Computer Image Analysis Method Research Articles

Automated 2D Segmentation of Prostate in T2-weighted MRI Scans

The prostate cancer is the second most frequent tumor amongst men. Statistics shows that biopsy reveals only 70-80% clinical cancer cases. Multiparametric magnetic resonance imaging (MRI) technique comes to play and is used to help to determine the location to perform a biopsy. With the aim to automating the biopsy localization, prostate segmentation has to be performed in magnetic resonance images. Computer image analysis methods play the key role here. The problem of automated prostate magnetic resonance (MR) image segmentation is burdened by the fact that MRI signal intensity is not standardized: field of view and image appearance is for a large part determined by acquisition protocol, field strength, coil profile and scanner type. Authors overview the most recent Prostate MR image segmentation challenge results and provide insights on T2-weighted MRI scan images automated prostate segmentation problem by comparing the best obtained automatic segmentation algorithms and applying them to 2D prostate segmentation case. The most important benefit of this research will have medical doctors involved in the management of the cancer.

Methods of high-throughput plant phenotyping for large-scale breeding and genetic experiments

Phenomics is a field of science at the junction of biology and informatics which solves the problems of rapid, accurate estimation of the plant phenotype; it was rapidly developed because of the need to analyze phenotypic characteristics in large scale genetic and breeding experiments in plants. It is based on using the methods of computer image analysis and integration of biological data. Owing to automation, new approaches make it possible to considerably accelerate the process of estimating the characteristics of a phenotype, to increase its accuracy, and to remove a subjectivism (inherent to humans). The main technologies of high-throughput plant phenotyping in both controlled and field conditions, their advantages and disadvantages, and also the prospects of their use for the efficient solution of problems of plant genetics and breeding are presented in the review.

Quantitative assessment of videolaryngostroboscopic images in patients with glottic pathologies

Introduction: Digital imaging techniques enable exploration of novel visualization modalities of the vocal folds during phonation and definition of parameters, facilitating more precise diagnosis of voice disorders.Aim: Application of computer vision algorithms for analysis of videolaryngostroboscopic (VLS) images aimed at qualitative and quantitative description of phonatory vibrations.Materials and methods: VLS examinations were conducted for 45 females, including 15 subjects with vocal nodules, 15 subjects with glottal incompetence, and 15 normophonic females. The recorded VLS images were preprocessed, the glottis area was segmented out, and the glottal cycles were identified. The glottovibrograms were built, and then the glottal area waveforms (GAW) were quantitatively described by computing the following parameters: open quotient (OQ), closing quotient (CQ), speed quotient (SQ), minimal relative glottal area (MRGA), and a new parameter termed closure difference index (CDI).Results: Profiles of the glottal widths assessed along the glottal length differentiated the study groups (P < 0.001). Moreover, it was shown that the OQ, CQ, CDI, and MRGA indices can be considered as viable parameters for quantifying kinematics of the vocal folds for normophonic subjects and patients with diagnosed vocal nodules and glottal incompetence (P < 0.001).Conclusions: Computer image processing and analysis methods applied to videolaryngostroboscopic images allow for their quantitative assessment. Computation of the size-related and time-related parameters characterizing glottic pathologies is of interest for evidence-based voice diagnostics. Results of the performed ROC curve analysis suggest that the evaluated parameters can distinguish patients with voice disorders from normophonic subjects.

Open source software for quantification of cell migration, protrusions, and fluorescence intensities.

Cell migration is frequently accompanied by changes in cell morphology (morphodynamics) on a range of spatial and temporal scales. Despite recent advances in imaging techniques, the application of unbiased computational image analysis methods for morphodynamic quantification is rare. For example, manual analysis using kymographs is still commonplace, often caused by lack of access to user-friendly, automated tools. We now describe software designed for the automated quantification of cell migration and morphodynamics. Implemented as a plug-in for the open-source platform, ImageJ, ADAPT is capable of rapid, automated analysis of migration and membrane protrusions, together with associated fluorescently labeled proteins, across multiple cells. We demonstrate the ability of the software by quantifying variations in cell population migration rates on different extracellular matrices. We also show that ADAPT can detect and morphologically profile filopodia. Finally, we have used ADAPT to compile an unbiased description of a "typical" bleb formed at the plasma membrane and quantify the effect of Arp2/3 complex inhibition on bleb retraction.

Neural image analysis for maturity classification of sewage sludge composted with maize straw

This study uses the methods of computer image analysis and neural modelling for the construction of classification models to identify the stage of early maturity in composted material based on sewage sludge and maize straw. The research material was produced in strictly controlled laboratory conditions with a six-chamber bioreactor. Samples of the material were subjected to image acquisition in visible light (VIS), ultraviolet light from the UV-A range and mixed light (MIX, VIS+UV-A). The acquired images were subjected to broad analysis. As a result the values of 46 parameters providing information about the colour and texture were obtained. The colour was analysed for the RGB, HSV and greyscale model. The texture analysis determined the grey level co-occurrence matrixes (GLCM). The parameters acquired from the image were the basis of train, validation and test sets which were used for the construction of neural classification models. The models were based on the MLP (Multilayer Perceptron) topology. The process of their construction went on in the iterative manner, where the potentially insignificant input parameters were eliminated by means of sensitivity analysis. Finally 21 such models were generated. The classification error for the best model in the MIX light was 1.56%. On the other hand, the models with the best accuracy in the UV-A and VIS light showed the error, which was 1.83% and 2.87% greater than the best model for the MIX light, respectively.

Deep Imaging: the next frontier in microscopy.

The microscope is the quintessential tool for discovery in cell biology. From its earliest incarnation as a tool to make the unseen visible, microscopes have been at the center of most revolutionizing developments in cell biology, histology and pathology. Major quantum leaps in imaging involved the dramatic improvements in resolution to see increasingly smaller structures, methods to visualize specific molecules inside of cells and tissues, and the ability to peer into living cells to study dynamics of molecules and cellular structures. The latest revolution in microscopy is Deep Imaging-the ability to look at very large numbers of samples by high-throughput microscopy at high spatial and temporal resolution. This approach is rooted in the development of fully automated high-resolution microscopes and the application of advanced computational image analysis and mining methods. Deep Imaging is enabling two novel, powerful approaches in cell biology: the ability to image thousands of samples with high optical precision allows every discernible morphological pattern to be used as a read-out in large-scale imaging-based screens, particularly in conjunction with RNAi-based screening technology; in addition, the capacity to capture large numbers of images, combined with advanced computational image analysis methods, has also opened the door to detect and analyze very rare cellular events. These two applications of Deep Imaging are revolutionizing cell biology.

Computational techniques for the assessment of fracture repair

The combination of high-resolution three-dimensional medical imaging, increased computing power, and modern computational methods provide unprecedented capabilities for assessing the repair and healing of fractured bone. Fracture healing is a natural process that restores the mechanical integrity of bone and is greatly influenced by the prevailing mechanical environment. Mechanobiological theories have been proposed to provide greater insight into the relationships between mechanics (stress and strain) and biology. Computational approaches for modelling these relationships have evolved from simple tools to analyze fracture healing at a single point in time to current models that capture complex biological events such as angiogenesis, stochasticity in cellular activities, and cell-phenotype specific activities. The predictive capacity of these models has been established using corroborating physical experiments. For clinical application, mechanobiological models accounting for patient-to-patient variability hold the potential to predict fracture healing and thereby help clinicians to customize treatment. Advanced imaging tools permit patient-specific geometries to be used in such models. Refining the models to study the strain fields within a fracture gap and adapting the models for case-specific simulation may provide more accurate examination of the relationship between strain and fracture healing in actual patients. Medical imaging systems have significantly advanced the capability for less invasive visualization of injured musculoskeletal tissues, but all too often the consideration of these rich datasets has stopped at the level of subjective observation. Computational image analysis methods have not yet been applied to study fracture healing, but two comparable challenges which have been addressed in this general area are the evaluation of fracture severity and of fracture-associated soft tissue injury. CT-based methodologies developed to assess and quantify these factors are described and results presented to show the potential of these analysis methods.

The Smooth Bootstrap Approach to the Distribution of a Shape in the Ferritic Stainless Steel AISI 434L Powders

The ferritic-austenitic stainless steel was obtained by sintering the mixture of ferritic stainless steel AISI 434L powders with different amount of additions: Mn, Ni and Si. The structure of obtained sintered samples was investigated by computer image analysis methods. In porous materials the character of the pore structure strongly effects on its mechanical properties. Accurate information about pores shape is important information for technological process and quality control of produced materials as well. The images of the porous microstructure were analyzed using ADCIS Aphelion analytical software. Analyzed pores have complex and irregular shapes and thus authors decided to appoint different shape factors to obtain its proper quantitative description. Data obtained from image analysis process were statistically analyzed. Authors used a special resampling approach known as smooth bootstrapping to smoothing cumulative empirical distributions. The results obtained during resampling procedure have been compared with raw data from verification set and guidelines for the application of the proposed approach have been formulated. The validity of the proposed approach was positively verified and it significantly improved quality of the results. The smoothing and imputing of data allow to avoid numerical artifacts that may arise during the classical statistical calculations on irregular data originated from image analysis obtained from sintered samples. Efficient, reliable and relatively fast method for accessing the distribution of any others quantitative parameters describing microstructure of the materials is very interesting proposal for wide spectrum of application.

Relationship between color and tannin content in sorghum grain: application of image analysis and artificial neural network

The relationship between sorghum grain color and tannin content was reported in several references. In this study, 33 phenotypes of sorghum grain differing in seed characteristics were collected and analyzed by Folin-Ciocalteu method. A computer image analysis method was used to determine the color characteristics of all 33 sorghum phenotypes. Two methods of multiple linear regression and artificial neural network (ANN) models were developed to describe tannin content in sorghum grain from three input parameters of color characteristics. The goodness of fit of the models was tested using R², MS error, and bias. The computer image analysis technique was a suitable method to estimate tannin through sorghum grain color strength. Therefore, the color quality of the samples was described according three color parameters: L* (lightness), a* (redness - from green to red) and b* (blueness - from blue to yellow. The developed regression and ANN models showed a strong relationship between color and tannin content of samples. The goodness of fit (in terms of R²), which corresponds to training the ANN model, showed higher accuracy of prediction of ANN compared with the equation established by the regression method (0.96 vs. 0.88). The ANN models in term of MS error showed lower residuals distribution than that of regression model (0.002 vs. 0.006). The platform of computer image analysis technique and ANN-based model may be used to estimate the tannin content of sorghum.

X-ray diffraction investigation of white layer development in hard-turned surfaces

Hard turning process is widely used for finishing the hardened surfaces of machine parts exposed to high load. Such parts can be gears, bearings, or short shafts. As in each machining process, hard turning may change the properties of the surface layer. In this research work, we deal with material structure changes developing in the bore of gears made of a material of 20MnCr5. Microscope images show that the structure of the material is altered on the surface of the bores due to the turning process under different conditions. In certain cases, a certain layer that cannot be etched develops, called a white layer. Surface layer modification may take place using different technological data. In this paper, we will prove that proper selection of these technological data (cutting speed, depth of cut, feed rate) may eliminate the formation of a white layer or may decrease its thickness. The fine structure of this surface layer was investigated using the X-ray diffraction method, whilst the thickness of the layer was measured by a computer image analysis method. In this paper, the different conditions for the development of the white layer are described, its composition and some of its properties are determined, and the relationship between technological data and the thickness of the white layer is investigated.

Effects of Different Kinds of Embedding Media on the Image Definition of Animal Fibers using Differential Interference Contrast Light Microscopy

In this study, the image definition of animal fibers captured by Axioskop 2 MOT PLUS differential interference contrast (DIC) light microscope using 1-bromonaphthalene, diethylene glycol, n-decyl alcohol, and ethylene glycol monomethyl ether as embedding medium, respectively, is compared with the image definition obtained using paraffin wax, the reference embedding medium. The effects of the four kinds of embedding media on the image definition of camel cashmere, yak cashmere, wool, and cashmere are also analyzed. Results show that topographical feature, scale structure, and pigmentation are better visualized when n-decyl alcohol is used as embedding medium compared to when the other kinds of embedding media are used. This meets the requirement of computer image analysis method for the identification of camel cashmereand yak cashmere fibers. Furthermore, the image definition of wool and cashmere is clearly improved when n-decyl alcohol is used as the embedding medium compared to when paraffin wax is used in the same condition.

Accelerating taxonomic discovery through automated character extraction

This paper discusses the following key messages. Taxonomy is (and taxonomists are) more important than ever in times of global change. Taxonomic endeavour is not occurring fast enough: in 250 years since the creation of the Linnean Systema Naturae, only about 20% of Earth’s species have been named. We need fundamental changes to the taxonomic process and paradigm to increase taxonomic productivity by orders of magnitude. Currently, taxonomic productivity is limited principally by the rate at which we capture and manage morphological information to enable species discovery. Many recent (and welcomed) initiatives in managing and delivering biodiversity information and accelerating the taxonomic process do not address this bottleneck. Development of computational image analysis and feature extraction methods is a crucial missing capacity needed to enable taxonomists to overcome the taxonomic impediment in a meaningful time frame.

The early signal substances induced by heat stress in brains of mice

To study the effects of early signal substances induced by heat stress in brains of Kunming mice, six-month-old mice (n = 72) were pretreated with heat stress and subsequent ischemia/reperfusion by clipping of their bilateral cervical common arteries for 7 min. According to different treatments, animals were randomly divided into four groups: (1) normal control group; (2) heat stress pretreatment followed by ischemia and reperfusion group (HS/IR); (3) ischemia and reperfusion group (IR); (4) heat stress group (HS). Animals in the later three groups were subdivided into 3 subgroups (1 day, 4 days, 14 days), respectively. The changes in the expression of cAMP response element binding protein (CREB) and calcitonin gene-related peptide (CGRP) were detected by immunohistochemistry and computer image analysis methods. The results showed that compared with the normal group, the expressions of CREB in the hippocampal CA1 region increased significantly in the HS, HS/IR and IR groups (P < 0.05). Compared to the normal group, heat stress could result in CGRP excretion and redistribution in the cerebrum, with the highest level in the 4 d HS/IR group. Following heat stress, CGRP immunoreactivity was observed in varicose fibers and neuronal perikarya within the CA1 region. The results indicate that heat stress can induce CREB expression, which in turn stimulates CGRP secretion.

Regional specificity of cerebrospinal fluid abnormalities in first episode schizophrenia

The timing and regional specificity of cerebrospinal fluid (CSF) enlargements have not been well described in schizophrenia. High-resolution magnetic resonance images and computational image analysis methods were used to localize cross-sectional changes in lateral ventricle and sulcal and subarachnoid CSF in first episode schizophrenia patients (51 males/21 females) and healthy subjects (37 males/41 females). Volumes were obtained for each lateral ventricle horn and regional differences identified by comparing the distances from the ventricular surfaces to the central core at anatomically matched locations. Extra-cortical CSF differences were compared by measuring the proportion of CSF voxels sampled from spatially homologous cortical surface points. Significant extra-cortical CSF enlargements were observed in first episode patients, where regional differences surrounded the temporal, anterior frontal and parietal cortices. Volume and ventricular surface analyses failed to show significant effects of diagnosis. However, interactions indicated dorsal superior horn expansions in female patients compared with same-sex controls. Since ventricular enlargements are widely reported in chronic patients, our observations at first episode suggest ventricular enlargement may progress after disease onset with early changes occurring around the dorsal superior horn. In contrast, sulcal and subarachnoid CSF increases may be manifest near or before the first episode but after brain development is complete, reflecting pronounced reductions in proximal brain tissue.

Cortical Thinning in Cingulate and Occipital Cortices in First Episode Schizophrenia

Postmortem studies examining discrete regions show reduced cortical thickness in schizophrenia. Computational image analysis methods allow spatially detailed cortical thickness measurements across the entire cortex in 3D, but have not addressed thickness changes in cingulate or other cortices bordering the medial walls of the cerebral hemispheres in first episode schizophrenia. Magnetic resonance images and cortical pattern matching methods were used to compare gray matter thickness, measured at sub-voxel resolution at thousands of spatially equivalent locations on the medial hemispheric surfaces, between 72 (51m/21f) first episode schizophrenia patients and 78 (37m/41f) healthy controls similar in age. Group differences were mapped in 3D, and their overall significance was confirmed by permutation testing. Patients with little or no prior antipsychotic medication treatment showed significant cortical thinning within cingulate, occipital and frontopolar cortices with no significant increases in any cortical location. Regional sex differences were observed with pronounced thinning in the left paracentral lobule and right posterior cingulate in male and female patients respectively compared to same sex controls. Cortical thinning may correspond to cytoarchitectural and neurochemical abnormalities observed in similar anatomic locations and may underlie systems-wise disturbances that include heteromodal association cortices, where cortical thinning has been previously observed in first episode schizophrenia.

Proposal, verification and comparison of three computer image analysis methods for detection and evaluation of colour glaucomatous changes within the optic disc of a human eye retina

The typical symptom of the human eye glaucoma is a rise and a progression of the bright area (named pallor area) within the retina blind spot. The image analysis manner proposed by the authors detects and suitably numerically describes the relative size of the representative pallor area in the colour digital image of the retina obtained by a suitable fundus camera connected with the computer. Three new different computer image analysis statistical methods for experimental diagnostic evaluation of the obtained characteristic data are proposed in this article: the quantile curves method, the neural net method and the probability density curves method. The quantile curves method is based on the graphical comparison of a relative representative pallor area size with its determined normal value. The neural net and probability density curves methods can automatically and objectively classify the investigated eyes in exactly defined glaucoma risk classes and diagnosed glaucoma with the rated probabilities of incorrect diagnosis determination. All mentioned methods are verified and mutually compared by their application to the large statistical sets of human retina images of various healthy and glaucomatous subjects.

Regional specificity of hippocampal volume reductions in first-episode schizophrenia

Hippocampal volume reductions are widely observed in schizophrenia. Some studies suggest anterior hippocampal regions are more susceptible and associated with frontal lobe dysfunctions, while others implicate posterior regions. Using high-resolution MR images and novel computational image analysis methods, we identified the hippocampal subregions most vulnerable to disease processes in 62 (45 m/17 f) first-episode schizophrenia patients compared to 60 (30 m/30 f) healthy controls, similar in age. The hippocampi were traced on coronal brain slices and hemispheric volumes were compared between diagnostic groups. Regional structural abnormalities were identified by comparing distances, measured from homologous hippocampal surface points to the central core of each individual's hippocampal surface model, between groups in 3D. CSF concentrations were also compared statistically at homologous hippocampal surface points to localize corresponding gray matter reductions. Significant bilateral hippocampal volume reductions were observed in schizophrenia irrespective of brain size corrections. Statistical mapping results, confirmed by permutation testing, showed pronounced left hemisphere shape differences in anterior and midbody CA1 and CA2 regions in patients. Significant CSF increases surrounding the hippocampus were observed in a similar spatial pattern in schizophrenia. Results confirm that hippocampal volume reductions are a robust neuroanatomical correlate of schizophrenia and are present by first episode. Mid- to antero-lateral hippocampal regions show pronounced volume changes and complementary increases in peri-hippocampal CSF, suggesting that these hippocampal regions are more susceptible to disease processes in schizophrenia. Targeting regional hippocampal abnormalities may help dissociate schizophrenia patients from other groups exhibiting global hippocampal volume changes, and better focus systems-level pathophysiological hypotheses.

Characterization via image analysis of cross-over trajectories and inhomogeneity in twin wire arc spraying

By using a computer image analysis method, the microstructure of deposits arc sprayed with steel and copper wires as anode and cathode, respectively, were investigated and it was demonstrated that there exists a cross-over of the particle trajectories in the plume. Along the cross-section of the deposit, flattened particles from the anode and cathode wires were distributed in an asymmetric way about the centreline of the plume and varied gradually. For the deposits sprayed by an open nozzle system and a closed nozzle system with converging orifice, the locations of the maximum particle density of the anode and cathode materials did not coincide with each other and were not located at the centreline of the plume. However, for the deposit sprayed with a closed nozzle system and a converging–diverging orifice, such phenomenon was not obvious. By quantitative evaluation of the lamella structure of the deposits, it was found that the lamella thickness distribution followed a similar rule with that of the material content distribution. The above results indicate that in the twin wire arc spray process asymmetry in droplet formation and atomization results in remarkable inhomogeneity in coating microstructure.

Computational morphology of the lung and its virtual imaging

The author proposes an entirely new approach called ‘virtual imaging’ of an organ based on ‘computational morphology’. Computational morphology describes mathematically design as principles of an organ structure to generate the organ model via computer, which can be called virtual organ. Virtual imaging simulates image data using the virtual organ. The virtual organ is divided into cubic voxels, and the CT value or other intensity value for each voxel is calculated according to the tissue properties within the voxel. The validity of the model is examined by comparing virtual images with clinical images. Computational image analysis methods can be developed based on validated models. In this paper, computational anatomy of the lung and its virtual X-ray imaging are introduced.

Computer image analysis for prediction of carcass composition from cross-sections of Japanese Black steers.

Carcasses from Japanese Black steers were used to obtain prediction equations for carcass composition from information derived by computer image analysis of carcass cross-section images. The total weights of lean, fat, and bone were obtained from the left sides of 55 carcasses (Data Set I) and 18 carcasses (Data Set II) by physical dissection. The information such as total lean, fat, and bone areas in the cross-sections; muscle area, muscle circumference, short and long radius axis lengths, and direction of long radius axis; and geometric distance between any two muscle centers of gravity was obtained by scanning and image analysis of pictures of the cross-sections of the beef side at the 6th/7th rib interface. The coefficients of determination of the multiple regression equations estimated from Data Set I for kilograms of lean, fat, and bone were 0.76, 0.82, and 0.69, respectively, whereas for the percentages of lean, fat, and bone they were 0.57, 0.66, and 0.42, respectively. The multiple regression equations from Data Set I was applied to Data Set II in order to test the applicability of the prediction equations obtained. The correlation coefficients between the value predicted by the multiple regression equation and the measurement obtained by physical dissection for kilograms of lean, fat, and bone were 0.71, 0.72, and 0.70, respectively, whereas those for the percentages of lean, fat, and bone were 0.63, 0.44, and 0.29, respectively. The results indicate that the information obtained from the carcass cross-sections by the computer image analysis method can be used to predict carcass composition in Japanese Black steers.