Hybrid Nature Inspired SMO-GBM Classifier for Exudate Classification on Fundus Retinal Images

Background: The diabetic retinopathy can result in loss of vision if not detected in the earlier stages. Exudates are the lesions which play a crucial role in early diagnosis of diabetic retinopathy. The localization of exudates lesions with high values of performance metrics is complicated due to presence of blood vessels and other noisy artifacts. Method: We present computer aided system for classification of retinal fundus images using a novel nature inspired spider monkey optimization for parameter tuning of gradient boosting machines classifier. The image enhancement has been performed with histogram equalization and contourlet transform. The pixels belonging to optic disc region are detected and eliminated using circular Hough transform and Otsu's segmentation method. We have employed Kirsch's matrices for blood vessel detection. The GLCM based feature vector extraction has been employed for textural features. The classification has been performed with hybrid SMO-GBM classifier. Result: We have utilized the STARE database for validation of proposed technique. The proposed system can effectively classify entire image set from test data. The SMO-GBM classifier can further sub-segregate into sub classes with an average accuracy of 97.5%. Conclusion: The proposed approach provides detection and grading of diabetic retinopathy. The abnormality is further categories as soft, moderate and severe. The hybrid SMO-GBM classifier yields a better statistical metrics than the existing exudates classification approaches.     

Automatic Detection and Classification of Mammograms Using Improved Extreme Learning Machine with Deep Learning

Background and objectiveBreast cancer, the most intrusive form of cancer affecting women globally. Next to lung cancer, breast cancer is the one that provides a greater number of cancer deaths among women. In recent times, several intelligent methodologies were come into existence for building an effective detection and classification of such noxious type of cancer. For further improving the rate of early diagnosis and for increasing the life span of victims, optimistic light of research is essential in breast cancer classification. Accordingly, a new customized method of integrating the concept of deep learning with the extreme learning machine (ELM), which is optimized using a simple crow-search algorithm (ICS-ELM). Thus, to enhance the state-of-the-art workings, an improved deep feature-based crow-search optimized extreme learning machine is proposed for addressing the health-care problem. The paper pours a light-of-research on detecting the input mammograms as either normal or abnormal. Subsequently, it focuses on further classifying the type of abnormal severities i.e., benign type or malignant.Materials and methodsThe digital mammograms for this work are taken from the Curated Breast Imaging Subset of DDSM (CBIS-DDSM), Mammographic Image Analysis Society (MIAS), and INbreast datasets. Herein, the work employs 570 digital mammograms (250 normal, 200 benign and 120 malignant cases) from CBIS-DDSM dataset, 322 digital mammograms (207 normal, 64 benign and 51 malignant cases) from MIAS database and 179 full-field digital mammograms (66 normal, 56 benign and 57 malignant cases) from INbreast dataset for its evaluation. The work utilizes ResNet-18 based deep extracted features with proposed Improved Crow-Search Optimized Extreme Learning Machine (ICS-ELM) algorithm.ResultsThe proposed work is finally compared with the existing Support Vector Machines (RBF kernel), ELM, particle swarm optimization (PSO) optimized ELM, and crow-search optimized ELM, where the maximum overall classification accuracy is obtained for the proposed method with 97.193% for DDSM, 98.137% for MIAS and 98.266% for INbreast datasets, respectively.ConclusionThe obtained results reveal that the proposed Computer-Aided-Diagnosis (CAD) tool is robust for the automatic detection and classification of breast cancer.     

Automated Detection of B Cell and T Cell Acute Lymphoblastic Leukaemia Using Deep Learning

PurposeLeukaemia is diagnosed conventionally by observing the peripheral blood and bone marrow smear using a microscope and with the help of advanced laboratory tests. Image processing-based methods, which are simple, fast, and cheap, can be used to detect and classify leukemic cells by processing and analysing images of microscopic smear. The proposed study aims to classify Acute Lymphoblastic Leukaemia (ALL) by Deep Learning (DL) based techniques.ProceduresThe study used Deep Convolutional Neural Networks (DNNs) to classify ALL according to WHO classification scheme without using any image segmentation and feature extraction that involves intense computations. Images from an online image bank of American Society of Haematology (ASH) were used for the classification.FindingsA classification accuracy of 94.12% is achieved by the study in isolating the B-cell and T-cell ALL images using a pretrained CNN AlexNet as well as LeukNet, a custom-made deep learning network designed by the proposed work. The study also compared the classification performances using three different training algorithms.ConclusionsThe paper detailed the use of DNNs to classify ALL, without using any image segmentation and feature extraction techniques. Classification of ALL into subtypes according to the WHO classification scheme using image processing techniques is not available in literature to the best of the knowledge of the authors. The present study considered the classification of ALL only, and detection of other types of leukemic images can be attempted in future research.     

鄱阳湖周边稻田对白鹤的承载力

鄱阳湖是世界重要的候鸟越冬地,承载了全球约98%的白鹤(Leucogeranus leucogeranus)。然而,近年来鄱阳湖沉水植被退化严重,白鹤的传统食物苦草(Vallisneria spp.)冬芽丰富度急剧减少。食物短缺导致白鹤的觅食生境由自然湿地的浅水生境转移至稻田、藕塘等人工生境,稻田在白鹤保护中发挥的作用日益突出。了解鄱阳湖周边稻田对白鹤的承载力可为白鹤的保护和管理提供科学依据。为此,调查了鄱阳湖周边稻田散落稻谷的生物量,测量了稻谷的营养成分,并利用遥感影像对鄱阳湖周边10 km范围内稻田的总面积进行估算,从而得出鄱阳湖周边稻田散落稻谷能提供的总能量。然后,计算稻田中以稻谷为主要食物的6种鸟类的日能量消耗,用以表示日能量摄入。最后,依据稻田能提供的总能量、鸟类的日能量摄入、鸟类的越冬时长、以及各种鸟类的数量占比,计算得出鄱阳湖周边稻田能承载的白鹤数量。结果表明,鄱阳湖周边稻田散落稻谷的生物量为6.494 g/m²,环鄱阳湖周边10 km范围内稻田的总面积为1984.46 km²,这些稻田能承载的鸟类总数量为140860只,其中能承载的白鹤数量为10775只,超过了全球白鹤的总数量(3500-4000只)。因此,鄱阳湖周边稻田能为白鹤等鸟类提供丰富的食物资源。     

Seeing the Past: Visual Media in Archaeology

As digital media become increasingly affordable and accessible, visual representation in archaeology is expanding across several dimensions. In this essay, I examine some emerging forms of visual media in archaeology, including online documentaries, maps and photographs, hypermedia, experimental films, and peripatetic video. Visual media offer powerful opportunities for engagement with the public. In addition, archaeologists are finding new ways to use the visual in interpretation, analysis, and critique. Experimental visual works often are self-consciously reflexive, questioning and exposing the ways archaeological knowledge is constructed, represented, and disseminated.     

A general rule for sensory cue summation: evidence from photographic, musical, phonetic and cross-modal stimuli

The Euclidean and MAX metrics have been widely used to model cue summation psychophysically and computationally. Both rules happen to be special cases of a more general Minkowski summation rule , where m = 2 and ∞, respectively. In vision research, Minkowski summation with power m = 3-4 has been shown to be a superior model of how subthreshold components sum to give an overall detection threshold. Recently, we have previously reported that Minkowski summation with power m = 2.84 accurately models summation of suprathreshold visual cues in photographs. In four suprathreshold discrimination experiments, we confirm the previous findings with new visual stimuli and extend the applicability of this rule to cue combination in auditory stimuli (musical sequences and phonetic utterances, where m = 2.95 and 2.54, respectively) and cross-modal stimuli (m = 2.56). In all cases, Minkowski summation with power m = 2.5-3 outperforms the Euclidean and MAX operator models. We propose that this reflects the summation of neuronal responses that are not entirely independent but which show some correlation in their magnitudes. Our findings are consistent with electrophysiological research that demonstrates signal correlations (r = 0.1-0.2) between sensory neurons when these are presented with natural stimuli.     

Arabidopsis mutants of AtABCG22, an ABC transporter gene, increase water transpiration and drought susceptibility

In plants, water vapour is released into the atmosphere through stomata in a process called transpiration. Abscisic acid (ABA) is a key phytohormone that facilitates stomatal closure through its action on guard cells. Recently, ATP-binding cassette (ABC) transporter genes, AtABCG25 and AtABCG40, were shown to be involved in ABA transport and responses. However, the functions of many other AtABCG family genes are still unknown. Here, we identified another ABCG gene (AtABCG22) that is required for stomatal regulation in Arabidopsis. The atabcg22 mutant plants had lower leaf temperatures and increased water loss, implying elevated transpiration through an influence on stomatal regulation. We also found that atabcg22 plants were more suspectible to drought stress than wild-type plants. AtABCG22 was expressed in aerial organs, mainly guard cells, in which the gene expression pattern was consistent with the mutant phenotypes. Using double mutants, we investigated the genetic relationships between the mutations. The atabcg22 mutation further increased the water loss of srk2e/ost1 mutants, which were defective in ABA signalling in guard cells. Also, the atabcg22 mutation enhanced the phenotype of nced3 mutants, which were defective in ABA biosynthesis. Accordingly, the additive roles of AtABCG22 functions in ABA signalling and ABA biosynthesis are discussed.     

Research on characteristics of biomass distribution in urban forests of Shanghai metropolis based on remote sensing and spatial analysis

Aims
Monitoring and quantifying the biomass and its distribution in urban trees and forests are crucial to understanding the role of vegetation in an urban environment. In this paper, an estimation method for biomass of urban forests was developed for the Shanghai metropolis, China, based on spatial analysis and a wide variety of data from field inventory and remote sensing.
Methods
An optimal regression model between forest biomass and auxiliary variables was established by stepwise regression analysis. The residual value of regression model was computed for each of the sites sampled and interpolated by Inverse-distance weighting (IDW) to predict residual errors of other sites not subjected to sampling. Forest biomass in the study area was estimated by combining the regression model based on remote sensing image data and residual errors of spatial distribution map. According to the distribution of plantations and management practices, a total of 93 sample plots were established between June 2011 and June 2012 in the Shanghai metropolis. To determine a suitable model, several spectral vegetation indices relating to forest biomass and structure such as normalized difference vegetation index (NDVI), ratio vegetation index (RVI), difference vegetation index (DVI), soil-adjusted vegetation index (SAVI), and modified soil-adjusted vegetation index (MSAVI), and new images synthesized through band combinations such as the sum of TM2, TM3 and TM4 (denoted Band 234), and the sum of TM3, TM4 and TM5 (denoted Band 345) were used as alternative auxiliary parameters .
Important findings
The biomass density in urban forests of the Shanghai metropolis varied from 15 to 120 t·hm^-2. The higher densities of forest biomass concentrated mostly in the urban areas, e.g. in districts of Jing’an and Huangpu, mostly ranging from 35 to 70 t·hm^-2. Suburban localities such as the districts of Jiading and Qingpu had lower biomass densities at around 15 to 50 t·hm^-2. The biomass density of Cinnamomum camphora trees across the Shanghai metropolis varied between 20 and 110 t·hm^-2. The spatial biomass distribution of urban forests displayed a tendency of higher densities in northeastern areas and lower densities in southwestern areas. The total biomass was 3.57 million tons (Tg) for urban forests and 1.33 Tg for C. camphora trees. The overall forest biomass was also found to be distributed mostly in the suburban areas with a fraction of 93.9%, whereas the urban areas shared a fraction of only 6.1%. In terms of the areas, the suburban and urban forests accounted for 95.44% and 4.56%, respectively, of the total areas in the Shanghai metropolis. Among all the administrative districts, the Chongming county and the new district of Pudong had the highest and the second highest biomass, accounting for 20.1% and 19.18% of the total forest biomass, respectively. In contrast, the Jing’an district accounted for only 0.11% of the total forest biomass. The root-mean-square error (RMSE), mean absolute error (MAE) and mean relative error (MRE) of the model for estimating urban forest biomass in this study were 8.39, 6.86 and 24.22%, respectively, decreasing by 57.69%, 55.43% and 64.00% compared to the original simple regression model and by 62.21%, 58.50%, 65.40% compared to the spatial analysis method. Our results indicated that a more efficient way to estimate urban forest biomass in the Shanghai metropolis might be achieved by combining spatial analysis with regression analysis. In fact, the estimated results based on the proposed model are also more comparable to the up-scaled forest inventory data at a city scale than the results obtained using regression analysis or spatial analysis alone.     

Domestic chicks' runway responses to video images of conspecifics

Because chickens are highly social animals, live conspecifics are often used to provide an incentive or goal in studies of gait, sociality or fear that require the bird to traverse a runway. However, the variable behaviour of the stimulus birds can influence the approach/avoidance responses of the test birds and thereby confound the results. Because chickens modify their behaviour readily and appropriately to a variety of video images, including social stimuli, we asked if video playback might represent easily controllable and standardized alternatives to live birds. Female ISA Brown chicks were housed in groups of eight and then exposed to a blank illuminated television for 10 min per day from 2 to 7 days of age. At 8 or 9 days of age they were placed individually in the start box of a 1.6 m long runway and we recorded their responses to a monitor displaying selected video images that was situated in the goal box at the opposite end of the runway. In Experiment 1 chicks approached a monitor playing the video image and soundtrack of feeding chicks significantly sooner than one of a goal box with the food dish and background noise. In Experiment 2, chicks were exposed to the same video of feeding conspecifics with or without the associated sounds or to a video of the goal box with or without the chick soundtrack. Both the videos of other chicks elicited faster approach than did those of the goal box and the sound and silent versions were equally attractive. Adding the soundtrack of feeding chicks to the goal-box video failed to increase its attractiveness. The present results suggest that chicks are attracted towards televised images of other chicks. They also indicate that the visual and auditory components of the video stimuli did not exert additive effects and that approach reflected attraction to the visual image. Collectively, our findings suggest that video playback of selected social stimuli, such as feeding conspecifics, could be a valuable tool in tests requiring voluntary locomotion along a predetermined path.