Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

National Institute of Standards and Technology における超伝導研究及び生活

Pattern recognition

Learning Deconvolution Network for Semantic Segmentation

This work explores and compares the plethora of metrics for the performance evaluation of object-detection algorithms. Average precision (AP),for instance, is a popular metric for evaluating the accuracy of object detectors by estimating the area under the curve (AUC) of the precision × recall relationship. Depending on the point interpolation used in the plot, two different AP variants can be defined and, therefore, different results are generated. AP has six additional variants increasing the possibilities of benchmarking. The lack of consensus in different works and AP implementations is a problem faced by the academic and scientific communities. Metric implementations written in different computational languages and platforms are usually distributed with corresponding datasets sharing a given bounding-box description. Such projects indeed help the community with evaluation tools, but demand extra work to be adapted for other datasets and bounding-box formats. This work reviews the most used metrics for object detection detaching their differences, applications, and main concepts. It also proposes a standard implementation that can be used as a benchmark among different datasets with minimum adaptation on the annotation files.

A Survey on Performance Metrics for Object-Detection Algorithms

This article presents an automatic identification method of mycobacterium tuberculosis with conventional microscopy images based on Red and Green color channels using global adaptive threshold segmentation. Differing from fluorescence microscopy, in the conventional microscopy the bacilli are not easily distinguished from the background. The key to the bacilli segmentation method employed in this work is the use of Red minus Green (R-G) images from RGB color format. In this image, the bacilli appear as white regions on a dark background. Some artifacts are present in the (R-G) segmented image. To remove them we used morphological, color and size filters. The best sensitivity achieved was about 76.65%. The main contribution of this work was the proposal of the first automatic identification method of tuberculosis bacilli for conventional light microscopy.

Automatic identification of mycobacterium tuberculosis with conventional light microscopy

In the past years a lot of effort has been made in the field of face detection. The human face contains important features that can be used by vision-based automated systems in order to identify and recognize individuals. Face location, the primary step of the vision-based automated systems, finds the face area in the input image. An accurate location of the face is still a challenging task. Viola-Jones framework has been widely used by researchers in order to detect the location of faces and objects in a given image. Face detection classifiers are shared by public communities, such as OpenCV. An evaluation of these classifiers will help researchers to choose the best classifier for their particular need. This work focuses of the evaluation of face detection classifiers minding facial landmarks. Keywords—Face datasets, face detection, facial landmarking, haar wavelets, Viola-Jones detectors.

/pdf/evaluation-of-haar-cascade-classifiers-designed-for-face-4do9n8kyc3.pdf

Evaluation of Haar Cascade Classifiers Designed for Face Detection

Introduction According to the Global TB control report of 2013, “Tuberculosis (TB) remains a major global health problem. In 2012, an estimated 8.6 million people developed TB and 1.3 million died from the disease. Two main sputum smear microscopy techniques are used for TB diagnosis: Fluorescence microscopy and conventional microscopy. Fluorescence microscopy is a more expensive diagnostic method because of the high costs of the microscopy unit and its maintenance. Therefore, conventional microscopy is more appropriate for use in developing countries. Methods This paper presents a new method for detecting tuberculosis bacillus in conventional sputum smear microscopy. The method consists of two main steps, bacillus segmentation and post-processing. In the first step, the scalar selection technique was used to select input variables for the segmentation classifiers from four color spaces. Thirty features were used, including the subtractions of the color components of different color spaces. In the post-processing step, three filters were used to separate bacilli from artifact: a size filter, a geometric filter and a Rule-based filter that uses the components of the RGB color space. Results In bacillus identification, an overall sensitivity of 96.80% and an error rate of 3.38% were obtained. An image database with 120-sputum-smear microscopy slices of 12 patients with objects marked as bacillus, agglomerated bacillus and artifact was generated and is now available online. Conclusions The best results were obtained with a support vector machine in bacillus segmentation associated with the application of the three post-processing filters.

/pdf/automatic-identification-of-tuberculosis-mycobacterium-21i1aka2ul.pdf

Automatic identification of tuberculosis mycobacterium

Tuberculosis (TB) has been included among the top ten leading causes of death worldwide. Since 2008, several investigations have been developed by scientific community for automatic detection of mycobacterium tuberculosis (MT) in light field microscopy images. Those authors have applied techniques as digital images processing and machine learning, combined with different sizes of datasets and color spaces such as: RGB, HSI and Lab. More recently, deep learning has been used, working with grayscale, though. We present a method for automatic classification of light field smear microscopy patches using RGB, R-G and grayscale patches versions as inputs of a Convolutional Neural Networks (CNN) model for identifying MT. A dataset of negative and positive patches was created for training three CNN Models applying regularization techniques, like: dataset normalization, data augmentation and dropout. The best result in the patch classification test was reached using R-G input version and a three convolutional layers Model with regularization, getting an Area under ROC Curve of 99%.

Automatic classification of light field smear microscopy patches using Convolutional Neural Networks for identifying mycobacterium tuberculosis

According to the World Health Organization, cardiovascular diseases are the leading cause of death worldwide, accounting for 17.3 million deaths per year, a number that is expected to grow to more than 23.6 million by 2030. Most cardiac pathologies involve the left ventricle; therefore, estimation of several functional parameters from a previous segmentation of this structure can be helpful in diagnosis. Manual delineation is a time consuming and tedious task that is also prone to high intra and inter-observer variability. Thus, there exists a need for automated cardiac segmentation method to help facilitate the diagnosis of cardiovascular diseases. In this work we propose a deep fully convolutional neural network architecture to address this issue and assess its performance. The model was trained end to end in a supervised learning stage from whole cardiac MRI images input and ground truth to make a per pixel classification. For its design, development and experimentation was used Caffe deep learning framework over an NVidia Quadro K4200 Graphics Processing Unit. The net architecture is: Conv64-ReLU (2x) – MaxPooling – Conv128-ReLU (2x) – MaxPooling – Conv256-ReLU (2x) – MaxPooling – Conv512-ReLu-Dropout (2x) – Conv2-ReLU – Deconv – Crop – Softmax. Training and testing processes were carried out using 5-fold cross validation with short axis cardiac magnetic resonance images from Sunnybrook Database. We obtained a Dice score of 0.92 and 0.90, Hausdorff distance of 4.48 and 5.43, Jaccard index of 0.97 and 0.97, sensitivity of 0.92 and 0.90 and specificity of 0.99 and 0.99, overall mean values with SGD and RMSProp, respectively.

Cicero Ferreira Fernandes Costa Filho

Papers

Automatic identification of mycobacterium tuberculosis with conventional light microscopy

Evaluation of Haar Cascade Classifiers Designed for Face Detection

Automatic identification of tuberculosis mycobacterium

Automatic classification of light field smear microscopy patches using Convolutional Neural Networks for identifying mycobacterium tuberculosis

Left ventricle segmentation in cardiac MRI images using fully convolutional neural networks