Search results

Diabetic retinopathy (DR) is one of the dangerous complications of diabetes. Its grade level must be tracked to manage its progress and to start the appropriate decision for treatment in time. Effective automated methods for the detection of DR and the classification of its severity stage are necessary to reduce the burden on ophthalmologists and diagnostic contradictions among manual readers.

In this research, convolutional neural network (CNN) was used based on colored retinal fundus images for the detection of DR and classification of its stages. CNN can recognize sophisticated features on the retina and provides an automatic diagnosis. The pre-trained VGG-16 CNN model was applied using a transfer learning (TL) approach to utilize the already learned parameters in the detection.

By conducting different experiments set up with different severity groupings, the achieved results are promising. The best-achieved accuracies for 2-class, 3-class, 4-class and 5-class classifications are 86.5, 80.5, 63.5 and 73.7, respectively.

In this research, VGG-16 was used to detect and classify DR stages using the TL approach. Different combinations of classes were used in the classification of DR severity stages to illustrate the ability of the model to differentiate between the classes and verify the effect of these changes on the performance of the model.

This paper provides practical advice for archaeologists and heritage specialists wishing to use ML approaches to identify archaeological features in high-resolution satellite imagery (or other remotely sensed data sources). We seek to balance the disproportionately optimistic literature related to the application of ML to archaeological prospection through a discussion of limitations, challenges and other difficulties. We further seek to raise awareness among researchers of the time, effort, expertise and resources necessary to implement ML successfully, so that they can make an informed choice between ML and manual inspection approaches.

Automated object detection has been the holy grail of archaeological remote sensing for the last two decades. Machine learning (ML) models have proven able to detect uniform features across a consistent background, but more variegated imagery remains a challenge. We set out to detect burial mounds in satellite imagery from a diverse landscape in Central Bulgaria using a pre-trained Convolutional Neural Network (CNN) plus additional but low-touch training to improve performance. Training was accomplished using MOUND/NOT MOUND cutouts, and the model assessed arbitrary tiles of the same size from the image. Results were assessed using field data.

Validation of results against field data showed that self-reported success rates were misleadingly high, and that the model was misidentifying most features. Setting an identification threshold at 60% probability, and noting that we used an approach where the CNN assessed tiles of a fixed size, tile-based false negative rates were 95–96%, false positive rates were 87–95% of tagged tiles, while true positives were only 5–13%. Counterintuitively, the model provided with training data selected for highly visible mounds (rather than all mounds) performed worse. Development of the model, meanwhile, required approximately 135 person-hours of work.

Our attempt to deploy a pre-trained CNN demonstrates the limitations of this approach when it is used to detect varied features of different sizes within a heterogeneous landscape that contains confounding natural and modern features, such as roads, forests and field boundaries. The model has detected incidental features rather than the mounds themselves, making external validation with field data an essential part of CNN workflows. Correcting the model would require refining the training data as well as adopting different approaches to model choice and execution, raising the computational requirements beyond the level of most cultural heritage practitioners.

Improving the pre-trained model’s performance would require considerable time and resources, on top of the time already invested. The degree of manual intervention required – particularly around the subsetting and annotation of training data – is so significant that it raises the question of whether it would be more efficient to identify all of the mounds manually, either through brute-force inspection by experts or by crowdsourcing the analysis to trained – or even untrained – volunteers. Researchers and heritage specialists seeking efficient methods for extracting features from remotely sensed data should weigh the costs and benefits of ML versus manual approaches carefully.

Our literature review indicates that use of artificial intelligence (AI) and ML approaches to archaeological prospection have grown exponentially in the past decade, approaching adoption levels associated with “crossing the chasm” from innovators and early adopters to the majority of researchers. The literature itself, however, is overwhelmingly positive, reflecting some combination of publication bias and a rhetoric of unconditional success. This paper presents the failure of a good-faith attempt to utilise these approaches as a counterbalance and cautionary tale to potential adopters of the technology. Early-majority adopters may find ML difficult to implement effectively in real-life scenarios.

Unlike many high-profile reports from well-funded projects, our paper represents a serious but modestly resourced attempt to apply an ML approach to archaeological remote sensing, using techniques like transfer learning that are promoted as solutions to time and cost problems associated with, e.g. annotating and manipulating training data. While the majority of articles uncritically promote ML, or only discuss how challenges were overcome, our paper investigates how – despite reasonable self-reported scores – the model failed to locate the target features when compared to field data. We also present time, expertise and resourcing requirements, a rarity in ML-for-archaeology publications.

Mouth segmentation is one of the challenging tasks of development in lip reading applications due to illumination, low chromatic contrast and complex mouth appearance. Recently, deep learning methods effectively solved mouth segmentation problems with state-of-the-art performances. This study presents a modified Mobile DeepLabV3 based technique with a comprehensive evaluation based on mouth datasets.

This paper presents a novel approach to mouth segmentation by Mobile DeepLabV3 technique with integrating decode and auxiliary heads. Extensive data augmentation, online hard example mining (OHEM) and transfer learning have been applied. CelebAMask-HQ and the mouth dataset from 15 healthy subjects in the department of rehabilitation medicine, Ramathibodi hospital, are used in validation for mouth segmentation performance.

Extensive data augmentation, OHEM and transfer learning had been performed in this study. This technique achieved better performance on CelebAMask-HQ than existing segmentation techniques with a mean Jaccard similarity coefficient (JSC), mean classification accuracy and mean Dice similarity coefficient (DSC) of 0.8640, 93.34% and 0.9267, respectively. This technique also achieved better performance on the mouth dataset with a mean JSC, mean classification accuracy and mean DSC of 0.8834, 94.87% and 0.9367, respectively. The proposed technique achieved inference time usage per image of 48.12 ms.

The modified Mobile DeepLabV3 technique was developed with extensive data augmentation, OHEM and transfer learning. This technique gained better mouth segmentation performance than existing techniques. This makes it suitable for implementation in further lip-reading applications.

This paper proposes a solution for recognizing human faces under mask-wearing. The lower part of human face is occluded and could not be used in the learning process of face recognition. So, the proposed solution is developed to recognize human faces on any available facial components which could be varied depending on wearing or not wearing a mask.

The proposed solution is developed based on the FaceNet framework, aiming to modify the existing facial recognition model to improve the performance of both scenarios of mask-wearing and without mask-wearing. Then, simulated masked-face images are computed on top of the original face images, to be used in the learning process of face recognition. In addition, feature heatmaps are also drawn out to visualize majority of parts of facial images that are significant in recognizing faces under mask-wearing.

The proposed method is validated using several scenarios of experiments. The result shows an outstanding accuracy of 99.2% on a scenario of mask-wearing faces. The feature heatmaps also show that non-occluded components including eyes and nose become more significant for recognizing human faces, when compared with the lower part of human faces which could be occluded under masks.

The convolutional neural network based solution is tuned up for recognizing human faces under a scenario of mask-wearing. The simulated masks on original face images are augmented for training the face recognition model. The heatmaps are then computed to prove that features generated from the top half of face images are correctly chosen for the face recognition.

Diagnosing brain tumors is a process that demands a significant amount of time and is heavily dependent on the proficiency and accumulated knowledge of radiologists. Over the traditional methods, deep learning approaches have gained popularity in automating the diagnosis of brain tumors, offering the potential for more accurate and efficient results. Notably, attention-based models have emerged as an advanced, dynamically refining and amplifying model feature to further elevate diagnostic capabilities. However, the specific impact of using channel, spatial or combined attention methods of the convolutional block attention module (CBAM) for brain tumor classification has not been fully investigated.

To selectively emphasize relevant features while suppressing noise, ResNet50 coupled with the CBAM (ResNet50-CBAM) was used for the classification of brain tumors in this research.

The ResNet50-CBAM outperformed existing deep learning classification methods like convolutional neural network (CNN), ResNet-CBAM achieved a superior performance of 99.43%, 99.01%, 98.7% and 99.25% in accuracy, recall, precision and AUC, respectively, when compared to the existing classification methods using the same dataset.

Since ResNet-CBAM fusion can capture the spatial context while enhancing feature representation, it can be integrated into the brain classification software platforms for physicians toward enhanced clinical decision-making and improved brain tumor classification.

This research has not been published anywhere else.

This study aims to analyze the impact of Artificial Intelligence (AI) and Machine Learning (ML) on heritage conservation and preservation, and to identify relevant future research trends, by applying scientometrics.

A total of 1,646 articles, published between 1985 and 2021, concerning research on the application of ML and AI in cultural heritage were collected from the Scopus database and analyzed using bibliometric methodologies.

The findings of this study have shown that although there is a very important increase in academic literature in relation to AI and ML, publications that specifically deal with these issues in relation to cultural heritage and its conservation and preservation are significantly limited.

This study enriches the academic outline by highlighting the limited literature in this context and therefore the need to advance the study of AI and ML as key elements that support heritage researchers and practitioners in conservation and preservation work.

This paper aims to propose a solution for detecting and grading diabetic retinopathy (DR) in retinal images using a convolutional neural network (CNN)-based approach. It could classify input retinal images into a normal class or an abnormal class, which would be further split into four stages of abnormalities automatically.

The proposed solution is developed based on a newly proposed CNN architecture, namely, DeepRoot. It consists of one main branch, which is connected by two side branches. The main branch is responsible for the primary feature extractor of both high-level and low-level features of retinal images. Then, the side branches further extract more complex and detailed features from the features outputted from the main branch. They are designed to capture details of small traces of DR in retinal images, using modified zoom-in/zoom-out and attention layers.

The proposed method is trained, validated and tested on the Kaggle dataset. The regularization of the trained model is evaluated using unseen data samples, which were self-collected from a real scenario from a hospital. It achieves a promising performance with a sensitivity of 98.18% under the two classes scenario.

The new CNN-based architecture (i.e. DeepRoot) is introduced with the concept of a multi-branch network. It could assist in solving a problem of an unbalanced dataset, especially when there are common characteristics across different classes (i.e. four stages of DR). Different classes could be outputted at different depths of the network.

The tourist industry has to adopt a big data-driven foresight approach to enhance decision-making in a post-COVID international landscape still marked by significant uncertainty and in which some megatrends have the potential to reshape society in the next decades. This paper, considering the opportunity offered by the application of the quantitative analysis on internet new data sources, proposes a prediction method using Google Trends data based on an estimated transfer function model.

The paper uses the time-series methods to model and predict Google Trends data. A transfer function model is used to transform the prediction of Google Trends data into predictions of tourist arrivals. It predicts the United States tourism demand in Italy.

The results highlight the potential expressed by the use of big data-driven foresight approach. Applying a transfer function model on internet search data, timely forecasts of tourism flows are obtained. The two scenarios emerged can be used in tourism stakeholders’ decision-making process. In a future perspective, the methodological path could be applied to other tourism origin markets, to other internet search engine or other socioeconomic and environmental contexts.

The study raises awareness of foresight literacy in the tourism sector. Secondly, it complements the research on tourism demand forecasting by evaluating the performance of quantitative forecasting techniques on new data sources. Thirdly, it is the first paper that makes the United States arrival predictions in Italy. Finally, the findings provide immediate valuable information to tourism stakeholders that could be used to make decisions.

Simulations exist for the prediction of the behaviour of building structural systems under fire, including two-way coupled fire-structure interaction. However, these simulations do not include detailed models of the connections, whereas these connections may impact the overall behaviour of the structure. Therefore, this paper proposes a two-scale method to include screw connections.

The two-scale method consists of (a) a global-scale model that models the overall structural system and (b) a small-scale model to describe a screw connection. Components in the global-scale model are connected by a spring element instead of a modelled screw, and the stiffness of this spring element is predicted by the small-scale model, updated at each load step. For computational efficiency, the small-scale model uses a proprietary technique to model the behaviour of the threads, verified by simulations that model the complete thread geometry, and validated by existing pull-out experiments. For four screw failure modes, load-deformation behaviour and failure predictions of the two-scale method are verified by a detailed system model. Additionally, the two-scale method is validated for a combined load case by existing experiments, and demonstrated for different temperatures. Finally, the two-scale method is illustrated as part of a two-way coupled fire-structure simulation.

It was shown that proprietary ”threaded connection interaction” can predict thread relevant failure modes, i.e. thread failure, shank tension failure, and pull-out. For bearing, shear, tension, and pull-out failure, load-deformation behaviour and failure predictions of the two-scale method correspond with the detailed system model and Eurocode predictions. Related to combined load cases, for a variety of experiments a good correlation has been found between experimental and simulation results, however, pull-out simulations were shown to be inconsistent.

More research is needed before the two-scale method can be used under all conditions. This relates to the failure criteria for pull-out, combined load cases, and temperature loads.

The two-scale method bridges the existing very detailed small-scale screw models with present global-scale structural models, that in the best case only use springs. It shows to be insightful, for it contains a functional separation of scales, revealing their relationships, and it is computationally efficient as it allows for distributed computing. Furthermore, local small-scale non-convergence (e.g. a screw failing) can be handled without convergence problems in the global-scale structural model.

In time and accurate detection of cancer can save the life of the person affected. According to the World Health Organization (WHO), breast cancer occupies the most frequent incidence among all the cancers whereas breast cancer takes fifth place in the case of mortality numbers. Out of many image processing techniques, certain works have focused on convolutional neural networks (CNNs) for processing these images. However, deep learning models are to be explored well.

In this work, multivariate statistics-based kernel principal component analysis (KPCA) is used for essential features. KPCA is simultaneously helpful for denoising the data. These features are processed through a heterogeneous ensemble model that consists of three base models. The base models comprise recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU). The outcomes of these base learners are fed to fuzzy adaptive resonance theory mapping (ARTMAP) model for decision making as the nodes are added to the F_2ˆa layer if the winning criteria are fulfilled that makes the ARTMAP model more robust.

The proposed model is verified using breast histopathology image dataset publicly available at Kaggle. The model provides 99.36% training accuracy and 98.72% validation accuracy. The proposed model utilizes data processing in all aspects, i.e. image denoising to reduce the data redundancy, training by ensemble learning to provide higher results than that of single models. The final classification by a fuzzy ARTMAP model that controls the number of nodes depending upon the performance makes robust accurate classification.

Research in the field of medical applications is an ongoing method. More advanced algorithms are being developed for better classification. Still, the scope is there to design the models in terms of better performance, practicability and cost efficiency in the future. Also, the ensemble models may be chosen with different combinations and characteristics. Only signal instead of images may be verified for this proposed model. Experimental analysis shows the improved performance of the proposed model. This method needs to be verified using practical models. Also, the practical implementation will be carried out for its real-time performance and cost efficiency.

The proposed model is utilized for denoising and to reduce the data redundancy so that the feature selection is done using KPCA. Training and classification are performed using heterogeneous ensemble model designed using RNN, LSTM and GRU as base classifiers to provide higher results than that of single models. Use of adaptive fuzzy mapping model makes the final classification accurate. The effectiveness of combining these methods to a single model is analyzed in this work.