Juan D. García-Arteaga

An automatic search of Alzheimer patterns using a nonnegative matrix factorization

This paper presents a fully automatic method that condenses relevant morphometric information from a database of magnetic resonance images (MR) labeled as either normal (NC) or Alzheimers disease (AD). The proposed method generates class templates using Nonnegative Matrix Factorization (NMF) which will be used to develop an NC/AD classi cator. It then nds regions of interest (ROI) with discerning inter-class properties. by inspecting the di erence volume of the two class templates. From these templates local probability distribution functions associated to low level features such as intensities, orientation and edges within the found ROI are calculated. A sample brain volume can then be characterized by a similarity measure in the ROI to both the normal and the pathological templates. These characteristics feed a simple binary SVM classi er which, when tested with an experimental group extracted from a public brain MR dataset (OASIS), reveals an equal error rate measure which is better than the state-of-the-art tested on the same dataset (0:9 in the former and 0:8 in the latter).

Virtual slide mosaicing using feature descriptors and a registration consistency measure

The advent of low-cost digital storage and automated microscope motor stages has made it possible to sequen- tially capture the full set of elds of view (FOV) covering a slide sample in a process commonly resulting in hundreds or even thousands of images. Aligning or verifying manually the alignment of thousands of images is an unrealistically labor intensive task that is, however, a fundamental step in the analysis of virtual-slides: For virtual-Slide creation the automation of the aligning process is not a mere an option but an absolute necessity. In the present work we propose the use of feature based methods and local consistency measures to improve the creation of mosaics from individual images captured with an in-house built microscope.

Characterizing Architectural Distortion in Mammograms by Linear Saliency

Architectural distortion (AD) is a common cause of false-negatives in mammograms. This lesion usually consists of a central retraction of the connective tissue and a spiculated pattern radiating from it. This pattern is difficult to detect due the complex superposition of breast tissue. This paper presents a novel AD characterization by representing the linear saliency in mammography Regions of Interest (ROI) as a graph composed of nodes corresponding to locations along the ROI boundary and edges with a weight proportional to the line intensity integrals along the path connecting any pair of nodes. A set of eigenvectors from the adjacency matrix is then used to extract discriminant coefficients that represent those nodes with higher salient lines. A dimensionality reduction is further accomplished by selecting the pair of nodes with major contribution for each of the computed eigenvectors. The set of main salient lines is then assembled as a feature vector that inputs a conventional Support Vector Machine (SVM). Experimental results with two benchmark databases, the mini-MIAS and DDSM databases, demonstrate that the proposed linear saliency domain method (LSD) performs well in terms of accuracy. The approach was evaluated with a set of 246 RoI extracted from the DDSM (123 normal tissues and 123 AD) and a set of 38 ROI from the mini-MIAS collections (19 normal tissues and 19 AD) respectively. The classification results showed respectively for both databases an accuracy rate of 89 % and 87 %, a sensitivity rate of 85 % and 95 %, and a specificity rate of 93 % and 84 %. Likewise, the area under curve (Az) of the Receiver Operating Characteristic (ROC) curve was 0.93 for both databases.

Morphometry-based comparison of relevant brain regions for Alzheimer's disease detection

Morphometry based methods allow the detection of subtle anatomical differences in the Magnetic Resonance Images (MRI) between healthy subjects and Alzheimers Disease (AD) patients. However, anatomical volumes are rarely used for clinical diagnosis as the changes induced by AD are hard to differentiate from normal brain aging. We present a morphometry method which uses brain models generated using Nonnegative Matrix Factorization (NMF) characterized by signatures calculated from perceptual features such as intensities, edges or orientations, of salient regions. The Earth Movers Distance (EMD), a robust measure of the cost of transforming signature A into signature B, is used to calculate volume-models distances. The discerning power of these distances is tested by using them as features for a Support Vector Machine classifier. This work shows the usefulness of the EMD as a metric in medical image applications as it has proven to be robust to bin selection, takes into account cross bin relations, and allows high sensitivity with lower dimensionality. This method is able to find discerning regions which, besides aiding in classification, may provide new insights of the diseases development.

Modelo para el análisis de la mortalidad en Colombia 2000-2012

OBJECTIVE To propose and evaluate a model for fitting and forecasting the mortality rates in Colombia that allows analyzing the trends by age, sex, region and cause of death. METHODOLOGY The national death registries were used as primary source of analysis. The data was pre-processed recodifying the cause of death and redistributing the garbage codes. The forecast model was formulated as a linear approximation with a set of variables of interest, in particular the population and gross domestic product (GDP) by region. RESULTS As study case we took the mortality under 5 years old, it decreased steadily since 2000 at the national level and at most of the regions. The predictive power of the proposed methodology was tested by fitting the model with the data from 2000 to 2011, the forecast for 2012 was compared with the actual rate, and these results show the model is reliable enough for most of the region-cause combinations. CONCLUSIONS The proposed methodology and model have the potential to become an instrument to guide health spending priorities using some kind of evidence.

A lymphocyte spatial distribution graph-based method for automated classification of recurrence risk on lung cancer images

Tumor-infiltrating lymphocytes occurs when various classes of white blood cells migrate from the blood stream towards the tumor, infiltrating it. The presence of TIL is predictive of the response of the patient to therapy. In this paper, we show how the automatic detection of lymphocytes in digital H and E histopathological images and the quantitative evaluation of the global lymphocyte configuration, evaluated through global features extracted from non-parametric graphs, constructed from the lymphocytes’ detected positions, can be correlated to the patient’s outcome in early-stage non-small cell lung cancer (NSCLC). The method was assessed on a tissue microarray cohort composed of 63 NSCLC cases. From the evaluated graphs, minimum spanning trees and K-nn showed the highest predictive ability, yielding F1 Scores of 0.75 and 0.72 and accuracies of 0.67 and 0.69, respectively. The predictive power of the proposed methodology indicates that graphs may be used to develop objective measures of the infiltration grade of tumors, which can, in turn, be used by pathologists to improve the decision making and treatment planning processes.

Quantifying stimulus-response rehabilitation protocols by auditory feedback in Parkinson's disease gait pattern

External auditory cues stimulate motor related areas of the brain, activating motor ways parallel to the basal ganglia circuits and providing a temporary pattern for gait. In effect, patients may re-learn motor skills mediated by compensatory neuroplasticity mechanisms. However, long term functional gains are dependent on the nature of the pathology, follow-up is usually limited and reinforcement by healthcare professionals is crucial. Aiming to cope with these challenges, several researches and device implementations provide auditory or visual stimulation to improve Parkinsonian gait pattern, inside and outside clinical scenarios. The current work presents a semiautomated strategy for spatio-temporal feature extraction to study the relations between auditory temporal stimulation and spatiotemporal gait response. A protocol for auditory stimulation was built to evaluate the integrability of the strategy in the clinic practice. The method was evaluated in transversal measurement with an exploratory group of people with Parkinson’s (n = 12 in stage 1, 2 and 3) and control subjects (n =6). The result showed a strong linear relation between auditory stimulation and cadence response in control subjects (R=0.98 ±0.008) and PD subject in stage 2 (R=0.95 ±0.03) and stage 3 (R=0.89 ±0.05). Normalized step length showed a variable response between low and high gait velocity (0.2> R >0.97). The correlation between normalized mean velocity and stimulus was strong in all PD stage 2 (R>0.96) PD stage 3 (R>0.84) and controls (R>0.91) for all experimental conditions. Among participants, the largest variation from baseline was found in PD subject in stage 3 (53.61 ±39.2 step/min, 0.12 ± 0.06 in step length and 0.33 ± 0.16 in mean velocity). In this group these values were higher than the own baseline. These variations are related with direct effect of metronome frequency on cadence and velocity. The variation of step length involves different regulation strategies and could need others specific external cues. In conclusion the current protocol (and their selected parameters, kind of sound time for training, step of variation, range of variation) provide a suitable gait facilitation method specially for patients with the highest gait disturbance (stage 2 and 3). The method should be adjusted for initial stages and evaluated in a rehabilitation program.

Nuclei graph local features for basal cell carcinoma classification in whole slide images

Evidence based medicine aims to provide a quantifiable framework to support cancer optimal treatment selection. Pathological examination is the main evidence used in medical management, yet the level of quantification is low and highly dependent on the examiner expertise. This paper presents and evaluates a method to extract graph based topological features from skin tissue images to identify cancerous regions associated to basal cell carcinoma. The graph features constitute a quantitative measure of the architectural tissue organization. Results show that graph topological features extracted from a nuclei based distance graph, particularly those related to local density, have a high predictive value in the automated detection of basal cell carcinoma. The method was evaluated using a leave-one-out validation scheme in a set of 9 skin Whole Slide Images obtaining a 0.76 F-score in distinguishing basal cell carcinoma regions in skin tissue whole slide images.

Finding regional models of the Alzheimer disease by fusing information from neuropsychological tests and structural MR images

Initial diagnosis of Alzheimers disease (AD) is based on the patients clinical history and a battery of neuropsy-chological tests. This work presents an automatic strategy that uses Structural Magnetic Resonance Imaging (MRI) to learn brain models for different stages of the disease using information from clinical assessments. Then, a comparison of the discriminant power of the models in different anatomical areas is made by using the brain region of the models as a reference frame for the classification problem, by using the projection into the AD model a Receiver Operating Characteristic (ROC) curve is constructed. Validation was performed using a leave- one-out scheme with 86 subjects (20 AD and 60 NC) from the Open Access Series of Imaging Studies (OASIS) database. The region with the best classification performance was the left amygdala where it is possible to achieve a sensibility and specificity of 85% at the same time. The regions with the best performance, in terms of the AUC, are in strong agreement with those described as important for the diagnosis of AD in clinical practice.

Image superresolution of cytology images using wavelet based patch search

Telecytology is a new research area that holds the potential of significantly reducing the number of deaths due to cervical cancer in developing countries. This work presents a novel super-resolution technique that couples high and low frequency information in order to reduce the bandwidth consumption of cervical image transmission. The proposed approach starts by decomposing into wavelets the high resolution images and transmitting only the lower frequency coefficients. The transmitted coefficients are used to reconstruct an image of the original size. Additional details are added by iteratively replacing patches of the wavelet reconstructed image with equivalent high resolution patches from a previously acquired image database. Finally, the original transmitted low frequency coefficients are used to correct the final image. Results show a higher signal to noise ratio in the proposed method over simply discarding high frequency wavelet coefficients or replacing directly down-sampled patches from the image-database.