integrated for effective prediction of hip fracture without interrupting the actual clinical workflow.

Cannabis use influence on peripheral brain-derived neurotrophic factor levels in antipsychotic-naïve first-episode psychosis.

The objective of this study is to determine whether cannabis influences BDNF levels in patients with psychosis (FEP) and healthy volunteers (HV) to help understand the role of BDNF in psychosis. We assessed the association between BDNF and cannabis in a cohort of FEP antipsychotic-naïve patients and HV, whilst controlling for other potential confounding factors. 70 FEP drug-naive patients and 57 HV were recruited. A sociodemographic variable collection, structured clinical interview, weight and height measurement, substance use determination, and blood collection to determine BDNF levels by ELISA analysis were done. In FEP patients, cannabis use was associated with BDNF levels (high cannabis use was associated with lower BDNF levels). Moreover, cannabis use was statistically significantly associated with age (high use of cannabis was associated with younger age). In HV, no relationship between cannabis use and BDNF levels was observed. Otherwise, cannabis use was significantly associated with tobacco use, so that high cannabis users were also high tobacco users. This study showed a different association between cannabis use and BDNF levels in FEP patients compared with HV, particularly, with high doses of cannabis. These findings may help understand the deleterious effects of cannabis in some vulnerable individuals, as well as discrepancies in the literature.

A finite element-based machine learning approach for modeling the mechanical behavior of the breast tissues under compression in real-time.

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 compression plates was simulated off-line using the finite element (FE) method. Three machine learning models were trained with the data from those simulations. Then, they were used to predict in real-time the deformation of the breast tissues during the compression. The models were a decision tree and two tree-based ensemble methods (extremely randomized trees and random forest). Two different experimental setups were designed to validate and study the performance of these models under different conditions. The mean 3D Euclidean distance between nodes predicted by the models and those extracted from the FE simulations was calculated to assess the performance of the models in the validation set. The experiments proved that extremely randomized trees performed better than the other two models. The mean error committed by the three models in the prediction of the nodal displacements was under 2 mm, a threshold usually set for clinical applications. The time needed for breast compression prediction is sufficiently short to allow its use in real-time (<0.2 s).

Patient-specific simulation of the intrastromal ring segment implantation in corneas with keratoconus.

PURPOSE: The purpose of this study was the simulation of the implantation of intrastromal corneal-ring segments for patients with keratoconus. The aim of the study was the prediction of the corneal curvature recovery after this intervention. METHODS: Seven patients with keratoconus diagnosed and treated by implantation of intrastromal corneal-ring segments were enrolled in the study. The 3D geometry of the cornea of each patient was obtained from its specific topography and a hyperelastic model was assumed to characterize its mechanical behavior. To simulate the intervention, the intrastromal corneal-ring segments were modeled and placed at the same location at which they were placed in the surgery. The finite element method was then used to obtain a simulation of the deformation of the cornea after the ring segment insertion. Finally, the predicted curvature was compared with the real curvature after the intervention. RESULTS: The simulation of the ring segment insertion was validated comparing the curvature change with the data after the surgery. Results showed a flattening of the cornea which was in consonance with the real improvement of the corneal curvature. The mean difference obtained was of 0.74 mm using properties of healthy corneas. CONCLUSIONS: For the first time, a patient-specific model of the cornea has been used to predict the outcomes of the surgery after the intrastromal corneal-ring segments implantation in real patients.

Methodology based on genetic heuristics for in-vivo characterizing the patient-specific biomechanical behavior of the breast tissues.

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 measurements of the mechanical response of the breast tissues. For the first time, a combination of overlap and distance coefficients were used for the evaluation of the similarity between a deformed MRI of the breast and a simulation of that deformation. The methodology was validated using breast software phantoms for virtual clinical trials, compressed to mimic MRI-guided biopsies. The biomechanical model chosen to characterize the breast tissues was an anisotropic neo-Hookean hyperelastic model. Results from this analysis showed that the algorithm is able to find the elastic constants of the constitutive equations of the proposed model with a mean relative error of about 10%. Furthermore, the overlap between the reference deformation and the simulated deformation was of around 95% showing the good performance of the proposed methodology. This methodology can be easily extended to characterize the real biomechanical behavior of the breast tissues, which means a great novelty in the field of the simulation of the breast behavior for applications such as surgical planing, surgical guidance or cancer diagnosis. This reveals the impact and relevance of the presented work.

A new methodology for the in vivo estimation of the elastic constants that characterize the patient-specific biomechanical behavior of the human cornea.

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 human corneas due to non-contact tonometry, and an iterative search controlled by a genetic heuristic was used to estimate the elastic parameters that most closely approximates the simulated deformation to the real one. The results from a synthetic experiment showed that these parameters can be estimated with an error of about 5%. The results of 24 in vivo corneas showed an overlap of about 90% between simulation and real deformed cornea and a modified Hausdorff distance of 25 µm, which indicates the great accuracy of the proposed methodology.

Markerless monocular tracking system for guided external eye surgery.

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 reflections, which remain partially or totally occluded in most surgeries. Two known algorithms that compute invariant points and correspondences between pairs of images were implemented in our system: Scalable Invariant Feature Transforms (SIFT) and Speed Up Robust Features (SURF). The results of experiments performed on phantom eyes show that, with either algorithm, the developed system tracks a sphere at a 360° rotation angle with an error that is lower than 0.5%. Some experiments have also been carried out on images of real eyes showing promising behavior of the system in the presence of blood or surgical instruments during real eye surgery.

A complete software application for automatic registration of x-ray mammography and magnetic resonance images.

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, craniocaudal (CC), and mediolateral oblique (MLO). This new approximation allows radiologists to mark points in the MR images and, without any manual intervention, it provides their corresponding points in both types of XR mammograms and vice versa. METHODS: The application automatically segments magnetic resonance images and x-ray images using the C-Means method and the Otsu method, respectively. It compresses the magnetic resonance images in both directions, CC and MLO, using a biomechanical model of the breast that distinguishes the specific biomechanical behavior of each one of its three tissues (skin, fat, and glandular tissue) separately. It makes a projection of both compressions and registers them with the original XR images using affine transformations and nonrigid registration methods. RESULTS: The application has been validated by two expert radiologists. This was carried out through a quantitative validation on 14 data sets in which the Euclidean distance between points marked by the radiologists and the corresponding points obtained by the application were measured. The results showed a mean error of 4.2 ± 1.9 mm for the MRI to CC registration, 4.8 ± 1.3 mm for the MRI to MLO registration, and 4.1 ± 1.3 mm for the CC and MLO to MRI registration. CONCLUSIONS: A complete software application that automatically registers XR and MR images of the breast has been implemented. The application permits radiologists to estimate the position of a lesion that is suspected of being a tumor in an imaging modality based on its position in another different modality with a clinically acceptable error. The results show that the application can accelerate the mammographic screening process for high risk populations or for dense breasts.

A new approach based on Machine Learning for predicting corneal curvature (K1) and astigmatism in patients with keratoconus after intracorneal ring implantation.

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 to predict the vision gain of KC patients after ring implantation. That vision gain is assessed by means of the corneal curvature and the astigmatism. Different models were proposed; the best results were achieved by an artificial neural network based on the Multilayer Perceptron. The error provided by the best model was 0.97D of corneal curvature and 0.93D of astigmatism.

Automatic supervision of gestures to guide novice surgeons during training.

BACKGROUND: 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. Currently, this information is usually known by surgeons only at the end of the training session, but very limited during the training performance. In this paper, we present a novel method for automatic and interactive evaluation of the surgeon's skills that is able to supervise inexperienced surgeons during their training session with surgical simulators. METHODS: The method is based on the assumption that the sequence of gestures carried out by an expert surgeon in the simulator can be translated into a sequence (a character string) that should be reproduced by a novice surgeon during a training session. In this work, a string-matching algorithm has been modified to calculate the alignment and distance between the sequences of both expert and novice during the training performance. RESULTS: The results have shown that it is possible to distinguish between different skill levels at all times during the surgical training session. CONCLUSIONS: The main contribution of this paper is a method where the difference between an expert's sequence of gestures and a novice's ongoing sequence is used to guide inexperienced surgeons. This is possible by indicating to novices the gesture corrections to be applied during surgical training as continuous expert supervision would do.

Estimation of the elastic parameters of human liver biomechanical models by means of medical images and evolutionary computation.

This paper presents a method to computationally estimate the elastic parameters of two biomechanical models proposed for the human liver. The method is aimed at avoiding the invasive measurement of its mechanical response. The chosen models are a second order Mooney-Rivlin model and an Ogden model. A novel error function, the geometric similarity function (GSF), is formulated using similarity coefficients widely applied in the field of medical imaging (Jaccard coefficient and Hausdorff coefficient). This function is used to compare two 3D images. One of them corresponds to a reference deformation carried out over a finite element (FE) mesh of a human liver from a computer tomography image, whilst the other one corresponds to the FE simulation of that deformation in which variations in the values of the model parameters are introduced. Several search strategies, based on GSF as cost function, are developed to accurately find the elastics parameters of the models, namely: two evolutionary algorithms (scatter search and genetic algorithm) and an iterative local optimization. The results show that GSF is a very appropriate function to estimate the elastic parameters of the biomechanical models since the mean of the relative mean absolute errors committed by the three algorithms is lower than 4%.

Analysis of several biomechanical models for the simulation of lamb liver behaviour using similarity coefficients from medical image.

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 deformation is simulated by means of the finite element method. The models used are a linear elastic model, a neo-Hookean model, a Mooney-Rivlin model, an Ogden model, a linear viscoelastic model and a viscohyperelastic model. The model that provided a behaviour that is closest to reality was the viscohyperelastic model, where the hyperelastic part was modelled with an Ogden model.

Segmentation of the breast skin and its influence in the simulation of the breast compression during an X-ray mammography.

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 of the breast compression during an X-ray mammography. The reaction forces of the plates are obtained and compared with the reaction forces obtained using classical methods that model the skin as a 2D membranes that cover all the breast. The results of this work show that, in most of the cases, the method of skin segmentation is accurate and that real skin should be considered in the simulation of the breast compression during the X-ray mammographies.

A study about coefficients to estimate the error in biomechanical models used to virtually simulate the organ behaviors.

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 fit error, proving that these are Jaccard coefficient and a modification of Hausdorff coefficient. In addition, the combination of both coefficients is applied to compare two given biomechanical models of the lamb liver.

MRI skin segmentation for the virtual deformation of the breast under mammographic compression.

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 biomechanical model to register X-Ray Images with Magnetic Resonance Images in the future, using a virtually deformed Magnetic Resonance Image.

Pneumoperitoneum technique simulation in laparoscopic surgery on lamb liver samples and 3D reconstruction.

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 images when either pneumoperitoneum is applied and when it is not. This methodology has showed that there are perceptible changes of volume when the pneumoperitoneum is applied.

A study of the viability of obtaining a generic animation of the foot while walking for the virtual testing of footwear using dorsal pressures.

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 presented. This paper describes the procedure carried out to obtain the foot animations used in this simulator. The virtual feet used in the simulator are feet without a standard form scanned in a static way. Their movements are rebuilt from the register of movements of several foot anatomical points during a complete step. The dorsal pressures exerted by some shoe uppers on these anatomical points were measured for several subjects and used to establish the viability of the use of these animations in a virtual simulator for footwear.

Peripheral plasma adenosine release in patients with chronic heart failure.

OBJECTIVE: Chronic heart failure (CHF) is accompanied by increased adenosine plasma levels (APLs). It is unknown whether adenosine release occurs at the peripheral level or whether the myocardium itself is the source of adenosine release. To answer this question, we evaluated APLs in the coronary sinus of CHF patients during a resynchronisation procedure and compared the values with those at the peripheral level. We also investigated a possible correlation between APLs and ischaemia-modified albumin (IMA) levels, a useful marker of tissue ischaemia. METHODS: 19 men and seven women were prospectively included. Blood samples for APLs were collected simultaneously from a brachial vein (peripheral) and from the coronary sinus. Blood samples for brain natriutretic peptide (BNP) and IMA were collected from a brachial vein. RESULTS: APLs from the brachial vein were higher than those from the coronary sinus (1.69 vs 0.75 muM p<0.01). IMA levels were correlated with APLs from the brachial vein (r = 0.59, p<0.01). BNP concentrations were correlated with APLs from the brachial vein (r = 0.73, p<0.001) but not with APLs from the coronary sinus (r = 0.38, p>0.05). BNP concentrations and IMA levels were correlated (r = 0.71, p<0.001). CONCLUSIONS: In CHF patients, adenosine release occurs at a peripheral level and not at the myocardium level.