Search

Accurate vertebrae segmentation from medical images plays an important role in clinical tasks of surgical planning, diagnosis, kyphosis, scoliosis, degenerative disc disease, spondylolisthesis, and post-operative assessment. Although the structures of bone have high contrast in medical images, vertebrae segmentation is a challenging task due to its complex structure, abnormal spine curves, and unclear boundaries. In recent years, deep learning has been widely applied in the segmentation of vertebrae images. In this paper, towards a robust and automatic segmentation system, we present an overlapping patch-based convNet (OP-convNet) model for automatic vertebrae CT images segmentation. Due to the greater memory and processing costs associated with 3D convolutional neural networks, as well as the risk of over-fitting, we employ overlapping patches in segmentation tasks using 2D convNet. In the proposed vertebrae segmentation method, OP-convNet effectively keeps the local information contained in CT images. We divide CT image slices into equal-sized square overlapping patches and applied the RUS-function on these patches for class balancing to minimize computational requirements. Then, these patches are input into the model along with their corresponding ground truth patches. This method has been evaluated on publicly available CT images from the MICCAI CSI workshop challenge. The results indicate that OP-convNet has precision (PRE) of 90.1%, specificity (SPE) of 99.4%, accuracy (ACC) of 98.8%, F-score of 90.1% in terms of the patch-based classification accuracy, and BF-score of 90.2%, sensitivity (SEN) of 90.3%, Jaccard index (JAC) of 82.3%, dice similarity score (DSC) of 89.9% in terms of the segmentation accuracy that outperform previous methods across all metrics

This study aims to analyze the potential of the computer vision (CV) approach to classify eight canola varieties. The input images of eight canola varieties were CON-I, CON-II, CON-III, Pakola, Canola Raya, Rainbow, PARC Canola Hybrid, and Tarnab-III. A digital camera acquired these images on an open sunny day without any complex laboratory setup. First-order histogram features, second-order statistical texture features, binary features, spectral features of three bands were, blue (B), green (G), and red (R), were employed in the artificial neural network (ANN). A 10-fold stratified cross-validation method was used for classification. The best results with accuracy ranging from 95% to 98% observed when the data of regions of interest (512 × 512) deployed to the classifier.

The main objective of this study was to assess the machine vision (MV) techniques to classify six Asian rice varieties commonly named as Kachi-Kainat, Kachi-Toota, Kainat-Pakki, Super-Basmati-Kachi, Super-Basmati-Pakki, and Super-Maryam-Kainat (A1, A2, A3, A4, A5, and A6), mainly cultivated in Pakistan, China, India, Bangladesh, and neighboring countries. The sample of each selected rice variety contained 1800 grains, giving a total of 10800 (1800 × 6) grain samples. A cell phone camera captured the actual field digital images dataset in an open climate. All the captured images were enhanced and converted into the standard 8-bit gray-scale format. Six radius-based non-overlapping regions of interest (ROI’s) were taken on each captured image inducing a total of 3600 (6 × 600) ROI’s image dataset. We have extracted Binary (B), Histogram (H), and Texture (T) features from each image. We converted these forty-three features for each image into 154800 (43 × 3600) feature vector (FV) space to discriminate rice varieties. After optimizing the FV, five MV classifiers, namely; LMT Tree (LMT-T), Meta Classifier via Regression (MCR), Meta Bagging (MB), Tree J48 (T-J48), and Meta Attribute Select Classifier (MAS-C), were deployed attaining the classification accuracies as 97.4%, 97.0%, 96.3%, 95.74%, and 95.2%, respectively. The maximum overall accuracy (MOA) observed was 97.4% by LMT-Tree.

The objective of this study was to observe the potential of machine vision (MV) approach for the classification of eight citrus varieties. The leaf images of eight citrus varieties that were grapefruit, Moussami, Malta, Lemon, Kinow, Local lemon, Fuetrells, and Malta Shakri. These were acquired by a digital camera in an open environment without any complex laboratory setup. The acquired digital images dataset was transformed into the multi-feature dataset that was the combination of binary, histogram, texture, spectral, rotational, scalability and translational (RST) invariant features. For each citrus leaf image, total 57 multi-features were acquired on every non-overlapping region of interest (ROI), i.e. (32x32), (64x64), (128x128), and (256x256). Furthermore, the optimized 15 features using the supervised correlation-based feature selection (CFS) technique were acquired. The optimized multi-features dataset to different MV classifiers namely Multilayer Perceptron (MLP), Random Forest (RF), J48 and Naïve Bayes using10-fold cross-validation method were plugged-in. The results produced by MLP presented an average overall accuracy of 98.14% on ROIs (256x256) outperforming the other classifiers. The classification accuracy values by MLP on the eight citrus leaf varieties, namely; Grapefruit, Moussami, Malta, Lemon, Kinow, Local lemon, Fuetrells, and Malta Shakri were observed 98%, 98.75%, 99.25%, 97.5%, 97%, 95.87%, 95.5%, and 99.37% respectively.

Software development is a complex task, and the introduction of multi‐site development teams spread across the globe makes it even harder. Requirement changes during the software developme...

An accurate vertebrae segmentation in the spine is an essential pre-requisite in many applications of image-based spine assessment, surgical planning, clinical diagnostic treatment, and biomechanical modeling. In this paper, we present the stacked sparse autoencoder (SSAE) model for the segmentation of vertebrae from CT images. After the preprocessing step, we extracted overlapped patches from the vertebrae CT images as the inputs of our proposed model. The SSAE model was trained in an unsupervised way to learn high-level features from the input pixels of the unlabeled images patch. To improve the discriminability of the learned features, we further refined the feature representation in a supervised fashion and fine-tuned the whole model by using the feedforward neural network parameters for classifying the overlapped patches. We then validated our model on a publicly available MICCAI CSI2014 dataset and found that our model outperforms the other state-of-the-art methods.

Precise automatic vertebra segmentation in computed tomography (CT) images is important for the quantitative analysis of vertebrae-related diseases but remains a challenging task due to high variation in spinal anatomy among patients. In this paper, we propose a deep learning approach for automatic CT vertebra segmentation named patch-based deep belief networks (PaDBNs). Our proposed PaDBN model automatically selects the features from image patches and then measures the differences between classes and investigates performance. The region of interest (ROI) is obtained from CT images. Unsupervised feature reduction contrastive divergence algorithm is applied for weight initialization, and the weights are optimized by layers in a supervised fine-tuning procedure. The discriminative learning features obtained from the steps above are used as input of a classifier to obtain the likelihood of the vertebrae. Experimental results demonstrate that the proposed PaDBN model can considerably reduce computational cost and produce an excellent performance in vertebra segmentation in terms of accuracy compared with state-of-the-art methods.

Automatic vertebra segmentation is a challenging task from CT images due to anatomically complexity, shape variation and vertebrae articulation with each other. Deep Learning is a machine learning paradigm that focuses on deep hierarchical learning modeling of input data. In this paper, we propose a novel approach of automatic vertebrae segmentation from computed tomography (CT) images by using deep belief networks (BDNs) modeling. Using the DBN model, the contexture features of vertebra from CT images are extracted automatically by an unsupervised pattern called pre-training and followed by supervised training called back-propagation algorithm; then segmentation the vertebra from other abdominal structure. To evaluate the performance, we computed the overall accuracy (94.2%), sensitivity (83.2%), specificity (94.8%) and mean Dice coefficients (0.85 ± 0.03) for segmentation evaluation. Experimental results show that our proposed model provides a more accuracy in vertebra segmentation compared to the previous state of art methods.

Deep learning methods have been successfully applied to feature learning in medical applications. In this paper, we proposed a Deep Stacked Auto-Encoder (DSAE) for liver segmentation from CT images. The proposed method composes of three major steps. First, we learned the features with unlabeled data using the auto encoder. Second, these features are fine-tuned to classify the liver among other abdominal organs. Using this technique we got promising classification results on 2D CT data. This classification of the data helps to segment the liver from the abdomen. Finally, segmentation of a liver is refined by post processing method. We focused on the high accuracy of the classification task because of its effect on the accuracy of a better segmentation. We trained the deep stacked auto encoder (DSAE) on 2D CT images and experimentally shows that this method has high classification accuracy and can speed up the clinical task to segment the liver. The mean DICE coefficient is noted to be 90.1% which is better than the state of art methods.

Green computing concept is to improve environmental condition. The main aim of green computing is to reduce toxic materials. We systematically analyze its energy consumption which is based on types of services and obtain the conditions to facilitate green cloud computing to save overall energy consumption in this system. Today it is the major issue to prepare such equipments by which we achieve efficient energy and to minimize of e-waste and use of non toxic chemicals/materials in preparation of e-equipments. We can implement green computing in computer’s fields as CPU servers and other peripheral devices (mobile devices). By using green computing we can reduce resources consumption and disposal of electric waste (e-waste). It has been seen that computers and other electronics devices are increasing day by day, so the amount of electricity consumed by them is also increasing. In this way the percentage of co2 in the atmosphere is also increasing. The other toxic materials which are used in computer/electronics industry are also harmful for environment. In this paper, we will elaborate comprehensively survey the concepts and architecture of green computing, as well as its heat and energy consumption issues. Their pros and cons are discussed for each green computing strategy with its friendly approach towards atmosphere. Green computing can facilitate us to safe, secure place and healthy environment all over in the world. This paper will help us to take some initiatives currently under in the field of computers/electronics industry and new ways to save vast amounts of energy which is wasted in very large scale.