Large Image Datasets Research Articles

Detection of cervical cancer using deep transfer learning

Cervical cancer, the fourth most prevalent disease among females, is one of the major threats to the health of women worldwide. The early detection of cervical cancer is crucial for efficient disease management and medical care since it increases the likelihood of treating and curing the disease. Medically cervical cancer is usually diagnosed using pap test or colposcopic examination. This tests are time consuming and lack of enough, but with technological advancement and application of artificial intelligence to health care, machine learning algorithms has be giving faster and more accurate prediction of illness. This study made use of the readily available SIPAKMed dataset which includes the collection of a large image dataset of 4049 images. Due to the fact that detection of this type of cancer using medical images gives a faster and more accurate detection compared to using structured dataset comprising of various attributes related to cervical cancer. Pretrained EfficientNet-B7 was used as the detection model. The dataset went through series of preprocessing techniques like filtering using low and high pass filter, normalization using Min-Max Scalar. Label and image data were then converted to an array. The developed system was evaluated and gave an average accuracy and precision of 87% and 87% respectively, it was then compared with Graph Neural Network and Random Forest but they were both outperformed. As a result of this, EfficientNet-B7 is a very good deep learning algorithm for detecting cervical cancer using medical images. Other architectures of convolutional neural networks should be experimented and their performance should be compared with the developed system in future work, while larger dataset should be utilized as well.

Bragg Spot Finder (BSF): a new machine-learning-aided approach to deal with spot finding for rapidly filtering diffraction pattern images

Macromolecular crystallography contributes significantly to understanding diseases and, more importantly, how to treat them by providing atomic resolution 3D structures of proteins. This is achieved by collecting X-ray diffraction images of protein crystals from important biological pathways. Spotfinders are used to detect the presence of crystals with usable data, and the spots from such crystals are the primary data used to solve the relevant structures. Having fast and accurate spot finding is essential, but recent advances in synchrotron beamlines used to generate X-ray diffraction images have brought us to the limits of what the best existing spotfinders can do. This bottleneck must be removed so spotfinder software can keep pace with the X-ray beamline hardware improvements and be able to see the weak or diffuse spots required to solve the most challenging problems encountered when working with diffraction images. In this paper, we first present Bragg Spot Detection (BSD), a large benchmark Bragg spot image dataset that contains 304 images with more than 66 000 spots. We then discuss the open source extensible U-Net-based spotfinder Bragg Spot Finder (BSF), with image pre-processing, a U-Net segmentation backbone, and post-processing that includes artifact removal and watershed segmentation. Finally, we perform experiments on the BSD benchmark and obtain results that are (in terms of accuracy) comparable to or better than those obtained with two popular spotfinder software packages (Dozor and DIALS), demonstrating that this is an appropriate framework to support future extensions and improvements.

Detecting white spot lesions on post-orthodontic oral photographs using deep learning based on the YOLOv5x algorithm: a pilot study

BackgroundDeep learning model trained on a large image dataset, can be used to detect and discriminate targets with similar but not identical appearances. The aim of this study is to evaluate the post-training performance of the CNN-based YOLOv5x algorithm in the detection of white spot lesions in post-orthodontic oral photographs using the limited data available and to make a preliminary study for fully automated models that can be clinically integrated in the future.MethodsA total of 435 images in JPG format were uploaded into the CranioCatch labeling software and labeled white spot lesions. The labeled images were resized to 640 × 320 while maintaining their aspect ratio before model training. The labeled images were randomly divided into three groups (Training:349 images (1589 labels), Validation:43 images (181 labels), Test:43 images (215 labels)). YOLOv5x algorithm was used to perform deep learning. The segmentation performance of the tested model was visualized and analyzed using ROC analysis and a confusion matrix. True Positive (TP), False Positive (FP), and False Negative (FN) values were determined.ResultsAmong the test group images, there were 133 TPs, 36 FPs, and 82 FNs. The model’s performance metrics include precision, recall, and F1 score values of detecting white spot lesions were 0.786, 0.618, and 0.692. The AUC value obtained from the ROC analysis was 0.712. The mAP value obtained from the Precision-Recall curve graph was 0.425.ConclusionsThe model’s accuracy and sensitivity in detecting white spot lesions remained lower than expected for practical application, but is a promising and acceptable detection rate compared to previous study. The current study provides a preliminary insight to further improved by increasing the dataset for training, and applying modifications to the deep learning algorithm.Clinical revelanceDeep learning systems can help clinicians to distinguish white spot lesions that may be missed during visual inspection.

Image Colorization Using Deep Learning

mage colorization using deep learning is a fascinating field that aims to add color to black and white images automatically. This project explores the use of advanced neural networks, specifically convolutional neural networks (CNNs), to achieve this task efficiently and accurately. By leveraging large datasets of color images paired with their black and white counterparts, the CNN learns to predict plausible colorizations for grayscale input images. The project demonstrates the effectiveness of deep learning techniques in recreating realistic colors while preserving the details of the original images. Through experimentation and evaluation, we showcase the potential of this approach in various applications, including historical photo restoration, artistic expression, and enhancing visual content in multimedia. Overall, this project contributes to the advancement of computer vision and image processing technologies, offering a powerful tool for enriching visual content effortlessly. Keywords: Image Colorization, Deep Learning, Convolutional Neural Networks, Generative

Predictive Modeling of Infection from Chest X-ray Images with ResNet34 and Deep learning CNN approach

The timely and precise detection of infected persons is still a major issue in the ongoing fight against the COVID-19 epidemic. With its ability to provide information on the severity and development of the illness, chest X-ray imaging has become a useful diagnostic and surveillance technique for COVID-19 pneumonia. This paper suggests a unique method for predictive modeling of COVID-19 infection from chest X-ray pictures by using the capabilities of deep learning, namely ResetNet architecture. In image classification tasks, ResNet34, a convolutional neural network (CNN) variation, has shown impressive performance, particularly in situations with little training data. Here, we report a comprehensive training and validation protocol for the ResetNet model using a large dataset of chest X-ray images that includes both positive and negative COVID-19 instances. To improve feature extraction, the dataset is preprocessed. All data size: 5856,Train size: 4684, Val size: 585, Test size: 587. To reduce overfitting and boost generalization, data augmentation methods are used. Our suggested method uses chest X-ray pictures to reliably distinguish COVID-19 patients from non-COVID-19 cases by using deep learning's discriminative capabilities. The result of model with TRAINING: Epoch: 19, Loss: 0.08099747663349896, Accuracy: 97.11and TEST: Epoch: 19, Loss: 0.08553645759820938, Accuracy: 96.76.

Child GAN: Face Aging and Rejuvenation to Find Missing Children

The Child GAN project seeks in order to resolve the crucial problem of missing child location by utilizing state-of-the art machine learning methods, particularly Generative Adversarial Networks (GANs). The project's main goal is to create a novel method that uses face aging and rejuvenation algorithms to create age- progressed images of missing children. Our GAN-based model learns complex patterns of facial aging and rejuvenation by utilizing large datasets of facial images taken at different ages. Over time, the model can produce realistic representations of how missing children might age or look rejuvenated by training on pairs of images that show individuals at different stages of life. The age-progressed images that are produced are extremely useful resources for communities, non-profits, and law enforcement agencies that are looking for missing children. Through the use of various media channels, such as social media and traditional media, we hope to increase the visibility of these images of missing children and expedite their prompt recovery.

NOVEL METHOD VITAMIN DEFICIENCY DETECTION USING ALEXNET DNN ALGORITHM

Vitamin deficiencies can have significant impacts on overall health and well-being. Early detection plays a crucial role in preventing complications and improving outcomes. However, traditional methods for detecting deficiencies can be time-consuming and costly. This project aims to develop a novel method for detecting vitamin deficiencies using the AlexNet DNN algorithm, a powerful deep learning model for image classification. The purpose of this project is to explore the feasibility of using image analysis and deep learning techniques to detect vitamin deficiencies accurately and efficiently. The objectives include improving the accuracy of detection, reducing false positives and negatives, and developing a reliable and accessible tool for early detection. To achieve our objectives, we will gather a large dataset of images depicting various vitamin deficiencies. These images will be preprocessed to enhance features and reduce noise. The AlexNet DNN algorithm will be trained on this dataset, learning to recognize patterns and features associated with different deficiencies. The algorithm will undergo rigorous testing and evaluation to ensure its effectiveness. KEYWORDS— Vitamins, Deficiency, AlexNet, Deep Neural Network(DNN), Effectiveness

Face Emotional Detection Using Neural Network

The face reaction detection using neural network is a very much popular topic in the artificial intelligence and compute vision field. The main goal of this research is to develop a system that can accurately detect the emotional system of a person based on these facial expression. The proposed system consists of so many stages, including face detection, feature extraction, and classification. In the first stage, the face is detected and cropped from the input image using a pre-trained face detection model. Then, the relevant facial features are, like, extracted, such as the position and shape of the eyebrows, mouth, and eyes, and other stuff. The extracted features are then fed into a neural network model, which is trained on a large dataset of labeled facial expressions to learn the relationship between the facial features and emotional states. To evaluate the performance of the proposed system, several metrics are used, like, accuracy, precision, recall, and F1-score. The system is tested on a large dataset of images with labeled emotional states, and the results show that, woah, the system achieves high accuracy and precision in detecting emotions. In conclusion, the proposed system is, you know, an effective approach for face emotional detection using neural networks. The system can be used in a variety of applications, such as human computer interaction and social robotics and emotion-based marketing, all those cool things.

Enhancing the medical diagnosis of COVID-19 with learning based decision support systems

Since late December 2019, the COVID-19 pandemic has had substantial impact and long-lasting impact on numerous lives. The surge in patients has overwhelmed hospitals and exhausted essential resources such as masks and gloves. However, in response to this crisis, we have developed a robust solution that can ease the burden on emergency services and manage the influx of patients. Our proposed framework comprises deep learning and machine learning models that can predict and manage patient demand with high accuracy. The first model, is specifically designed to classify computed tomography (CT) scan images for COVID or non-COVID cases. We trained multiple convolutional neural network (CNN) models on a large dataset of CT scan images and evaluated their performance on a separate test set. Our evaluation showed that the ResNet50 model was the most effective, achieving an accuracy of 93.28%. The second model uses patient measurements dataset to predict the likelihood of intensive care unit (ICU) admission for COVID-19 patients. We experimented with the XGBoost machine learning algorithm and found that the accuracy score achieved 88.40%.

A Learnable Counter-Condition Analysis Framework for Functional Connectivity-Based Neurological Disorder Diagnosis.

To understand the biological characteristics of neurological disorders with functional connectivity (FC), recent studies have widely utilized deep learning-based models to identify the disease and conducted post-hoc analyses via explainable models to discover disease-related biomarkers. Most existing frameworks consist of three stages, namely, feature selection, feature extraction for classification, and analysis, where each stage is implemented separately. However, if the results at each stage lack reliability, it can cause misdiagnosis and incorrect analysis in afterward stages. In this study, we propose a novel unified framework that systemically integrates diagnoses (i.e., feature selection and feature extraction) and explanations. Notably, we devised an adaptive attention network as a feature selection approach to identify individual-specific disease-related connections. We also propose a functional network relational encoder that summarizes the global topological properties of FC by learning the inter-network relations without pre-defined edges between functional networks. Last but not least, our framework provides a novel explanatory power for neuroscientific interpretation, also termed counter-condition analysis. We simulated the FC that reverses the diagnostic information (i.e., counter-condition FC): converting a normal brain to be abnormal and vice versa. We validated the effectiveness of our framework by using two large resting-state functional magnetic resonance imaging (fMRI) datasets, Autism Brain Imaging Data Exchange (ABIDE) and REST-meta-MDD, and demonstrated that our framework outperforms other competing methods for disease identification. Furthermore, we analyzed the disease-related neurological patterns based on counter-condition analysis.

From data to diagnosis: skin cancer image datasets forartificial intelligence.

Artificial intelligence (AI) solutions for skin cancer diagnosis continue to gain momentum, edging closer towards broad clinical use. These AI models, particularly deep-learning architectures, require large digital image datasets for development. This review provides an overview of the datasets used to develop AI algorithms and highlights the importance of dataset transparency for the evaluation of algorithm generalizability across varying populations and settings. Current challenges for curation of clinically valuable datasets are detailed, which include dataset shifts arising from demographic variations and differences in data collection methodologies, along with inconsistencies in labelling. These shifts can lead to differential algorithm performance, compromise of clinical utility, and the propagation of discriminatory biases when developed algorithms are implemented in mismatched populations. Limited representation of rare skin cancers and minoritized groups in existing datasets are highlighted, which can further skew algorithm performance. Strategies to address these challenges are presented, which include improving transparency, representation and interoperability. Federated learning and generative methods, which may improve dataset size and diversity without compromising privacy, are also examined. Lastly, we discuss model-level techniques that may address biases entrained through the use of datasets derived from routine clinical care. As the role of AI in skin cancer diagnosis becomes more prominent, ensuring the robustness of underlying datasets is increasingly important.

Coupling randomisation and sparse modelling for the exploratory analysis of large hyperspectral datasets

Sparse-based models are a powerful tools for data compression, variable reduction, and model complexity reduction. Nevertheless, their major issue is the high computational time needed in large matrices. This manuscript proposes, for the first time, to couple randomised decomposition as a first step before sparsity calculations, followed by a projection of the full data onto a reduced-sparse set of loadings that will drastically reduce the time needed for calculations and built models that are equally reliable as their sparse-based homologous. While this new approach might be valid for several scenarios (exploration, regression and classification), we will focus on exploration methods (like Principal Component Analysis – PCA) applied to large datasets of hyperspectral images. Two datasets of different complexity have been tested, and the benefits of the coupled randomisation and sparse PCA (rsPCA) are extensively studied.

Mammo-Light: A lightweight convolutional neural network for diagnosing breast cancer from mammography images

People of all countries, developed and developing alike endure cancer-related fatal diseases. The rate of breast cancer in females is increasing daily, partly due to ignorance and misdiagnosis in the early stages. Diagnosis of breast cancer accurately during its earlier stages of development can result in proper initial treatment for breast cancer. Artificial intelligence can aid in the acceleration and automation of breast cancer detection. Deep learning is decisive in effectively recognizing and classifying cancer on large datasets of medical images. In this paper, we propose a novel computer-aided classification approach, Mammo-Light for breast cancer prediction. Preprocessing strategies have been utilized to eradicate the noise and enhance mammogram lesions. Photometric augmentation techniques adapted to the preprocessed classes to balance and increase the size of the dataset. After that, a lightweight yet intuitive convolutional neural network is applied to classify breast cancer on the publicly available dataset CBIS-DDSM. For further validation of the proposed approach, we have used the MIAS dataset. Mammo-Light attained a 99.17% and 98.42% test accuracy respectively for CBIS-DDSM and MIAS datasets and outperformed state-of-the-art methods in terms of accuracy and other metrics. Due to being the lightweight model, Mammo-Light performs exceptionally well with fewer parameters and computational time, which can potentially contribute to the field of breast cancer early diagnosis and enable fast treatment.

Bridging ethnography and AI: a reciprocal methodology for studying visual political action

ABSTRACT This article proposes a methodological approach to address the analytical challenge of meaningfully studying visual politics in the current abundance of online image data. It proposes a novel methodological process of bridging ethnography and computational methods for analysing visual data in a manner that avoids mixing and blurring the boundaries of the two methods while establishing a continuous interchange between them. The methodological development enables investigating how youth participate through image creation and usage, both online and offline, by outlining repertoires of visual political action. We argue that combining ethnographic research with supervised deep learning-based AI and pre-trained neural networks allows for systematically analysing large image datasets while maintaining interpretative, analytical perspectives and contextual sensitivity. Our results show that the existence of common visual features in images, the re-evaluation of the image categories and the creation of subcategories constitute key aspects for an ethnographically-informed image classification.

An Automated Tool to Classify and Transform Unstructured MRI Data into BIDS Datasets.

The increasing use of neuroimaging in clinical research has driven the creation of many large imaging datasets. However, these datasets often rely on inconsistent naming conventions in image file headers to describe acquisition, and time-consuming manual curation is necessary. Therefore, we sought to automate the process of classifying and organizing magnetic resonance imaging (MRI) data according to acquisition types common to the clinical routine, as well as automate the transformation of raw, unstructured images into Brain Imaging Data Structure (BIDS) datasets. To do this, we trained an XGBoost model to classify MRI acquisition types using relatively few acquisition parameters that are automatically stored by the MRI scanner in image file metadata, which are then mapped to the naming conventions prescribed by BIDS to transform the input images to the BIDS structure. The model recognizes MRI types with 99.475% accuracy, as well as a micro/macro-averaged precision of 0.9995/0.994, a micro/macro-averaged recall of 0.9995/0.989, and a micro/macro-averaged F1 of 0.9995/0.991. Our approach accurately and quickly classifies MRI types and transforms unstructured data into standardized structures with little-to-no user intervention, reducing the barrier of entry for clinical scientists and increasing the accessibility of existing neuroimaging data.

Computer vision enables taxon-specific identification of African carnivore tooth marks on bone

Taphonomic works aim at discovering how paleontological and archaeofaunal assemblages were formed. They also aim at determining how hominin fossils were preserved or destroyed. Hominins and other mammal carnivores have been co-evolving, at least during the past two million years, and their potential interactions determined the evolution of human behavior. In order to understand all this, taxon-specific carnivore agency must be effectively identified in the fossil record. Until now, taphonomists have been able to determine, to some degree, hominin and carnivore inputs in site formation, and their interactions in the modification of part of those assemblages. However, the inability to determine agency more specifically has hampered the development of taphonomic research, whose methods are virtually identical to those used several decades ago (lagged by a high degree of subjectivity). A call for more objective and agent-specific methods would be a major contribution to the advancement of taphonomic research. Here, we present one of these advances. The use of computer vision (CV) on a large data set of images of tooth marks has enabled the objective discrimination of taxon-specific carnivore agency up to 88% of the testing sample. We highlight the significance of this method in an interdisciplinary interplay between traditional taphonomic-paleontological analysis and artificial intelligence-based computer science. The new questions that can be addressed with this will certainly bring important changes to several ideas on important aspects of the human evolutionary process.

ENHANCEMENT OF UNDERWATER IMAGES WITH ARTIFICIAL INTELLIGENCE

Abstract. Camera systems using optical sensors have made great progress in recent years in obtaining underwater images. Very high-resolution images can be obtained with underwater cameras. However, it becomes difficult to process the image obtained due to many distorting factors such as inhomogeneous underwater lighting, low contrast, blur, sea snow. For this purpose, image enhancement algorithms are used to minimise the problems in the captured images. The physical characteristics of the underwater environment cause degradation effects that are not found in normal images captured in air. Due to the effects such as colour distortion, low contrast and brightness, blurred details in images captured in this type of environment, the usage area of these images is very limited. This project aims to use artificial intelligence methods for colour enhancement of underwater images. In addition, one of the main objectives of the project is to examine the effect of colour enhancement applied to images on other areas of use. Within the scope of the project, An Underwater Image Enhancement Benchmark Dataset and Beyond Dataset and Large Scale Underwater Image Dataset datasets were used. In our study, U-Shape Transformer architecture and Pix2Pix architecture were tested for their usability for image enhancement. According to the results obtained, Pix2Pix architecture achieved the highest accuracy. Accuracy results were obtained as 22.5884 and 0.8764 for PSNR and SSIM, respectively using UIEB dataset. The accuracy values for LSUI dataset are 25.1010 and 0.8440 for PSNR and SSIM, respectively.

Calibration-free structured-light-based 3D scanning system in laparoscope for robotic surgery.

Accurate 3D shape measurement is crucial for surgical support and alignment in robotic surgery systems. Stereo cameras in laparoscopes offer a potential solution; however, their accuracy in stereo image matching diminishes when the target image has few textures. Although stereo matching with deep learning has gained significant attention, supervised learning requires a large dataset of images with depth annotations, which are scarce for laparoscopes. Thus, there is a strong demand to explore alternative methods for depth reconstruction or annotation for laparoscopes. Active stereo techniques are a promising approach for achieving 3D reconstruction without textures. In this study, a 3D shape reconstruction method is proposed using an ultra-small patterned projector attached to a laparoscopic arm to address these issues. The pattern projector emits a structured light with a grid-like pattern that features node-wise modulation for positional encoding. To scan the target object, multiple images are taken while the projector is in motion, and the relative poses of the projector and a camera are auto-calibrated using a differential rendering technique. In the experiment, the proposed method is evaluated by performing 3D reconstruction using images obtained from a surgical robot and comparing the results with a ground-truth shape obtained from X-rayCT.

Development of a deep learning model to distinguish the cause of optic disc atrophy using retinal fundus photography

The differential diagnosis for optic atrophy can be challenging and requires expensive, time-consuming ancillary testing to determine the cause. While Leber's hereditary optic neuropathy (LHON) and optic neuritis (ON) are both clinically significant causes for optic atrophy, both relatively rare in the general population, contributing to limitations in obtaining large imaging datasets. This study therefore aims to develop a deep learning (DL) model based on small datasets that could distinguish the cause of optic disc atrophy using only fundus photography. We retrospectively reviewed fundus photographs of 120 normal eyes, 30 eyes (15 patients) with genetically-confirmed LHON, and 30 eyes (26 patients) with ON. Images were split into a training dataset and a test dataset and used for model training with ResNet-18. To visualize the critical regions in retinal photographs that are highly associated with disease prediction, Gradient-Weighted Class Activation Map (Grad-CAM) was used to generate image-level attention heat maps and to enhance the interpretability of the DL system. In the 3-class classification of normal, LHON, and ON, the area under the receiver operating characteristic curve (AUROC) was 1.0 for normal, 0.988 for LHON, and 0.990 for ON, clearly differentiating each class from the others with an overall total accuracy of 0.93. Specifically, when distinguishing between normal and disease cases, the precision, recall, and F1 scores were perfect at 1.0. Furthermore, in the differentiation of LHON from other conditions, ON from others, and between LHON and ON, we consistently observed precision, recall, and F1 scores of 0.8. The model performance was maintained until only 10% of the pixel values of the image, identified as important by Grad-CAM, were preserved and the rest were masked, followed by retraining and evaluation.

Model-agnostic adversarial example detection via high-frequency amplification

Image classification based on Deep Neural Networks (DNNs) is vulnerable to adversarial examples, which make the classifier output incorrect predictions. One approach to defending against this attack is to detect whether the input is an adversarial example. Unfortunately, existing adversarial example detection methods heavily rely on the underlying classifier and may fail when the classifier is upgraded. In this paper, we propose a model-agnostic detection method that leverages high-frequency signals from adversarial noises in adversarial examples and does not need interactions with the underlying classifier. We amplify redundant high-frequency signals brought by adversarial noises and represent object boundaries with these signals in an image. Our key insight is that the boundaries extracted by redundant high-frequency signals have a strong correlation with the boundaries of images in adversarial examples, while this correlation does not exist in clean images. Furthermore, adversarial examples of large images have more high-frequency signals and make adversarial detection easier on large image datasets. Experimental results show that our method has good transferability and can accurately detect various adversarial examples on different datasets.