Fig 4 - uploaded by Yann Lecun
Content may be subject to copyright.
convolutional network applied to a sample image. (a) top: raw input image; bottom: pre-processed image; (b) state of layer C1; (c) layer C3; (d): layer C5; (e): output layer. The five output maps correspond to the five categories, respectively from top to bottom: nucleus, nucleus membrane, cytoplasm, cell wall, external medium. The properly segmented regions are clearly visible on the output maps.
Source publication
We describe a trainable system for analyzing videos of developing C. elegans embryos. The system automatically detects, segments, and locates cells and nuclei in microscopic images. The system was designed as the central component of a fully automated phenotyping system. The system contains three modules 1) a convolutional network trained to classi...
Contexts in source publication
Context 1
... 2 shows the size of each layer when a 40 × 40 pixel input is used and a single output vector is produced. Figure 4 shows the result of applying the convolutional network to an image, which produces a label image with 1/4 the resolution of the input. It would be straightforward to modify the method to produce a label image with the same resolution as the input. ...
Context 2
... DIC process creates an embossed "bas relief" look that, while pleasing to the human eye, makes processing the images quite challenging. For example the cell wall in the upper left region of the raw image in figure 4 looks quite different from the cell wall in the lower right region. We decided to design a linear filter that would make the images more isotropic, while preserving the texture information. ...
Context 3
... linear filter used was equivalent to computing the difference between the image and a suitably shifted version of it. A typical resulting image is shown in figure 4((a), bottom). The pixel intensities were then centered so that each image had zero mean, and scaled so that the standard deviation was 1.0. ...
Context 4
... and more importantly, the usefulness of the overall system will be determined by how well the cells and nuclei can be detected, located, counted, and measured. Figure 4 shows a sample image (top left), a pre-processed version of the image (bottom left), and the corresponding internal state and output of the convolutional network. Layers ...
Similar publications
Giant phospholipid vesicles obtained by the method of electroformation were observed by the phase contrast microscope. Most of these vesicles contain a protrusion which shortens in a slow shape transformation process until it is absorbed into the main vesicle body. We are concerned with the last stages of this shape transformation process, where th...
Little is known about the regulation of apoptosis in fibroblasts although several model systems including serum deprivation and treatment with staurosporine or topoisomerase inhibitors have been used to induce apoptosis in vitro. To validate a reproducible in vitro model for the study of apoptosis in fibroblasts, we cultured density-inhibited monol...
Citations
... A CNN is used to recognize human actions through the extraction of spatial features from the video frame sequences. Inspired by the performance of CNNs in various disciplines of computer vision and image processing, Ning et al. [22] presented a 2D CNN model that represents a single frame architecture, with the spatial feature vector extracted for each frame. The 2D CNN networks are primarily used to extract spatial features from video frame sequences and they are incapable to extract motion information from frame sequence. ...
... Texture-based human action recognition techniques [16] [18] [22] [24] are robust and efficient at extracting spatio-temporal features and have attracted researchers due to their outstanding performance and computational effectiveness. However, there are a number of limitations associated with these texture-based techniques, which are variants of LBP, such as their sensitivity to noise and limited capacity to collect more discriminative information. ...
Human action recognition research has become increasingly sophisticated in recent years because it is used in many different applications. Many techniques have been introduced to solve the problem of action recognition, but a number of challenges still need to be resolved. To describe a video in action recognition, it is necessary to efficiently collect and aggregate the spatial-temporal information. In this paper, we address the problem of human action recognition by combining handcrafted spatio-temporal features with deep spatial features. This paper proposed a novel method for recognizing human actions in videos by concatenating handcrafted spatio-temporal texture features extracted by our proposed feature descriptor Volume Local Derivative Gradient Ternary Patterns (VLDGTP) and deep spatial features extracted from a modified Inception-v4 network. In order to reduce the dimension and to get equal sized feature vectors, we employed the PCA dimensionality reduction technique for both types of features. Then, the dimensionality reduced feature vectors are combined and passed to the SVM classifier for action recognition. Extensive experimentation is carried out on three standard datasets, such as KTH, UCF101 and HMDB-51 datasets. The experimental results and analysis show that our proposed method (VLDGTP + DEEP_FEATURES) yields better recognition accuracy for all three benchmark datasets compared to its competitors.
... It is popular in computer vision applications. CNN has been successfully applied to biological image segmentation [28], traffic sign recognition [29][30][31], face recognition [32,33], and other fields. Later studies added rectified linear units, dropout, data augmentation, and other techniques to CNN. ...
According to the World Health Organization statistics, as of 25 October 2022, there have been 625,248,843 confirmed cases of COVID-19, including 65,622,281 deaths worldwide. The spread and severity of COVID-19 are alarming. The economy and life of countries worldwide have been greatly affected. The rapid and accurate diagnosis of COVID-19 directly affects the spread of the virus and the degree of harm. Currently, the classification of chest X-ray or CT images based on artificial intelligence is an important method for COVID-19 diagnosis. It can assist doctors in making judgments and reduce the misdiagnosis rate. The convolutional neural network (CNN) is very popular in computer vision applications, such as applied to biological image segmentation, traffic sign recognition, face recognition, and other fields. It is one of the most widely used machine learning methods. This paper mainly introduces the latest deep learning methods and techniques for diagnosing COVID-19 using chest X-ray or CT images based on the convolutional neural network. It reviews the technology of CNN at various stages, such as rectified linear units, batch normalization, data augmentation, dropout, and so on. Several well-performing network architectures are explained in detail, such as AlexNet, ResNet, DenseNet, VGG, GoogleNet, etc. We analyzed and discussed the existing CNN automatic COVID-19 diagnosis systems from sensitivity, accuracy, precision, specificity, and F1 score. The systems use chest X-ray or CT images as datasets. Overall, CNN has essential value in COVID-19 diagnosis. All of them have good performance in the existing experiments. If expanding the datasets, adding GPU acceleration and data preprocessing techniques, and expanding the types of medical images, the performance of CNN will be further improved. This paper wishes to make contributions to future research.
... O. Ronneberger et al.'s U-Net architecture for medical imaging [14] draws inspiration from the Fully Convolutional Networks (FCN), particularly in its design of expanding and contracting segments. Similar to DeConvNet [24], U-Net utilizes a combination of downsampling and upsampling techniques to retain information that could be lost in the downsampling process. The contracting part of the network is chiefly tasked with generating feature maps, using 3x3 convolutions to effectively extract features during the downsampling phase.On the other hand, deconvolution is utilised during the expanding stage, which results in fewer feature maps but increased spatial dimensions (height, width). ...
Recently, new advances in deep learning algorithms have yielded some fascinating results in the field of computer vision technology. As a result, it can now perform activities that formerly required the use of human vision and the brain. Classification, object identification, and semantic segmentation have all seen substantial advancements in deep learning architecture in the last few years. For still images and movies, there has been a major advancement in the field of semantic segmentation. In practical uses like autonomous vehicles, segmenting semantic video continues to be difficult due to high-performance standards, the high cost of convolutional neural networks (CNNs), and the significant need for low latency. An effective machine-learning environment will be developed to meet the performance and latency challenges outlined above. The use of deep learning architectures like SegNet and FlowNet2.0 on the CamVid dataset enables this environment to conduct pixel-wise semantic segmentation of video properties while maintaining low latency. As a result, it is ideally suited for real-world applications since it takes advantage of both SegNet and FlowNet topologies. The decision network determines whether an image frame should be processed by a segmentation network or an optical flow network based on the expected confidence score. In conjunction with adaptive scheduling of the key frame approach, this technique for decision-making can help to speed up the procedure. Using the ResNet50 SegNet model, a mean Intersection on Union (IoU) of "54.27 percent" and an average frame per second of "19.57" were observed. Aside from decision network and adaptive key frame sequencing, it was discovered that FlowNet2.0 increased the frames processed per second9(fps) to "30.19" on GPU with a mean IoU of "47.65%". Because the GPU was utilized "47.65%" of the time, this resulted. There has been an increase in the speed of the Video semantic segmentation network without sacrificing quality, as demonstrated by this improvement in performance.
... Deep learning methods, often utilized in video and audio processing, have been demonstrated to learn language processing and picture processing from text repositories and text vectors in neural networks (NNs) [18][19][20][21][22]. CNN has played a vital role as a particular type of NN and is currently at the center of several profound learning applications. ...
Sentiment analysis holds great importance within the domain of natural language processing as it examines both the expressed and underlying emotions conveyed through review content. Furthermore, researchers have discovered that relying solely on the overall sentiment derived from the textual content is inadequate. Consequently, sentiment analysis was developed to extract nuanced expressions from textual information. One of the challenges in this field is effectively extracting emotional elements using multi-label data that covers various aspects. This article presents a novel approach called the Ensemble of DenseNet based on Aquila Optimizer (EDAO). EDAO is specifically designed to enhance the precision and diversity of multi-label learners. Unlike traditional multi-label methods, EDAO strongly emphasizes improving model diversity and accuracy in multi-label scenarios. To evaluate the effectiveness of our approach, we conducted experiments on seven distinct datasets, including emotions, hotels, movies, proteins, automobiles, medical, news, and birds. Our initial strategy involves establishing a preprocessing mechanism to obtain precise and refined data. Subsequently, we used the Vader tool with Bag of Words (BoW) for feature extraction. In the third stage, we created word associations using the word2vec method. The improved data were also used to train and test the DenseNet model, which was fine-tuned using the Aquila Optimizer (AO). On the news, emotion, auto, bird, movie, hotel, protein, and medical datasets, utilizing the aspect-based multi-labeling technique, we achieved accuracy rates of 95%, 97%, and 96%, respectively, with DenseNet-AO. Our proposed model demonstrates that EDAO outperforms other standard methods across various multi-label datasets with different dimensions. The implemented strategy has been rigorously validated through experimental results, showcasing its effectiveness compared to existing benchmark approaches.
... For instance, face recognition (1) utilizes semantic segmentation to identify specific parts of a person's face, while autonomous driving (2,3) requires the real-time monitoring of various types of semantic information on the road to avoid accidents. Medical image segmentation (4) plays a crucial role in accurately identifying disease-specific areas at the pixel level, assisting doctors in formulating treatment plans for patients. ...
... One of the main issues is a class imbalance, which can cause the learning process to ignore infrequent tumor sub-regions. Additionally, treating the entire image as a single entity is similar to processing numerous patches in a batch (Ning et al., 2005), which can lead to slower convergence rates hence increasing the use of computational resources. Previous research has shown that updating weights after each sample can be a more efficient approach than updating weights after accumulating gradients over a batch of samples (LeCun et al., 2002). ...
The realm of medical image diagnosis has advanced significantly with the integration of computer-aided diagnosis and surgical systems. However, challenges persist, particularly in achieving precise image segmentation. While deep learning techniques show potential, obstacles like limited resources, slow convergence, and class imbalance impede their effectiveness. Traditional patch-based methods, though common, struggle to capture intricate tumor boundaries and often lead to redundant samples, compromising computational efficiency and feature quality. To tackle these issues, this research introduces an innovative approach centered on the tumor itself for patch-based image analysis. This novel tumor-centered patching method aims to address the class imbalance and boundary deficiencies, enabling focused and accurate tumor segmentation. By aligning patches with the tumor's anatomical context, this technique enhances feature extraction accuracy and reduces computational load. Experimental results demonstrate improved class imbalance, with segmentation scores of 0.78, 0.76, and 0.71 for whole, core, and enhancing tumors, respectively using a lightweight simple U-Net. This approach shows potential for enhancing medical image segmentation and improving computer-aided diagnosis systems.
... 61], ilerleme göstermektedir. Önceki yaklaşımlarda, her pikselin, nesnenin sınıfı ile etiketlendiği semantik segmentasyon[55,[62][63][64][65][66], için evrişimli sinir ağları kullanılmakta olup piksel bazında sınıflandırmaya dayalı yöntemler için, görüntünün merkez vokseli, çoklu evrişimli katmanlardan geçirildikten sonra düzleştirilmiş (flatten) özellik haritalarına dayalı olarak tam bağlantılı katmanlar tarafından sınıflandırılmaktadır. Shelhamer vd.[67] ilk önce, tam bağlı katmanı bir evrişim katmanı ile değiştiren bir ESA önerdi. ...
Tıbbi görüntüleme alanında derin öğrenme modeli, mevcut yöntemlere kıyasla zaman ve performans açısından daha fazla katkıda bulunan bir modeldir. Görüntülerin otomatik olarak bölütlenmesini veya sınıflandırılmasını kapsar. Mevcut yöntemler ile tek katmanlı görüntüler üzerinden işlem yapılırken, derin öğrenme modeli ile çok katmanlı görüntüler üzerinden çalışma performansı daha yüksek ve daha kesin sonuçlar elde edilebilir. Son zamanlardaki gelişmeler, bu yaklaşımların tıbbi görüntülerdeki örüntülerin tanımlanması ve nicelendirilmesinde oldukça etkili olduğunu göstermektedir. Bu ilerlemelerin en önemli nedeni, derin öğrenme yaklaşımlarının doğrudan görüntülerden hiyerarşik özellik temsilleri elde etme yeteneğidir. Bu nedenle, derin öğrenme yöntemlerinin tıbbi görüntü işleme ve bölütleme alanındaki uygulamaları hızla en son teknolojiye dönüşmektedir ve klinik uygulamalarda performans iyileştirmeleri sağlamaktadır. Bu makalede, derin öğrenme yaklaşımlarının biyomedikal görüntülerin bölütlenmesi için uygulamaları, yöntemleri ve içerikleri genel bir bakış açısıyla incelenmiştir.
... The former is better answered by translation-invariant global features, while the latter requires well-localized features, posing a challenge to deep models. CNNs for pixel-level classification first appeared in the mid-2000s (Ning et al., 2005), but their use accelerated after the seminal paper on FCNs by Long et al. (2015), which, along with U-Net (Ronneberger et al., 2015), have become the basis for many state-of-the-art segmentation models. In contrast to classification CNNs (e.g., LeNet, AlexNet, VGG, GoogLeNet, ResNet), FCNs easily cope with arbitrary-sized input images. ...
Skin cancer is a major public health problem that could benefit from computer-aided diagnosis to reduce the burdenof this common disease. Skin lesion segmentation from images is an important step toward achieving this goal.However, the presence of natural and artificial artifacts (e.g., hair and air bubbles), intrinsic factors (e.g., lesionshape and contrast), and variations in image acquisition conditions make skin lesion segmentation a challenging task.Recently, various researchers have explored the applicability of deep learning models to skin lesion segmentation. Inthis survey, we cross-examine 177 research papers that deal with deep learning-based segmentation of skin lesions.We analyze these works along several dimensions, including input data (datasets, preprocessing, and synthetic datageneration), model design (architecture, modules, and losses), and evaluation aspects (data annotation requirementsand segmentation performance). We discuss these dimensions both from the viewpoint of select seminal works, andfrom a systematic viewpoint, examining how those choices have influenced current trends, and how their limitationsshould be addressed. To facilitate comparisons, we summarize all examined works in a comprehensive table as wellas an interactive table available online
... Deep learning technology has lately been used in various industries to evaluate medical images, and it is particularly good at tasks like segmentation and registration. Several CNN architectures have been suggested that feed whole images with increased image resolution (6). Fully CNN (fCNN) was developed for segmenting images and was first introduced by Long et al. (7). ...
Introduction
The primary factor for cardiovascular disease and upcoming cardiovascular events is atherosclerosis. Recently, carotid plaque texture, as observed on ultrasonography, is varied and difficult to classify with the human eye due to substantial inter-observer variability. High-resolution magnetic resonance (MR) plaque imaging offers naturally superior soft tissue contrasts to computed tomography (CT) and ultrasonography, and combining different contrast weightings may provide more useful information. Radiation freeness and operator independence are two additional benefits of M RI. However, other than preliminary research on MR texture analysis of basilar artery plaque, there is currently no information addressing MR radiomics on the carotid plaque.
Methods
For the automatic segmentation of MRI scans to detect carotid plaque for stroke risk assessment, there is a need for a computer-aided autonomous framework to classify MRI scans automatically. We used to detect carotid plaque from MRI scans for stroke risk assessment pre-trained models, fine-tuned them, and adjusted hyperparameters according to our problem.
Results
Our trained YOLO V3 model achieved 94.81% accuracy, RCNN achieved 92.53% accuracy, and MobileNet achieved 90.23% in identifying carotid plaque from MRI scans for stroke risk assessment. Our approach will prevent incorrect diagnoses brought on by poor image quality and personal experience.
Conclusion
The evaluations in this work have demonstrated that this methodology produces acceptable results for classifying magnetic resonance imaging (MRI) data.
... rock images. To address the lack of traditional image segmentation, some scholars have used convolution neural network (CNN) for image semantic segmentation (Ning et al., 2005;Cireşan et al., 2012;Farabet et al., 2013;Ganin and Lempitsky, 2014;Gupta et al., 2014;Pinheiro and Collobert, 2014). ...
