2016
Use of scaling analysis in the geometric studying of astrocytomas
Francisco J. Torres, Miguel Martín-Landrove, Juan R. Vergara
Proceedings of the 13th International Congress on Numerical Methods in Engineering and Applied Sciences, CIMENICS 2016, PI 117-125
Tumor growth can be characterized by using scaling analysis methods performed upon the tumor interface; the procedure yields key parameters that define growth geometry according different universality classes. In the present work, results obtained by scaling analysis are shown for tumor astrocytomas, of primary origin, i.e, pilocytic, diffuse and anaplastic, is used to calculate fractal dimension and local roughness exponents to characterize in vivo 3-D tumor growth. Image acquisition was made according to the standard protocol used for brain radiotherapy and radiosurgery, i.e., axial, coronal and sagittal magnetic resonance T1-weighted images, and comprising the brain volume for image registration. Image segmentation was performed by the application of the k-means procedure upon contrasted images. The results show significant variations of the parameters depending on the tumor stage and histological origin.
Dimensión fractal de radiografías dentales y su relación con la evolución de tratamientos implantológicos
Kathiuska Díaz, Gladys Uzcátegui, Miguel Martín-Landrove, Iván Sánchez, Erich Marcano
Proceedings of the 13th International Congress on Numerical Methods in Engineering and Applied Sciences, CIMENICS 2016, PI 97-104 [article in spanish]
The self-similar fractal geometry studies forms and provides tools to quantify the complexity of partially self-similar natural structures. From digital radiographs where a structure is shown, its complexity can be quantified using any of the existing algorithms for determining the fractal dimension, such as box counting method. For implant osseointegration, there is preliminary evidence that the fractal dimension of trabecular bone can be used as a parameter to quantify the regeneration degree after treatment. The objective of this study is to obtain quantitative information on bone structure to assess quickly implant therapy. Specific areas nearby implant placement were analyzed on panoramic radiographs in two time points: before the placement of the dental implant and newly placed. Preliminary results by analyzing ten cases of implants suggest that the fractal dimension tends to decrease during the first days of the placement. Currently it is carrying out the analysis process of the cases within six months after the implants were placed. It is necessary to analyze a larger number of cases to establish whether with this technique it is possible to make predictions of cases treated with dental implants. This would constitute an important contribution to dental clinical practice, which has a growing demand for implant therapy.
Segmentation of dynamic contrast-enhanced magnetic resonance images of the prostate
Wuilian Torres, Leonardo Cordero, Miguel Martín-Landrove, Antonio Rueda-Toicen
Proceedings of the 13th International Congress on Numerical Methods in Engineering and Applied Sciences, CIMENICS 2016, PI 1-11
Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) is a valuable tool to localize, characterize, and evaluate anomalous prostate tissue. The DCE-MRI technique produces ultrafast gradient-echo acquisitions of MRI volumes with high temporal resolution that are generated at regular intervals while the patient receives, in a controlled manner, a paramagnetic contrast agent. Angiogenesis in malignant tumors of the prostate is characterized by aggressive vessel growth that contributes to the rapid evacuation (“wash-out”) of the contrast agent. This feature becomes evident in a time series where each voxel exhibits a particular behavior of contrast uptake (“wash-in”) and posterior wash-out; this behavior depends on both the properties of the tissue represented by the voxel and its degree of vascularization. In this work, we propose a segmentation method that groups together neighboring voxels with similar contrast wash-out responses. The segmentation algorithm uses a region growing technique that is a variant of the “GrowCut” cellular automaton. This cellular automaton has cells associated with each voxel in the volume, and every cell in the automaton evolves iteratively according to its relationship with its neighboring cells. Initially, the cellular automaton is given “seed” cells, which are determined in an automatic manner through morphological filters that identify homogeneous regions in the volume. Seed cells are representatives of the clinically relevant types of tissues in the prostate. Each cell in the automaton has three parameters: a label that identifies the type of tissue in the associated voxel, a vector with the values of the DCE-MRI time series, and a coefficient called “strength” with values between 0 and 1 that is interpreted as the probability with which the cell belongs to its assigned label. Every non-seed cell can modify its state; this occurs when each cell is attacked by its neighbors with a strength of attack that is inversely proportional to the similarity of the values of the time series between the cells. If the strength of the attacked cell is less than the strength of the attack of one of its neighbors, the state of the attacked cell changes and it takes the label of the attacking cell. The attacked cell also updates its strength making it equal to the strength of the attack with which it was conquered. To perform a clinical validation of the resulting segmentations, we used various cases from the database of The Cancer Imaging Archive (TCIA), National Cancer Institute (NCI) [5, 6].
Pharmacokinetic parameters quantification in DCE-MRI for prostate cancer
Jhonalbert Aponte, Álvaro Ruíz, Miguel Martín-Landrove, Wuilian Torres, Leonardo Cordero
Proceedings of the 13th International Congress on Numerical Methods in Engineering and Applied Sciences, CIMENICS 2016, PI 85-95
Tumor vascularity detection and quantification is of high relevance in the assessment of cancer lesions not only for disease diagnostics but for therapy considerations and monitoring. The present work is addressed to the quantification of pharmacokinetic parameters, as derived from the two-compartment Brix model, through the analysis and processing of Dynamic Contrast Enhanced Magnetic Resonance Images (DCE-MRI) of prostate cancer lesions. The 3D image sets were acquired at regular time intervals, covering all the phases implied in contrast injection (wash-in and wash-out phases) and the standardized image intensity is determined for each voxel, conforming a 4D data set. Previous voxel classification was carried out by the three-time-point method proposed by Degani et al. (1997) and Furman-Haran et al. (1998) in order to identify regions of interest. Relevant pharmacokinetic parameters, such as 𝑘𝑒𝑙, the vascular elimination rate, and 𝑘𝑒𝑝, the extravascular transfer rate, are extracted by a Levenberg-Marquardt algorithm and parameter distributions maps were obtained for either pathological and unaffected glandular regions. Results can be applied to prostate cancer diagnostic evaluation and therapy follow up.
Geometrical quality assessment in stereotactic images
Eric Sira, Wuilian Torres, Miguel Martín-Landrove
Proceedings of the 13th International Congress on Numerical Methods in Engineering and Applied Sciences, CIMENICS 2016, PI 55-63
Gamma Knife stereotactic radiosurgery is a very specialized medical procedure for the treatment of intracranial lesions. For this type of treatment, magnetic resonance images are used not only for diagnostic and treatment monitoring but also for a precise stereotactic lesion localization to be used by a treatment planning system. Correct positioning of the stereotactic frame on the patient and correct patient positioning in the MRI scanner are crucial for a precise and time optimized treatment. In the present work, fiducial marks left by the stereotactic frame are used to estimate the distance between the segmented lesion and the center of the stereotactic frame. Image segmentation was performed on axial images by a region growing algorithm using the mathematical morphology connectivity on a von Neumann neighborhood. The lesion border is estimated by subtraction from the segmented binary image of a binary image generated by the morphological erosion operator with a 3x3 square structuring element. Different lesions were classified according to the distance parameter and the degree of difficulty of the general procedure, i.e., number of shots, collimator diameters, manual or automatic mode and collision probability, even the possibility of no treatment at all. A retrospective study using the information of 70 treatments was performed and as a result a difficulty map was obtained, which can be used as a reference for the evaluation of further treatments. Also, the fiducial marks were used to estimate the aligning between stereotactic frame axis and the MRI scanner axis for quality assurance of the patient positioning at the MRI scanner, determining tolerance levels appropriate for the treatment planning system. The proposed method can be easily implemented in any gamma knife facility and become a part of the quality assurance protocol.
Characterizing the structure of complex protein-protein interaction networks
Allan A. Zea, Antonio Rueda-Toicen
Proceedings of the 13th International Congress on Numerical Methods in Engineering and Applied Sciences, CIMENICS 2016, BSB 93-102
Network theorists have developed methods to characterize the complex interactions in natural phenomena. The structure of the network of interactions between proteins is important in the field of proteomics, and has been subject to intensive research in recent years, as scientists have become increasingly capable and interested in describing the underlying structure of interactions in both normal and pathological biological processes. In this paper, we survey the graph-theoretic characterization of protein-protein interaction networks {PINs) in terms of structural features, and discuss its possible applications in biomedical research. We also perform a brief revision of network theory's classical literature and discuss modem statistical and computational techniques to describe the structure of PINs.
Tumor growth in the brain: complexity and fractality
Miguel Martín-Landrove, Antonio Brú, Antonio Rueda-Toicen, Francisco Torres-Hoyos
The Fractal Geometry of the Brain, Part of the Springer Series in Computational Neuroscience pp 351-369, 2016
Tumor growth is a complex process characterized by uncontrolled cell proliferation and invasion of neighboring tissues. The understanding of these phenomena is of vital importance to establish appropriate diagnosis and therapy strategies and starts with the evaluation of their complexity with suitable descriptors produced by scaling analysis. There has been considerable effort in the evaluation of fractal dimension as a suitable parameter to describe differences between normal and pathological tissues, and it has been used for brain tumor grading with great success. In the present work, several contributions, which exploit scaling analysis in the context of brain tumors, are reviewed. These include very promising results in tumor segmentation, grading, and therapy monitoring. Emphasis is done on scaling analysis techniques applicable to multifractal systems, proposing new descriptors to advance the understanding of tumor growth dynamics in brain. These techniques serve as a starting point to develop innovative practical growth models for therapy simulation and optimization, drug delivery, and the evaluation of related neurological disorders.
An attractor network-based model with Darwinian dynamics
Harold P. de Vladar, Anna Fedor, András Szilágyi, István Zachar, Eörs Szathmáry
GECCO '16 Companion Proceedings of the 2016 on Genetic and Evolutionary Computation Conference Companion, Pages 1049-1052, ACM New York, NY, USA, 2016
The human brain can generate new ideas, hypotheses and candidate solutions to difficult tasks with surprising ease. We argue that this process has evolutionary dynamics, with multiplication, inheritance and variability all implemented in neural matter. This inspires our model, whose main component is a population of recurrent attractor networks with palimpsest memory that can store correlated patterns. The candidate solutions are represented as output patterns of the attractor networks and they are maintained in implicit working memory until they are evaluated by selection. The best patterns are then multiplied and fed back to attractor networks as a noisy version of these patterns (inheritance with variability), thus generating a new generation of candidate hypotheses. These components implement a truly Darwinian process which is more efficient than both natural selection on genetic inheritance or learning, on their own. We argue that this type of evolutionary search with learning can be the basis of high-level cognitive processes, such as problem solving or language.
Breeding novel solutions in the brain: a model of Darwinian neurodynamics
András Szilágyi, István Zachar, Anna Fedor, Harold P. de Vladar, Eörs Szathmáry
F1000Research 2016, 5:2416
Background: The fact that surplus connections and neurons are pruned during development is well established. We complement this selectionist picture by a proof-of-principle model of evolutionary search in the brain, that accounts for new variations in theory space. We present a model for Darwinian evolutionary search for candidate solutions in the brain.
Methods: We combine known components of the brain – recurrent neural networks (acting as attractors), the action selection loop and implicit working memory – to provide the appropriate Darwinian architecture. We employ a population of attractor networks with palimpsest memory. The action selection loop is employed with winners-share-all dynamics to select for candidate solutions that are transiently stored in implicit working memory.
Results: We document two processes: selection of stored solutions and evolutionary search for novel solutions. During the replication of candidate solutions attractor networks occasionally produce recombinant patterns, increasing variation on which selection can act. Combinatorial search acts on multiplying units (activity patterns) with hereditary variation and novel variants appear due to (i) noisy recall of patterns from the attractor networks, (ii) noise during transmission of candidate solutions as messages between networks, and, (iii) spontaneously generated, untrained patterns in spurious attractors.
Conclusions: Attractor dynamics of recurrent neural networks can be used to model Darwinian search. The proposed architecture can be used for fast search among stored solutions (by selection) and for evolutionary search when novel candidate solutions are generated in successive iterations. Since all the suggested components are present in advanced nervous systems, we hypothesize that the brain could implement a truly evolutionary combinatorial search system, capable of generating novel variants.
Interventricular asynchrony determination by equilibrium radioisotope ventriculography image analysis
Larry Suárez, Lila Carrizales-Silva, Aisa Manzo, Miguel Martín-Landrove
Acta Científica Venezolana 67(3):1-13, 2016
Interventricular cardiac asynchrony assessment is very important in patients’ evaluation for Cardiac Resynchronization Therapy. In the present work, images obtained by Equilibrium Radioisotope Ventriculography Imaging are used to evaluate the ventricular asynchrony and ventricular temporal parameters. The method was applied to a 17 female patients set with an average age >50 years. Image segmentation proceeds by histogram thresholding with Opsu’s method throughout the whole set of frames, and Time-Activity Curves (TAC) were obtained. Fourier analysis was performed up to the fourth harmonic to measured TACs, obtaining the Ventricular Ejection Fraction (VEF) and ventricular volume parameters. Results indicate an average ventricular contraction time of 371 ± 41 ms for the left ventricle and 355 ± 45 ms for the right ventricle, 70.6 % of the total population exhibit a VEF for the left ventricle superior to 50 % and the average initial contraction time for the left ventricle was 39.9 ± 36 ms compared to the time for the right ventricle of 49.18 ± 40 ms, which indicates a ventricular asynchrony of 9.7 ± 16 ms. Further research work is under development to extend the use of the proposed method as a general practice of equilibrium radioisotope ventriculography.
Computational Prediction of Tooth Shades in Direct Restorations with Composite Resins
Allan Zea, Aljhadys Zea, Mabel Sáenz-Guzmán
JIFI 2016
The VITA shade guide is a dental instrument that allows accurate and secure determination of tooth color and shade, which constitutes one of the most important but still complex aspects of dental restorations in aesthetics dentistry. In this study, we survey the modern literature in color matching techniques and discuss a logistic regression approach for computational prediction and selection of tooth color in direct restorations using dental composite resins. Test cases were obtained at the school of dentistry of Universidad Central de Venezuela and were further analyzed in a personal computer through the capabilities of the Wolfram language's built-in machine learning package in Mathematica. Altogether, our results exhibit significantly high accuracy rates in front of the classification problem under study, thus emphasizing the impact of this sort of models in current aesthetics dentistry.
Can Dynamic Conformal arc be an Option in Epidermoid Cervical Cancer Treatment?
Jacksson Sánchez, Nelly Muñoz, Luis Moreno Sánchez, Frank Montero
Cancer therapy & Oncology, Volume 1 Issue 1 - January 2016
Historically locally advanced cancer of the cervix has been treated with radiotherapy and brachytherapy and it was not until 1999 that the use of concurrent chemotherapy was formalized due to excellent results in terms of rate of overall and disease-free survival. Box technique in radiotherapy is the most widely known providing excellent results, with some variations as oblique fields, but greatly increasing irradiation potentially healthy tissue, leading to the higher proportion of own side effects of each treatment. Therefore present a radiant treatment planning mode Dynamic Conformal Arc for cervical carcinoma.Treatment with dynamic conformal arc achieves better conformation of tumor and area to be treated, avoiding unnecessary doses to organs at risk (OAR), compared to conventional four fields irradiation technique (box technique), further significantly reduces the treatment time.Dynamic Conformal Arc (DAT) technique in the pelvis reduces irradiation dose in the organs at risk, making a good coverage of the clinical area to be treated, further decreasing the side effects. It could be considered as an alternative to conventional treatment of 4 fields or to the impossibility of intensity modulated radiation therapy (IMRT).
Brachial Plexus Neurofibroma Treated with Volumetrically Modulated Arc Therapy (VMAT): A Case Report.
Luis Moreno Sánchez, Jacksson Sánchez, Moisés Dieguez
Cancer therapy & Oncology, Volume 2 Issue 2 - October 2016
Neurofibromatosis was first described in 1882 by Friedrich Daniel von Reckling hausen, a German pathologist. Neurofibroma is a benign peripheral nerve sheath tumor that consists of Schwann cells, associated or unassociated with axons, perineural cells, and fibroblasts. Whenever possible, the treatment of choice should be surgical, but the management depends on the location and growth pattern. We present the case of a patient with left axillary neurofibroma without neurofibromatosis (NF) in whom surgery was delayed due to involvement of the brachial plexus, so was sent to radiotherapy, planned and treated with volumetrically modulated arc therapy (VMAT).
Delay in fructose-mediated renaturalization process for irradiated triple-helix gelatin chains in aqueous medium
Jesús Enrique Dávila-Pérez, Rafael Martín-Landrove
Rev. LatinAm. Metal. Mat. 2016; 36 (2) [article in spanish]
The triple-helix denaturalization in gelatin as a result of its interaction with ionizing radiation and evolution of renaturalization process were studied in a hydrated mixture of 250 Bloom type B gelatin, fructose and tridistilled water (gelatin + fructose), which was irradiated with X-rays produced by a clinical linear accelerator with an operating potential of 6 MV in order to deliver a dose between 30 and 300 Gy. Irradiated and non-irradiated samples were analyzed with the use of mass spectroscopy and magnetic resonance imaging (MRI) with diffusion techniques. The mass spectroscopy results indicate that no formation of new compounds took place for delivered doses. With diffusion techniques in a magnetic resonance 1.5 T scanner a follow-up of macroscopic changes in gelatin + fructose mixture could be done as they correspond to a triple-helix denaturalization followed by its incomplete renaturalization, which is an indication that changes in the material porous structure at microscopic level happened with a renaturalization half life time of 12.46 +/- 0.15 h.
