Improving Known-Unknown Cattle's Face Recognition for Smart Livestock Farm Management

Accurate identification of individual cattle is of paramount importance in precision livestock farming, enabling the monitoring of cattle behavior, disease prevention, and enhanced animal welfare. Unlike human faces, the faces of most Hanwoo cattle, a native breed of Korea, exhibit significant similarities and have the same body color, posing a substantial challenge in accurately distinguishing between individual cattle. In this study, we sought to extend the closed-set scope (only including identifying known individuals) to a more-adaptable open-set recognition scenario (identifying both known and unknown individuals) termed Cattle's Face Open-Set Recognition (CFOSR). By integrating open-set techniques to enhance the closed-set accuracy, the proposed method simultaneously addresses the open-set scenario. In CFOSR, the objective is to develop a trained model capable of accurately identifying known individuals, while effectively handling unknown or novel individuals, even in cases where the model has been trained solely on known individuals. To address this challenge, we propose a novel approach that integrates Adversarial Reciprocal Points Learning (ARPL), a state-of-the-art open-set recognition method, with the effectiveness of Additive Margin Softmax loss (AM-Softmax). ARPL was leveraged to mitigate the overlap between spaces of known and unknown or unregistered cattle. At the same time, AM-Softmax was chosen over the conventional Cross-Entropy loss (CE) to classify known individuals. The empirical results obtained from a real-world dataset demonstrated the effectiveness of the ARPL and AM-Softmax techniques in achieving both intra-class compactness and inter-class separability. Notably, the results of the open-set recognition and closed-set recognition validated the superior performance of our proposed method compared to existing algorithms. To be more precise, our method achieved an AUROC of 91.84 and an OSCR of 87.85 in the context of open-set recognition on a complex dataset. Simultaneously, it demonstrated an accuracy of 94.46 for closed-set recognition. We believe that our study provides a novel vision to improve the classification accuracy of the closed set. Simultaneously, it holds the potential to significantly contribute to herd monitoring and inventory management, especially in scenarios involving the presence of unknown or novel cattle.

An iterative noisy annotation correction model for robust plant disease detection

Previous work on plant disease detection demonstrated that object detectors generally suffer from degraded training data, and annotations with noise may cause the training task to fail. Well-annotated datasets are therefore crucial to build a robust detector. However, a good label set generally requires much expert knowledge and meticulous work, which is expensive and time-consuming. This paper aims to learn robust feature representations with inaccurate bounding boxes, thereby reducing the model requirements for annotation quality. Specifically, we analyze the distribution of noisy annotations in the real world. A teacher-student learning paradigm is proposed to correct inaccurate bounding boxes. The teacher model is used to rectify the degraded bounding boxes, and the student model extracts more robust feature representations from the corrected bounding boxes. Furthermore, the method can be easily generalized to semi-supervised learning paradigms and auto-labeling techniques. Experimental results show that applying our method to the Faster-RCNN detector achieves a 26% performance improvement on the noisy dataset. Besides, our method achieves approximately 75% of the performance of a fully supervised object detector when 1% of the labels are available. Overall, this work provides a robust solution to real-world location noise. It alleviates the challenges posed by noisy data to precision agriculture, optimizes data labeling technology, and encourages practitioners to further investigate plant disease detection and intelligent agriculture at a lower cost. The code will be released at https://github.com/JiuqingDong/TS_OAMIL-for-Plant-disease-detection.

A New Deep Learning-based Dynamic Paradigm Towards Open-World Plant Disease Detection

Plant disease detection has made significant strides thanks to the emergence of deep learning. However, existing methods have been limited to closed-set and static learning settings, where models are trained using a specific dataset. This confinement restricts the model's adaptability when encountering samples from unseen disease categories. Additionally, there is a challenge of knowledge degradation for these static learning settings, as the acquisition of new knowledge tends to overwrite the old when learning new categories. To overcome these limitations, this study introduces a novel paradigm for plant disease detection called open-world setting. Our approach can infer disease categories that have never been seen during the model training phase and gradually learn these unseen diseases through dynamic knowledge updates in the next training phase. Specifically, we utilize a well-trained unknown-aware region proposal network to generate pseudo-labels for unknown diseases during training and employ a class-agnostic classifier to enhance the recall rate for unknown diseases. Besides, we employ a sample replay strategy to maintain recognition ability for previously learned classes. Extensive experimental evaluation and ablation studies investigate the efficacy of our method in detecting old and unknown classes. Remarkably, our method demonstrates robust generalization ability even in cross-species disease detection experiments. Overall, this open-world and dynamically updated detection method shows promising potential to become the future paradigm for plant disease detection. We discuss open issues including classification and localization, and propose promising approaches to address them. We encourage further research in the community to tackle the crucial challenges in open-world plant disease detection. The code will be released at https://github.com/JiuqingDong/OWPDD.

Embracing limited and imperfect training datasets: opportunities and challenges in plant disease recognition using deep learning

Recent advancements in deep learning have brought significant improvements to plant disease recognition. However, achieving satisfactory performance often requires high-quality training datasets, which are challenging and expensive to collect. Consequently, the practical application of current deep learning-based methods in real-world scenarios is hindered by the scarcity of high-quality datasets. In this paper, we argue that embracing poor datasets is viable and aims to explicitly define the challenges associated with using these datasets. To delve into this topic, we analyze the characteristics of high-quality datasets, namely, large-scale images and desired annotation, and contrast them with the limited and imperfect nature of poor datasets. Challenges arise when the training datasets deviate from these characteristics. To provide a comprehensive understanding, we propose a novel and informative taxonomy that categorizes these challenges. Furthermore, we offer a brief overview of existing studies and approaches that address these challenges. We point out that our paper sheds light on the importance of embracing poor datasets, enhances the understanding of the associated challenges, and contributes to the ambitious objective of deploying deep learning in real-world applications. To facilitate the progress, we finally describe several outstanding questions and point out potential future directions. Although our primary focus is on plant disease recognition, we emphasize that the principles of embracing and analyzing poor datasets are applicable to a wider range of domains, including agriculture. Our project is public available at https://github.com/xml94/EmbracingLimitedImperfectTrainingDatasets.

Local refinement mechanism for improved plant leaf segmentation in cluttered backgrounds

Plant phenotyping is a critical field in agriculture, aiming to understand crop growth under specific conditions. Recent research uses images to describe plant characteristics by detecting visual information within organs such as leaves, flowers, stems, and fruits. However, processing data in real field conditions, with challenges such as image blurring and occlusion, requires improvement. This paper proposes a deep learning-based approach for leaf instance segmentation with a local refinement mechanism to enhance performance in cluttered backgrounds. The refinement mechanism employs Gaussian low-pass and High-boost filters to enhance target instances and can be applied to the training or testing dataset. An instance segmentation architecture generates segmented masks and detected areas, facilitating the derivation of phenotypic information, such as leaf count and size. Experimental results on a tomato leaf dataset demonstrate the system's accuracy in segmenting target leaves despite complex backgrounds. The investigation of the refinement mechanism with different kernel sizes reveals that larger kernel sizes benefit the system's ability to generate more leaf instances when using a High-boost filter, while prediction performance decays with larger Gaussian low-pass filter kernel sizes. This research addresses challenges in real greenhouse scenarios and enables automatic recognition of phenotypic data for smart agriculture. The proposed approach has the potential to enhance agricultural practices, ultimately leading to improved crop yields and productivity.

Multiview Monitoring of Individual Cattle Behavior Based on Action Recognition in Closed Barns Using Deep Learning

Cattle behavior recognition is essential for monitoring their health and welfare. Existing techniques for behavior recognition in closed barns typically rely on direct observation to detect changes using wearable devices or surveillance cameras. While promising progress has been made in this field, monitoring individual cattle, especially those with similar visual characteristics, remains challenging due to numerous factors such as occlusion, scale variations, and pose changes. Accurate and consistent individual identification over time is therefore essential to overcome these challenges. To address this issue, this paper introduces an approach for multiview monitoring of individual cattle behavior based on action recognition using video data. The proposed system takes an image sequence as input and utilizes a detector to identify hierarchical actions categorized as part and individual actions. These regions of interest are then inputted into a tracking and identification mechanism, enabling the system to continuously track each individual in the scene and assign them a unique identification number. By implementing this approach, cattle behavior is continuously monitored, and statistical analysis is conducted to assess changes in behavior in the time domain. The effectiveness of the proposed framework is demonstrated through quantitative and qualitative experimental results obtained from our Hanwoo cattle video database. Overall, this study tackles the challenges encountered in real farm indoor scenarios, capturing spatiotemporal information and enabling automatic recognition of cattle behavior for precision livestock farming.

Isolation of Streptomycin-Resistant Erwinia pyrifoliae in Korea

As a black shoot blight disease-causing agent, Erwinia pyrifoliae was first reported in 1995 in Korea. A total of 101 isolates of E. pyrifoliae were isolated from samples showing bacterial symptoms collected from apple and pear orchards between 2020 and 2021. These isolates were screened for streptomycin resistance, with one from an orchard in Gwangju showing resistance at 100 μg/ml streptomycin. This streptomycin-resistant E. pyrifoliae (EpSmR) isolate was identified via polymerase chain reaction amplification of the strA/strB gene and an internal region of the ribosomal rpsL gene containing codon 43. EpSmR has a point mutation that altered this codon from lysine (AAA) to threonine (ACA). The strA and strB genes were not identified in EpSmR. EpSmR showed a high resistance to streptomycin (>50,000 μg/ml). This is the first study reporting EpSmR, which emerged due to a mutation in codon 43 of the rpsL gene.

Transfer learning for versatile plant disease recognition with limited data

Deep learning has witnessed a significant improvement in recent years to recognize plant diseases by observing their corresponding images. To have a decent performance, current deep learning models tend to require a large-scale dataset. However, collecting a dataset is expensive and time-consuming. Hence, the limited data is one of the main challenges to getting the desired recognition accuracy. Although transfer learning is heavily discussed and verified as an effective and efficient method to mitigate the challenge, most proposed methods focus on one or two specific datasets. In this paper, we propose a novel transfer learning strategy to have a high performance for versatile plant disease recognition, on multiple plant disease datasets. Our transfer learning strategy differs from the current popular one due to the following factors. First, PlantCLEF2022, a large-scale dataset related to plants with 2,885,052 images and 80,000 classes, is utilized to pre-train a model. Second, we adopt a vision transformer (ViT) model, instead of a convolution neural network. Third, the ViT model undergoes transfer learning twice to save computations. Fourth, the model is first pre-trained in ImageNet with a self-supervised loss function and with a supervised loss function in PlantCLEF2022. We apply our method to 12 plant disease datasets and the experimental results suggest that our method surpasses the popular one by a clear margin for different dataset settings. Specifically, our proposed method achieves a mean testing accuracy of 86.29over the 12 datasets in a 20-shot case, 12.76 higher than the current state-of-the-art method's accuracy of 73.53. Furthermore, our method outperforms other methods in one plant growth stage prediction and the one weed recognition dataset. To encourage the community and related applications, we have made public our codes and pre-trained model.

Flexible and high quality plant growth prediction with limited data

Predicting plant growth is a fundamental challenge that can be employed to analyze plants and further make decisions to have healthy plants with high yields. Deep learning has recently been showing its potential to address this challenge in recent years, however, there are still two issues. First, image-based plant growth prediction is currently taken either from time series or image generation viewpoints, resulting in a flexible learning framework and clear predictions, respectively. Second, deep learning-based algorithms are notorious to require a large-scale dataset to obtain a competing performance but collecting enough data is time-consuming and expensive. To address the issues, we consider the plant growth prediction from both viewpoints with two new time-series data augmentation algorithms. To be more specific, we raise a new framework with a length-changeable time-series processing unit to generate images flexibly. A generative adversarial loss is utilized to optimize our model to obtain high-quality images. Furthermore, we first recognize three key points to perform time-series data augmentation and then put forward T-Mixup and T-Copy-Paste. T-Mixup fuses images from a different time pixel-wise while T-Copy-Paste makes new time-series images with a different background by reusing individual leaves extracted from the existing dataset. We perform our method in a public dataset and achieve superior results, such as the generated RGB images and instance masks securing an average PSNR of 27.53 and 27.62, respectively, compared to the previously best 26.55 and 26.92.

Data-centric annotation analysis for plant disease detection: Strategy, consistency, and performance

Object detection models have become the current tool of choice for plant disease detection in precision agriculture. Most existing research improved the performance by ameliorating networks and optimizing the loss function. However, because of the vast influence of data annotation quality and the cost of annotation, the data-centric part of a project also needs more investigation. We should further consider the relationship between data annotation strategies, annotation quality, and the model's performance. In this paper, a systematic strategy with four annotation strategies for plant disease detection is proposed: local, semi-global, global, and symptom-adaptive annotation. Labels with different annotation strategies will result in distinct models' performance, and their contrasts are remarkable. An interpretability study of the annotation strategy is conducted by using class activation maps. In addition, we define five types of inconsistencies in the annotation process and investigate the severity of the impact of inconsistent labels on model's performance. Finally, we discuss the problem of label inconsistency during data augmentation. Overall, this data-centric quantitative analysis helps us to understand the significance of annotation strategies, which provides practitioners a way to obtain higher performance and reduce annotation costs on plant disease detection. Our work encourages researchers to pay more attention to annotation consistency and the essential issues of annotation strategy. The code will be released at: https://github.com/JiuqingDong/PlantDiseaseDetection_Yolov5 .

Improving Accuracy of Tomato Plant Disease Diagnosis Based on Deep Learning With Explicit Control of Hidden Classes

Recognizing plant diseases is a major challenge in agriculture, and recent works based on deep learning have shown high efficiency in addressing problems directly related to this area. Nonetheless, weak performance has been observed when a model trained on a particular dataset is evaluated in new greenhouse environments. Therefore, in this work, we take a step towards these issues and present a strategy to improve model accuracy by applying techniques that can help refine the model's generalization capability to deal with complex changes in new greenhouse environments. We propose a paradigm called "control to target classes." The core of our approach is to train and validate a deep learning-based detector using target and control classes on images collected in various greenhouses. Then, we apply the generated features for testing the inference of the system on data from new greenhouse conditions where the goal is to detect target classes exclusively. Therefore, by having explicit control over inter- and intra-class variations, our model can distinguish data variations that make the system more robust when applied to new scenarios. Experiments demonstrate the effectiveness and efficiency of the proposed approach on our extended tomato plant diseases dataset with 14 classes, from which 5 are target classes and the rest are control classes. Our detector achieves a recognition rate of target classes of 93.37% mean average precision on the inference dataset. Finally, we believe that our study offers valuable guidelines for researchers working in plant disease recognition with complex input data.

Genome-wide analysis in 756,646 individuals provides first genetic evidence that ACE2 expression influences COVID-19 risk and yields genetic risk scores predictive of severe disease

SARS-CoV-2 enters host cells by binding angiotensin-converting enzyme 2 (ACE2). Through a genome-wide association study, we show that a rare variant (MAF = 0.3%, odds ratio 0.60, P=4.5×10^-13) that down-regulates ACE2 expression reduces risk of COVID-19 disease, providing human genetics support for the hypothesis that ACE2 levels influence COVID-19 risk. Further, we show that common genetic variants define a risk score that predicts severe disease among COVID-19 cases.

Open Set Self and Across Domain Adaptation for Tomato Disease Recognition With Deep Learning Techniques

Recent advances in automatic recognition systems based on deep learning technology have shown the potential to provide environmental-friendly plant disease monitoring. These systems are able to reliably distinguish plant anomalies under varying environmental conditions as the basis for plant intervention using methods such as classification or detection. However, they often show a performance decay when applied under new field conditions and unseen data. Therefore, in this article, we propose an approach based on the concept of open-set domain adaptation to the task of plant disease recognition to allow existing systems to operate in new environments with unseen conditions and farms. Our system specifically copes diagnosis as an open set learning problem, and mainly operates in the target domain by exploiting a precise estimation of unknown data while maintaining the performance of the known classes. The main framework consists of two modules based on deep learning that perform bounding box detection and open set self and across domain adaptation. The detector is built based on our previous filter bank architecture for plant diseases recognition and enforces domain adaptation from the source to the target domain, by constraining data to be classified as one of the target classes or labeled as unknown otherwise. We perform an extensive evaluation on our tomato plant diseases dataset with three different domain farms, which indicates that our approach can efficiently cope with changes of new field environments during field-testing and observe consistent gains from explicit modeling of unseen data.

Style-Consistent Image Translation: A Novel Data Augmentation Paradigm to Improve Plant Disease Recognition

Deep learning shows its advantages and potentials in plant disease recognition and has witnessed a profound development in recent years. To obtain a competing performance with a deep learning algorithm, enough amount of annotated data is requested but in the natural world, scarce or imbalanced data are common, and annotated data is expensive or hard to collect. Data augmentation, aiming to create variations for training data, has shown its power for this issue. But there are still two challenges: creating more desirable variations for scarce and imbalanced data, and designing a data augmentation to ease object detection and instance segmentation. First, current algorithms made variations only inside one specific class, but more desirable variations can further promote performance. To address this issue, we propose a novel data augmentation paradigm that can adapt variations from one class to another. In the novel paradigm, an image in the source domain is translated into the target domain, while the variations unrelated to the domain are maintained. For example, an image with a healthy tomato leaf is translated into a powdery mildew image but the variations of the healthy leaf are maintained and transferred into the powdery mildew class, such as types of tomato leaf, sizes, and viewpoints. Second, current data augmentation is suitable to promote the image classification model but may not be appropriate to alleviate object detection and instance segmentation model, mainly because the necessary annotations can not be obtained. In this study, we leverage a prior mask as input to tell the area we are interested in and reuse the original annotations. In this way, our proposed algorithm can be utilized to do the three tasks simultaneously. Further, We collect 1,258 images of tomato leaves with 1,429 instance segmentation annotations as there is more than one instance in one single image, including five diseases and healthy leaves. Extensive experimental results on the collected images validate that our new data augmentation algorithm makes useful variations and contributes to improving performance for diverse deep learning-based methods.

Human Placenta Hydrolysate Promotes Liver Regeneration via Activation of the Cytokine/Growth Factor-Mediated Pathway and Anti-oxidative Effect

Liver regeneration is a very complex process and is regulated by several cytokines and growth factors. It is also known that liver transplantation and the regeneration process cause massive oxidative stress, which interferes with liver regeneration. The placenta is known to contain various physiologically active ingredients such as cytokines, growth factors, and amino acids. In particular, human placenta hydrolysate (hPH) has been found to contain many amino acids. Most of the growth factors found in the placenta are known to be closely related to liver regeneration. Therefore, in this study, we investigated whether hPH is effective in promoting liver regeneration in rats undergoing partial hepatectomy. We confirmed that cell proliferation was significantly increased in HepG2 and human primary cells. Hepatocyte proliferation was also promoted in partial hepatectomized rats by hPH treatment. hPH increased liver regeneration rate, double nucleic cell ratio, mitotic cell ratio, proliferating cell nuclear antigen (PCNA), and K_i-67 positive cells in vivo as well as interleukin (IL)-6, tumor necrosis factor alpha (TNF-α), and hepatocyte growth factor (HGF). Moreover, Kupffer cells secreting IL-6 and TNF-α were activated by hPH treatment. In addition, hPH reduced thiobarbituric acid reactive substances (TBARs) and significantly increased glutathione (GSH), glutathione peroxidase (GPx), and superoxide dismutase (SOD). Taken together, these results suggest that hPH promotes liver regeneration by activating cytokines and growth factors associated with liver regeneration and eliminating oxidative stress.

Improved Real-Time Facial Expression Recognition Based on a Novel Balanced and Symmetric Local Gradient Coding

In the field of Facial Expression Recognition (FER), traditional local texture coding methods have a low computational complexity, while providing a robust solution with respect to occlusion, illumination, and other factors. However, there is still need for improving the accuracy of these methods while maintaining their real-time nature and low computational complexity. In this paper, we propose a feature-based FER system with a novel local texture coding operator, named central symmetric local gradient coding (CS-LGC), to enhance the performance of real-time systems. It uses four different directional gradients on 5 × 5 grids, and the gradient is computed in the center-symmetric way. The averages of the gradients are used to reduce the sensitivity to noise. These characteristics lead to symmetric of features by the CS-LGC operator, thus providing a better generalization capability in comparison to existing local gradient coding (LGC) variants. The proposed system further transforms the extracted features into an eigen-space using a principal component analysis (PCA) for better representation and less computation; it estimates the intended classes by training an extreme learning machine. The recognition rate for the JAFFE database is 95.24%, whereas that for the CK+ database is 98.33%. The results show that the system has advantages over the existing local texture coding methods.

Deep Learning-Based Phenotyping System With Glocal Description of Plant Anomalies and Symptoms

Recent advances in Deep Neural Networks have allowed the development of efficient and automated diagnosis systems for plant anomalies recognition. Although existing methods have shown promising results, they present several limitations to provide an appropriate characterization of the problem, especially in real-field scenarios. To address this limitation, we propose an approach that besides being able to efficiently detect and localize plant anomalies, allows to generate more detailed information about their symptoms and interactions with the scene, by combining visual object recognition and language generation. It uses an image as input and generates a diagnosis result that shows the location of anomalies and sentences describing the symptoms as output. Our framework is divided into two main parts: First, a detector obtains a set of region features that contain the anomalies using a Region-based Deep Neural Network. Second, a language generator takes the features of the detector as input and generates descriptive sentences with details of the symptoms using Long-Short Term Memory (LSTM). Our loss metric allows the system to be trained end-to-end from the object detector to the language generator. Finally, the system outputs a set of bounding boxes along with the sentences that describe their symptoms using glocal criteria into two different ways: a set of specific descriptions of the anomalies detected in the plant and an abstract description that provides general information about the scene. We demonstrate the efficiency of our approach in the challenging tomato diseases and pests recognition task. We further show that our approach achieves a mean Average Precision (mAP) of 92.5% in our newly created Tomato Plant Anomalies Description Dataset. Our objective evaluation allows users to understand the relationships between pathologies and their evolution throughout their stage of infection, location in the plant, symptoms, etc. Our work introduces a cost-efficient tool that provides farmers with a technology that facilitates proper handling of crops.

High-Performance Deep Neural Network-Based Tomato Plant Diseases and Pests Diagnosis System With Refinement Filter Bank

A fundamental problem that confronts deep neural networks is the requirement of a large amount of data for a system to be efficient in complex applications. Promising results of this problem are made possible through the use of techniques such as data augmentation or transfer learning of pre-trained models in large datasets. But the problem still persists when the application provides limited or unbalanced data. In addition, the number of false positives resulting from training a deep model significantly cause a negative impact on the performance of the system. This study aims to address the problem of false positives and class unbalance by implementing a Refinement Filter Bank framework for Tomato Plant Diseases and Pests Recognition. The system consists of three main units: First, a Primary Diagnosis Unit (Bounding Box Generator) generates the bounding boxes that contain the location of the infected area and class. The promising boxes belonging to each class are then used as input to a Secondary Diagnosis Unit (CNN Filter Bank) for verification. In this second unit, misclassified samples are filtered through the training of independent CNN classifiers for each class. The result of the CNN Filter Bank is a decision of whether a target belongs to the category as it was detected (True) or not (False) otherwise. Finally, an integration unit combines the information from the primary and secondary units while keeping the True Positive samples and eliminating the False Positives that were misclassified in the first unit. By this implementation, the proposed approach is able to obtain a recognition rate of approximately 96%, which represents an improvement of 13% compared to our previous work in the complex task of tomato diseases and pest recognition. Furthermore, our system is able to deal with the false positives generated by the bounding box generator, and class unbalances that appear especially on datasets with limited data.

Simple Coiling versus Stent-Assisted Coiling of Paraclinoid Aneurysms: Radiological Outcome in a Single Center Study

Objective

Paraclinoid aneurysms are a group of aneurysms arising at the distal internal carotid artery. Due to a high incidence of small, wide-necked aneurysms in this zone, it is often challenging to achieve complete occlusion when solely using detachable coils, thus stent placement is often required. In the present study, we aimed to investigate the effect of stent placement in endovascular treatment of paraclinoid aneurysms.

Methods

Data of 98 paraclinoid aneurysms treated by endovascular approach in our center from August 2005 to June 2016 were retrospectively reviewed. They were divided into two groups: simple coiling and stent-assisted coiling. Differences in the recurrence and progressive occlusion between the two groups were mainly analyzed. The recurrence was defined as more than one grade worsening according to Raymond-Roy Classification or major recanalization that is large enough to permit retreatment in the follow-up study compared to the immediate post-operative results.

Results

Complete occlusion was achieved immediately after endovascular treatment in eight out of 37 patients (21.6%) in the stent-assisted group and 18 out of 61 (29.5%) in the simple coiling group. In the follow-up imaging studies, the recurrence rate was lower in the stent-assisted group (one out of 37, 2.7%) compared to the simple coiling group (13 out of 61, 21.3%) (p=0.011). Multivariate logistic regression model showed lower recurrence rate in the stent-assisted group than the simple coiling group (odds ratio [OR] 0.051, 95% confidence interval [CI] 0.005-0.527). Furthermore there was also a significant difference in the rate of progressive occlusion between the stent-assisted group (16 out of 29 patients, 55.2%) and the simple coiling group (10 out of 43 patients, 23.3%) (p=0.006). The stent-assisted group also exhibited a higher rate of progressive occlusion than the simple coiling group in the multivariate logistic regression model (OR 3.208, 95% CI 1.106-9.302).

Conclusion

Use of stents results in good prognosis not only by reducing the recurrence rate but also by increasing the rate of progressive occlusion in wide-necked paraclinoid aneurysms. Stent-assisted coil embolization can be an important treatment strategy for paraclinoid aneurysms when considering the superiority of long term outcome.

A Robust Deep-Learning-Based Detector for Real-Time Tomato Plant Diseases and Pests Recognition

Plant Diseases and Pests are a major challenge in the agriculture sector. An accurate and a faster detection of diseases and pests in plants could help to develop an early treatment technique while substantially reducing economic losses. Recent developments in Deep Neural Networks have allowed researchers to drastically improve the accuracy of object detection and recognition systems. In this paper, we present a deep-learning-based approach to detect diseases and pests in tomato plants using images captured in-place by camera devices with various resolutions. Our goal is to find the more suitable deep-learning architecture for our task. Therefore, we consider three main families of detectors: Faster Region-based Convolutional Neural Network (Faster R-CNN), Region-based Fully Convolutional Network (R-FCN), and Single Shot Multibox Detector (SSD), which for the purpose of this work are called "deep learning meta-architectures". We combine each of these meta-architectures with "deep feature extractors" such as VGG net and Residual Network (ResNet). We demonstrate the performance of deep meta-architectures and feature extractors, and additionally propose a method for local and global class annotation and data augmentation to increase the accuracy and reduce the number of false positives during training. We train and test our systems end-to-end on our large Tomato Diseases and Pests Dataset, which contains challenging images with diseases and pests, including several inter- and extra-class variations, such as infection status and location in the plant. Experimental results show that our proposed system can effectively recognize nine different types of diseases and pests, with the ability to deal with complex scenarios from a plant's surrounding area.

Flavivirus and Filovirus EvoPrinters: New alignment tools for the comparative analysis of viral evolution

Background

Flavivirus and Filovirus infections are serious epidemic threats to human populations. Multi-genome comparative analysis of these evolving pathogens affords a view of their essential, conserved sequence elements as well as progressive evolutionary changes. While phylogenetic analysis has yielded important insights, the growing number of available genomic sequences makes comparisons between hundreds of viral strains challenging. We report here a new approach for the comparative analysis of these hemorrhagic fever viruses that can superimpose an unlimited number of one-on-one alignments to identify important features within genomes of interest.

Methodology/Principal finding

We have adapted EvoPrinter alignment algorithms for the rapid comparative analysis of Flavivirus or Filovirus sequences including Zika and Ebola strains. The user can input a full genome or partial viral sequence and then view either individual comparisons or generate color-coded readouts that superimpose hundreds of one-on-one alignments to identify unique or shared identity SNPs that reveal ancestral relationships between strains. The user can also opt to select a database genome in order to access a library of pre-aligned genomes of either 1,094 Flaviviruses or 460 Filoviruses for rapid comparative analysis with all database entries or a select subset. Using EvoPrinter search and alignment programs, we show the following: 1) superimposing alignment data from many related strains identifies lineage identity SNPs, which enable the assessment of sublineage complexity within viral outbreaks; 2) whole-genome SNP profile screens uncover novel Dengue2 and Zika recombinant strains and their parental lineages; 3) differential SNP profiling identifies host cell A-to-I hyper-editing within Ebola and Marburg viruses, and 4) hundreds of superimposed one-on-one Ebola genome alignments highlight ultra-conserved regulatory sequences, invariant amino acid codons and evolutionarily variable protein-encoding domains within a single genome.

Conclusions/Significance

EvoPrinter allows for the assessment of lineage complexity within Flavivirus or Filovirus outbreaks, identification of recombinant strains, highlights sequences that have undergone host cell A-to-I editing, and identifies unique input and database SNPs within highly conserved sequences. EvoPrinter’s ability to superimpose alignment data from hundreds of strains onto a single genome has allowed us to identify unique Zika virus sublineages that are currently spreading in South, Central and North America, the Caribbean, and in China. This new set of integrated alignment programs should serve as a useful addition to existing tools for the comparative analysis of these viruses.

Flaviviruses, including Zika and Dengue viruses, and Filoviruses, including Ebola and Marburg viruses, are significant global public health threats. Genetic surveillance of viral isolates provides important insights into the origin of outbreaks, reveals lineage heterogeneity and diversification, and facilitates identification of novel recombinant strains and host cell modified viral genomes. We report the development of EvoPrinter, a web-accessed alignment tool for the rapid comparative analysis of viral genomes. EvoPrinter superimposes alignment data from multiple pairwise comparisons onto a single reference sequence of interest, to reveal both similarities and differences detected in hundreds of selected viral isolates. Evoprinter databases provide easy access to hundreds of non-redundant Flavivirus and Filovirus genomes. allowing the user to distinguish between sublineage identity SNPs and unique strain-specific SNPs, thus facilitating analysis of the history of viral diversification during an epidemic. EvoPrinter also proves useful in identifying recombinant strains and their parental lineages and detecting host-cell genomic editing. EvoPrinter should serve as a useful addition to existing tools for the comparative analysis of these viruses.

Relationship between Gyrus Rectus Resection and Cognitive Impairment after Surgery for Ruptured Anterior Communicating Artery Aneurysms

OBJECTIVE

The gyrus rectus (GR) is known as a non-functional gyrus; hence, its resection is agreed to be a safe procedure frequently practiced to achieve a better surgical view during specific surgeries. This study aimed at comparing the cognitive outcomes following GR resection in patients who underwent surgery for ruptured anterior communicating artery (ACoA) aneurysms.

MATERIALS AND METHODS

From 2012 to 2015, 39 patients underwent surgical clipping for ruptured ACoA aneurysms. Mini-mental state examinations (MMSE) were performed in 2 different periods. The statistical relationship between GR resection and MMSE results was evaluated, and further analysis of MMSE subgroup was performed.

RESULTS

Twenty-five out of the 39 patients (64.19%) underwent GR resection. Mean initial and final MMSE scores in the GR resection group were 16.3 ± 9.8 and 20.8 ± 7.3, respectively. In the non-resection group, the mean initial and final MMSE scores were 17.1 ± 8.6 and 21.9 ± 4.5, respectively. Neither group's scores showed a significant change. Subgroup analysis of initial MMSE showed a significant difference in memory recall and language (p = 0.02) but not in the final MMSE scores.

CONCLUSION

There was no significant relationship between the GR resection and cognitive outcomes in terms of total MMSE scores after surgery for ruptured ACoA aneurysm. However, subgroup analysis revealed a temporary negative effect of GR resection in the categories of language and memory recall. This study suggests that GR resection should be executed superficially, owing to its close anatomical relationship with the limbic system.

Serial Mini-Mental Status Examination to Evaluate Cognitive Outcome in Patients with Traumatic Brain Injury

OBJECTIVE

This study was aimed at finding out the changes in cognitive dysfunction in patients with traumatic brain injury (TBI) and investigating the factors limiting their cognitive improvement.

METHODS

Between January 2010 and March 2014, 33 patients with TBI participated in serial mini-mental status examination (MMSE). Their cognitive functions were statistically analyzed to clarify their relationship with different TBI status. Patients who developed hydrocephalus were separately analyzed in regards to their cognitive function depending on the placement of ventriculoperitoneal shunt (VPS).

RESULTS

Bi-frontal lobe injury (β=-10.441, p<0.001), contre-coup injury (β=-6.592, p=0.007), severe parenchymal injury (β=-7.210, p=0.012), temporal lobe injury (β=-5.524, p=0.027), and dominant hemisphere injury (β=-5.388, p=0.037) significantly lowered the final MMSE scores. The risk of down-grade in the prognosis was higher in severe parenchymal injury [odds ratio (OR)=13.41, 95% confidence interval (CI)=1.31-136.78], temporal lobe injury (OR=12.3, 95% CI=2.07-73.08), dominant hemisphere injury (OR=8.19, 95% CI=1.43-46.78), and bi-frontal lobe injury (OR=7.52, 95% CI=1.31-43.11). In the 11 post-traumatic hydrocephalus patients who underwent VPS, the final MMSE scores (17.7±6.8) substantially increased from the initial MMSE scores (11.2±8.6).

CONCLUSION

Presence of bi-frontal lobe injury, temporal lobe injury, dominant hemisphere injury, and contre-coup injury and severe parenchymal injury adversely influenced the final MMSE scores. They can be concluded to be poor prognostic factors in terms of cognitive function in TBI patients. Development of hydrocephalus aggravates cognitive impairment with unpredictable time of onset. Thus, close observation and routine image follow-up are mandatory for early detection and surgical intervention for hydrocephalus.

Clinical Utility of an Automated Pupillometer in Patients with Acute Brain Lesion

Objective

The purpose of this study was to evaluate the clinical utility and validity of using a pupillometer to assess patients with acute brain lesions.

Methods

Pupillary examinations using an automated pupillometer (NeurOptics®NPi™-100 Pupillometer) were performed every 4 hours and were simultaneously assessed using the Glasgow Coma Scale (GCS) and for intracranial pressure (ICP), from admission to discharge or expire in neuro-intensive care unit (NICU). Manual pupillary examinations were also recorded for comparison. By comparing these data, we evaluated the validity of using automated pupillometers to predict clinical outcomes.

Results

The mean values of the Neurologic Pupillary index (NPi) were different in the groups examined manually. The GCS correlated well with NPi values, especially in severe brain injury patients (GCS below 9). However, the NPi values were weakly correlated with intracranial pressure (ICP) when the ICP was lower than 30 cm H₂O. The NPi value was not affected by age or intensity of illumination. In patients with a "poor" prognosis who had a Glasgow Outcome Scale (GOS) of 1 or 2, the mean initial NPi score was 0.88±1.68, whereas the value was 3.89±0.97 in patients with a "favorable" prognosis who had a GOS greater than 2 (p<0.001). For predicting clinical outcomes, the initial NPi value of 3.4 had the highest sensitivity and specificity.

Conclusion

An automated pupillometer can serve as a simple and useful tool for the accurate measurement of pupillary reactivity in patients with acute brain lesions.

Intensity Variation Normalization for Finger Vein Recognition Using Guided Filter Based Singe Scale Retinex

Finger vein recognition has been considered one of the most promising biometrics for personal authentication. However, the capacities and percentages of finger tissues (e.g., bone, muscle, ligament, water, fat, etc.) vary person by person. This usually causes poor quality of finger vein images, therefore degrading the performance of finger vein recognition systems (FVRSs). In this paper, the intrinsic factors of finger tissue causing poor quality of finger vein images are analyzed, and an intensity variation (IV) normalization method using guided filter based single scale retinex (GFSSR) is proposed for finger vein image enhancement. The experimental results on two public datasets demonstrate the effectiveness of the proposed method in enhancing the image quality and finger vein recognition accuracy.

Clinical characteristics of cerebral venous thrombosis in a single center in Korea

OBJECTIVE

The purpose of this study is to investigate the clinical characteristics of cerebral venous thrombosis (CVT) in a single center in Korea.

METHODS

A total of 36 patients were diagnosed with CVT from August 2005 to May 2013. The patient data regarding age, sex, disease stage, pathogenesis, location, laboratory findings, radiological findings, and treatment modalities were retrospectively collected. The results were compared with those of previous studies in other countries.

RESULTS

The patient group comprised 21 men and 15 women with a mean age of 46.9 years (ranging from three months to 77 years). The most common cause was a prothrombotic condition (8 patients, 22.2%). Within the patient group, 13 patients (36.1%) had a hemorrhagic infarction, whereas 23 (63.9%) had a venous infarction without hemorrhage. By location, the incidence of hemorrhagic infarction was the highest in the group with a transverse and/or sigmoid sinus thrombosis (n=9); however, the proportion of hemorrhagic infarction was higher in the cortical venous thrombosis group (75%) and the deep venous thrombosis group (100%). By pathogenesis, the incidence of hemorrhagic infarction was the highest in the prothrombotic group (n=6), which was statistically significant (p=0.016).

CONCLUSION

According to this study, CVT was more prevalent in men, and the peak age group comprised patients in the sixth decade. The most common cause was a prothrombotic condition. This finding was comparable with reports from Europe or America, in which CVT was more common in younger women. Hemorrhagic infarction was more common in the prothrombotic group (p=0.016) than in the non-prothrombotic group in this study.

Efficient descriptor of histogram of salient edge orientation map for finger vein recognition

Finger vein images are rich in orientation and edge features. Inspired by the edge histogram descriptor proposed in MPEG-7, this paper presents an efficient orientation-based local descriptor, named histogram of salient edge orientation map (HSEOM). HSEOM is based on the fact that human vision is sensitive to edge features for image perception. For a given image, HSEOM first finds oriented edge maps according to predefined orientations using a well-known edge operator and obtains a salient edge orientation map by choosing an orientation with the maximum edge magnitude for each pixel. Then, subhistograms of the salient edge orientation map are generated from the nonoverlapping submaps and concatenated to build the final HSEOM. In the experiment of this paper, eight oriented edge maps were used to generate a salient edge orientation map for HSEOM construction. Experimental results on our available finger vein image database, MMCBNU_6000, show that the performance of HSEOM outperforms that of state-of-the-art orientation-based methods (e.g., Gabor filter, histogram of oriented gradients, and local directional code). Furthermore, the proposed HSEOM has advantages of low feature dimensionality and fast implementation for a real-time finger vein recognition system.

Glomus tumor in the femoral nerve

The glomus tumor of the peripheral nerve is one of the mesenchymal tumors originating in the epineurium, and is extremely rare. A 56-year-old man presented complaining of lancinating pain on the left thigh, which was provoked by pressure or exercise. Subsequent image study revealed a mass in the femoral nerve. Total surgical excision with the aid of intraoperative ultrasonography was performed and the pain was successfully controlled. The authors report an unusual case of a patient diagnosed with glomus tumor in peripheral nerve, with a review of the clinical features, imaging, and pathological findings.

Robust finger vein ROI localization based on flexible segmentation

Finger veins have been proved to be an effective biometric for personal identification in the recent years. However, finger vein images are easily affected by influences such as image translation, orientation, scale, scattering, finger structure, complicated background, uneven illumination, and collection posture. All these factors may contribute to inaccurate region of interest (ROI) definition, and so degrade the performance of finger vein identification system. To improve this problem, in this paper, we propose a finger vein ROI localization method that has high effectiveness and robustness against the above factors. The proposed method consists of a set of steps to localize ROIs accurately, namely segmentation, orientation correction, and ROI detection. Accurate finger region segmentation and correct calculated orientation can support each other to produce higher accuracy in localizing ROIs. Extensive experiments have been performed on the finger vein image database, MMCBNU_6000, to verify the robustness of the proposed method. The proposed method shows the segmentation accuracy of 100%. Furthermore, the average processing time of the proposed method is 22 ms for an acquired image, which satisfies the criterion of a real-time finger vein identification system.

Effective fingerprint quality estimation for diverse capture sensors

Recognizing the quality of fingerprints in advance can be beneficial for improving the performance of fingerprint recognition systems. The representative features to assess the quality of fingerprint images from different types of capture sensors are known to vary. In this paper, an effective quality estimation system that can be adapted for different types of capture sensors is designed by modifying and combining a set of features including orientation certainty, local orientation quality and consistency. The proposed system extracts basic features, and generates next level features which are applicable for various types of capture sensors. The system then uses the Support Vector Machine (SVM) classifier to determine whether or not an image should be accepted as input to the recognition system. The experimental results show that the proposed method can perform better than previous methods in terms of accuracy. In the meanwhile, the proposed method has an ability to eliminate residue images from the optical and capacitive sensors, and the coarse images from thermal sensors.

Loss of the Parkinson's disease-linked gene DJ-1 perturbs mitochondrial dynamics

Growing evidence highlights a role for mitochondrial dysfunction and oxidative stress as underlying contributors to Parkinson's disease (PD) pathogenesis. DJ-1 (PARK7) is a recently identified recessive familial PD gene. Its loss leads to increased susceptibility of neurons to oxidative stress and death. However, its mechanism of action is not fully understood. Presently, we report that DJ-1 deficiency in cell lines, cultured neurons, mouse brain and lymphoblast cells derived from DJ-1 patients display aberrant mitochondrial morphology. We also show that these DJ-1-dependent mitochondrial defects contribute to oxidative stress-induced sensitivity to cell death since reversal of this fragmented mitochondrial phenotype abrogates neuronal cell death. Reactive oxygen species (ROS) appear to play a critical role in the observed defects, as ROS scavengers rescue the phenotype and mitochondria isolated from DJ-1 deficient animals produce more ROS compared with control. Importantly, the aberrant mitochondrial phenotype can be rescued by the expression of Pink1 and Parkin, two PD-linked genes involved in regulating mitochondrial dynamics and quality control. Finally, we show that DJ-1 deficiency leads to altered autophagy in murine and human cells. Our findings define a mechanism by which the DJ-1-dependent mitochondrial defects contribute to the increased sensitivity to oxidative stress-induced cell death that has been previously reported.

Fusion of CT and MR images using an improved wavelet based method

The fusion of multimodal medical images plays an important role in many clinical applications as it can support more accurate information than any individual source image. This paper presents a novel approach for fusion of computed tomography (CT) and magnetic resonance (MR) images based on wavelet transform. The medical images to be fused are firstly decomposed into multiscale representations by the wavelet transform. Then, by considering the physical meaning of wavelet coefficients and the characteristics of the CT and MR images, the coefficients of the low frequency band and high frequency bands are treated with different schemes: the former is performed with a maximum-selection (MS) rule, and the latter is convolved with a Laplacian operator followed by a MS rule. Finally, the fused image is reconstructed by using the inverse wavelet transform with the combined wavelet coefficients. The performance of our method is qualitatively and quantitatively compared with some existing fusion approaches. The experimental results can demonstrate that the proposed method is a promising and effective technique for fusion of CT and MR images.

WITHDRAWN: Prognostic factors of biochemical recurrence after radical prostatectomy in Korean men with high-risk prostate cancer

This article has been withdrawn consistent with Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy). The Publisher apologizes for any inconvenience this may cause.

Cell cycle machinery and stroke

Stroke results from a transient or permanent reduction in blood flow to the brain. The mechanisms involving neuronal death following ischemic insult are complex and not fully understood. One signal which may control ischemic neuronal death is the inappropriate activation of cell cycle regulators including cyclins, cyclin dependent kinases (CDKs) and endogenous cyclin dependent kinase inhibitors (CDKIs). In dividing cells, activation of cell cycle machinery induces cell proliferation. In the context of terminally differentiated-neurons, however, aberrant activation of these elements triggers neuronal death. Indeed, there are several lines of correlative and functional evidence supporting this "cell cycle/neuronal death hypothesis". The objective of this review is to summarize the findings implicating cell cycle machinery in ischemic neuronal death from in vitro and in vivo studies. Importantly, determining and blocking the signaling pathway(s) by which these molecules act to mediate ischemic neuronal death, in conjunction with other targets may provide a viable therapeutic strategy for stroke damage.

Surgical Outcome of Endoscopic Carpal Tunnel Release in 100 Patients with Carpal Tunnel Syndrome

The purpose of this study is to present the surgical outcome of endoscopic carpal tunnel release (ECTR) for the treatment of carpal tunnel syndrome (CTS). One hundred and thirty-one procedures (36 right hands, 33 left hands and 31 bilateral hands) of single portal ECTR were performed upon 100 patients (age range: 36-77 years, mean age: 52.9 years; 98 women and 2 men) with electrodiagnostically proven CTS for 2.5 years from 2001. Preoperative clinical severity and results of electrodiagnostic studies were compared with surgical outcomes at the minimal 3-month postoperative period. Among 131 cases 125 (95.4 %) with complete or significant relief of symptoms were satisfied and 6 (4.6 %) with partial or no relief of symptoms were dissatisfied. There were 2 cases of major complications (one with ulnar nerve injury and the other with ulnar artery injury) that developed in our early experience of ECTR and 1 case of recurrence. The grade of electrodiagnostic abnormalities was associated with surgical outcome but there was no statistical significance between them. The severity of clinical findings, age at onset and symptom duration were not correlated with surgical outcome. In conclusion, ECTR surgery was effective in relieving the symptoms of CTS with a low complication rate after the learning curve period. Thus, ECTR can be an alternative to the traditional open surgery and can be the first procedure for CTS with several advantages over open methods.

C-shaped root canals of mandibular second molars in a Korean population: clinical observation andin vitroanalysis

AIM

To investigate the incidence and morphology of C-shaped root canals of the mandibular second molar in a Korean population.

METHODOLOGY

Through clinical observation, randomly selected 272 mandibular second molars of Korean patients were accessed and evaluated after taking radiographs for determination of working length. In an in vitro analysis, 96 extracted mandibular second molars of Korean patients were collected and embedded in resin using an Endodontic cube technique, and were sectioned at intervals of 1 mm. The specimens were then observed with a surgical microscope and were photographed. Canal configurations were assigned to one of three categories: Category I defined a C-shaped outline without any separation; Category II referred to those with canal configurations, where dentine separated one distinct canal from a buccal or lingual C-shaped canal; Category III had two or more discrete and separate canals.

RESULTS

In clinical observation, 89 of 272 teeth (32.7%) had C-shaped canals. Of the 96 teeth examined in vitro, 30 (31.3%) had C-shaped canals. Upon in vitro analysis, only 1 tooth at the subpulpal level and 10 teeth at the apical 1 mm level were categorized under Category III.

CONCLUSION

There was high prevalence of C-shaped root canals in the mandibular second molars of Koreans. C-shaped canals having semicolon and continuous shapes at the canal orifice have a high possibility of being divided into two or three canals in the apical region.