RESUMEN

OBJECTIVES: The purpose of this study was to generate radiographs including dentigerous cysts by applying the latest generative adversarial network (GAN; StyleGAN3) to panoramic radiography. METHODS: A total of 459 cystic lesions were selected, and 409 images were randomly assigned as training data and 50 images as test data. StyleGAN3 training was performed for 500 000 images. Fifty generated images were objectively evaluated by comparing them with 50 real images according to four metrics: Fréchet inception distance (FID), kernel inception distance (KID), precision and recall, and inception score (IS). A subjective evaluation of the generated images was performed by three specialists who compared them with the real images in a visual Turing test. RESULTS: The results of the metrics were as follows: FID, 199.28; KID, 0.14; precision, 0.0047; recall, 0.00; and IS, 2.48. The overall results of the visual Turing test were 82.3%. No significant difference was found in the human scoring of root resorption. CONCLUSIONS: The images generated by StyleGAN3 were of such high quality that specialists could not distinguish them from the real images.

RESUMEN

Images, texts, voices, and signals can be synthesized by latent spaces in a multidimensional vector, which can be explored without the hurdles of noise or other interfering factors. In this paper, we present a practical use case that demonstrates the power of latent space in exploring complex realities such as image space. We focus on DaVinciFace, an AI-based system that explores the StyleGAN2 space to create a high-quality portrait for anyone in the style of the Renaissance genius Leonardo da Vinci. The user enters one of their portraits and receives the corresponding Da Vinci-style portrait as an output. Since most of Da Vinci's artworks depict young and beautiful women (e.g., "La Belle Ferroniere", "Beatrice de' Benci"), we investigate the ability of DaVinciFace to account for other social categorizations, including gender, race, and age. The experimental results evaluate the effectiveness of our methodology on 1158 portraits acting on the vector representations of the latent space to produce high-quality portraits that retain the facial features of the subject's social categories, and conclude that sparser vectors have a greater effect on these features. To objectively evaluate and quantify our results, we solicited human feedback via a crowd-sourcing campaign. Analysis of the human feedback showed a high tolerance for the loss of important identity features in the resulting portraits when the Da Vinci style is more pronounced, with some exceptions, including Africanized individuals.

RESUMEN

Background and objective: Due to the high prevalence of dental caries, fixed dental restorations are regularly required to restore compromised teeth or replace missing teeth while retaining function and aesthetic appearance. The fabrication of dental restorations, however, remains challenging due to the complexity of the human masticatory system as well as the unique morphology of each individual dentition. Adaptation and reworking are frequently required during the insertion of fixed dental prostheses (FDPs), which increase cost and treatment time. This article proposes a data-driven approach for the partial reconstruction of occlusal surfaces based on a data set that comprises 92 3D mesh files of full dental crown restorations. Methods: A Generative Adversarial Network (GAN) is considered for the given task in view of its ability to represent extensive data sets in an unsupervised manner with a wide variety of applications. Having demonstrated good capabilities in terms of image quality and training stability, StyleGAN-2 has been chosen as the main network for generating the occlusal surfaces. A 2D projection method is proposed in order to generate 2D representations of the provided 3D tooth data set for integration with the StyleGAN architecture. The reconstruction capabilities of the trained network are demonstrated by means of 4 common inlay types using a Bayesian Image Reconstruction method. This involves pre-processing the data in order to extract the necessary information of the tooth preparations required for the used method as well as the modification of the initial reconstruction loss. Results: The reconstruction process yields satisfactory visual and quantitative results for all preparations with a root mean square error (RMSE) ranging from 0.02 mm to 0.18 mm. When compared against a clinical procedure for CAD inlay fabrication, the group of dentists preferred the GAN-based restorations for 3 of the total 4 inlay geometries. Conclusions: This article shows the effectiveness of the StyleGAN architecture with a downstream optimization process for the reconstruction of 4 different inlay geometries. The independence of the reconstruction process and the initial training of the GAN enables the application of the method for arbitrary inlay geometries without time-consuming retraining of the GAN.

RESUMEN

BACKGROUND AND OBJECTIVE: Melanoma, a malignant form of skin cancer, is a critical health concern worldwide. Early and accurate detection plays a pivotal role in improving patient's conditions. Current diagnosis of skin cancer largely relies on visual inspections such as dermoscopy examinations, clinical screening and histopathological examinations. However, these approaches are characterized by low efficiency, high costs, and a lack of guaranteed accuracy. Consequently, deep learning based techniques have emerged in the field of melanoma detection, successfully aiding in improving the accuracy of diagnosis. However, the high similarity between benign and malignant melanomas, combined with the class imbalance issue in skin lesion datasets, present a significant challenge in further improving the diagnosis accuracy. We propose a two-stage framework for melanoma detection to address these issues. METHODS: In the first stage, we use Style Generative Adversarial Networks with Adaptive discriminator augmentation synthesis to generate realistic and diverse melanoma images, which are then combined with the original dataset to create an augmented dataset. In the second stage, we utilize a vision Transformer of BatchFormer to extract features and detect melanoma or non-melanoma skin lesions on the augmented dataset obtained in the previous step, specifically, we employed a dual-branch training strategy in this process. RESULTS: Our experimental results on the ISIC2020 dataset demonstrate the effectiveness of the proposed approach, showing a significant improvement in melanoma detection. The method achieved an accuracy of 98.43%, an AUC value of 98.63%, and an F1 value of 99.01%, surpassing some existing methods. CONCLUSION: The method is feasible, efficient, and achieves early melanoma screening. It significantly enhances detection accuracy and can assist physicians in diagnosis to a great extent.

Asunto(s)

RESUMEN

With the rapid development of artificial intelligence, automated endoscopy-assisted diagnostic systems have become an effective tool for reducing the diagnostic costs and shortening the treatment cycle of patients. Typically, the performance of these systems depends on deep learning models which are pre-trained with large-scale labeled data, for example, early gastric cancer based on endoscopic images. However, the expensive annotation and the subjectivity of the annotators lead to an insufficient and class-imbalanced endoscopic image dataset, and these datasets are detrimental to the training of deep learning models. Therefore, we proposed a Swin Transformer encoder-based StyleGAN (STE-StyleGAN) for unbalanced endoscopic image enhancement, which is composed of an adversarial learning encoder and generator. Firstly, a pre-trained Swin Transformer is introduced into the encoder to extract multi-scale features layer by layer from endoscopic images. The features are subsequently fed into a mapping block for aggregation and recombination. Secondly, a self-attention mechanism is applied to the generator, which adds detailed information of the image layer by layer through recoded features, enabling the generator to autonomously learn the coupling between different image regions. Finally, we conducted extensive experiments on a private intestinal metaplasia grading dataset from a Grade-A tertiary hospital. The experimental results show that the images generated by STE-StyleGAN are closer to the initial image distribution, achieving a Fréchet Inception Distance (FID) value of 100.4. Then, these generated images are used to enhance the initial dataset to improve the robustness of the classification model, and achieved a top accuracy of 86 %.

Asunto(s)

RESUMEN

Recent evidence shows that AI-generated faces are now indistinguishable from human faces. However, algorithms are trained disproportionately on White faces, and thus White AI faces may appear especially realistic. In Experiment 1 (N = 124 adults), alongside our reanalysis of previously published data, we showed that White AI faces are judged as human more often than actual human faces-a phenomenon we term AI hyperrealism. Paradoxically, people who made the most errors in this task were the most confident (a Dunning-Kruger effect). In Experiment 2 (N = 610 adults), we used face-space theory and participant qualitative reports to identify key facial attributes that distinguish AI from human faces but were misinterpreted by participants, leading to AI hyperrealism. However, the attributes permitted high accuracy using machine learning. These findings illustrate how psychological theory can inform understanding of AI outputs and provide direction for debiasing AI algorithms, thereby promoting the ethical use of AI.

Asunto(s)

RESUMEN

Reconstructing visual experience from brain responses measured by functional magnetic resonance imaging (fMRI) is a challenging yet important research topic in brain decoding, especially it has proved more difficult to decode visually similar stimuli, such as faces. Although face attributes are known as the key to face recognition, most existing methods generally ignore how to decode facial attributes more precisely in perceived face reconstruction, which often leads to indistinguishable reconstructed faces. To solve this problem, we propose a novel neural decoding framework called VSPnet (voxel2style2pixel) by establishing hierarchical encoding and decoding networks with disentangled latent representations as media, so that to recover visual stimuli more elaborately. And we design a hierarchical visual encoder (named HVE) to pre-extract features containing both high-level semantic knowledge and low-level visual details from stimuli. The proposed VSPnet consists of two networks: Multi-branch cognitive encoder and style-based image generator. The encoder network is constructed by multiple linear regression branches to map brain signals to the latent space provided by the pre-extracted visual features and obtain representations containing hierarchical information consistent to the corresponding stimuli. We make the generator network inspired by StyleGAN to untangle the complexity of fMRI representations and generate images. And the HVE network is composed of a standard feature pyramid over a ResNet backbone. Extensive experimental results on the latest public datasets have demonstrated the reconstruction accuracy of our proposed method outperforms the state-of-the-art approaches and the identifiability of different reconstructed faces has been greatly improved. In particular, we achieve feature editing for several facial attributes in fMRI domain based on the multiview (i.e., visual stimuli and evoked fMRI) latent representations.

Asunto(s)

RESUMEN

Accurate semantic editing of the generated images is extremely important for machine learning and sample enhancement of big data. Aiming at the problem of semantic entanglement in generated image latent space of the StyleGAN2 network, we proposed a generated image editing method based on global-local Jacobi disentanglement. In terms of global disentanglement, we extract the weight matrix of the style layer in the pre-trained StyleGAN2 network; obtain the semantic attribute direction vector by using the weight matrix eigen decomposition method; finally, utilize this direction vector as the initialization vector for the Jacobi orthogonal regularization search algorithm. Our method improves the speed of the Jacobi orthogonal regularization search algorithm with the proportion of effective semantic attribute editing directions. In terms of local disentanglement, we design a local contrast regularized loss function to relax the semantic association local area and non-local area and utilize the Jacobi orthogonal regularization search algorithm to obtain a more accurate semantic attribute editing direction based on the local area prior MASK. The experimental results show that the proposed method achieves SOTA in semantic attribute disentangled metrics and can discover more accurate editing directions compared with the mainstream unsupervised generated image editing methods.

RESUMEN

Introduction: Artificial intelligence (AI) in healthcare can enhance clinical workflows and diagnoses, particularly in large-scale operations like COVID-19 mass testing. This study presents a deep Convolutional Neural Network (CNN) model for automated COVID-19 RATD image classification. Methods: To address the absence of a RATD image dataset, we crowdsourced 900 real-world images focusing on positive and negative cases. Rigorous data augmentation and StyleGAN2-ADA generated simulated images to overcome dataset limitations and class imbalances. Results: The best CNN model achieved a 93% validation accuracy. Test accuracies were 88% for simulated datasets and 82% for real datasets. Augmenting simulated images during training did not significantly improve real-world test image performance but enhanced simulated test image performance. Discussion: The findings of this study highlight the potential of the developed model in expediting COVID-19 testing processes and facilitating large-scale testing and tracking systems. The study also underscores the challenges in designing and developing such models, emphasizing the importance of addressing dataset limitations and class imbalances. Conclusion: This research contributes to the deployment of large-scale testing and tracking systems, offering insights into the potential applications of AI in mitigating outbreaks similar to COVID-19. Future work could focus on refining the model and exploring its adaptability to other healthcare scenarios.

RESUMEN

Generative artificial intelligence can be applied to medical imaging on tasks such as privacy-preserving image generation and superresolution and denoising of existing images. Few prior approaches have used cardiac magnetic resonance imaging (cMRI) as a modality given the complexity of videos (the addition of the temporal dimension) as well as the limited scale of publicly available datasets. We introduce GANcMRI, a generative adversarial network that can synthesize cMRI videos with physiological guidance based on latent space prompting. GANcMRI uses a StyleGAN framework to learn the latent space from individual video frames and leverages the timedependent trajectory between end-systolic and end-diastolic frames in the latent space to predict progression and generate motion over time. We proposed various methods for modeling latent time-dependent trajectories and found that our Frame-to-frame approach generates the best motion and video quality. GANcMRI generated high-quality cMRI image frames that are indistinguishable by cardiologists, however, artifacts in video generation allow cardiologists to still recognize the difference between real and generated videos. The generated cMRI videos can be prompted to apply physiologybased adjustments which produces clinically relevant phenotypes recognizable by cardiologists. GANcMRI has many potential applications such as data augmentation, education, anomaly detection, and preoperative planning.

RESUMEN

Research in person and face perception has broadly focused on group-level consensus that individuals hold when making judgments of others (e.g., "X type of face looks trustworthy"). However, a growing body of research demonstrates that individual variation is larger than shared, stimulus-level variation for many social trait judgments. Despite this insight, little research to date has focused on building and explaining individual models of face perception. Studies and methodologies that have examined individual models are limited in what visualizations they can reliably produce to either noisy and blurry or computer avatar representations. Methods that produce low-fidelity visual representations inhibit generalizability by being clearly computer manipulated and produced. In the present work, we introduce a novel paradigm to visualize individual models of face judgments by leveraging state-of-the-art computer vision methods. Our proposed method can produce a set of photorealistic face images that correspond to an individual's mental representation of a specific attribute across a variety of attribute intensities. We provide a proof-of-concept study which examines perceived trustworthiness/untrustworthiness and masculinity/femininity. We close with a discussion of future work to substantiate our proposed method.

RESUMEN

StyleGAN is now capable of achieving excellent results, especially high-quality face synthesis. Recently, some studies have tried to invert real face images into style latent space via StyleGAN. However, morphing real faces via latent representation is still in its infancy. Training costs are high and/or require huge samples with labels. By adding regularization to style optimization, we propose a novel method to morph real faces based on StyleGAN. To do the supervised task, we label latent vectors via synthesized faces and release the label set; then we utilize logistic regression to fast discover interpretable directions in latent space. Appropriate regularization helps us to optimize both latent vectors (faces and directions). Moreover, we use learned directions under different layer representations to handle real face morphing. Compared to the existing methods, our method faster yields a larger diverse and realistic output. Code and cases are available at https://github.com/disanda/RFM.

Asunto(s)

RESUMEN

Although significant progress has been made in synthesizing high-quality and visually realistic face images by unconditional Generative Adversarial Networks (GANs), there is still a lack of control over the generation process in order to achieve semantic face editing. In this paper, we propose a novel learning framework, called GuidedStyle, to achieve semantic face editing on pretrained StyleGAN by guiding the image generation process with a knowledge network. Furthermore, we allow an attention mechanism in StyleGAN generator to adaptively select a single layer for style manipulation. As a result, our method is able to perform disentangled and controllable edits along various attributes, including smiling, eyeglasses, gender, mustache, hair color and attractive. Both qualitative and quantitative results demonstrate the superiority of our method over other competing methods for semantic face editing. Moreover, we show that our model can be also applied to different types of real and artistic face editing, demonstrating strong generalization ability.

Asunto(s)

RESUMEN

Microscopic wood identification plays a critical role in many economically important areas in wood science. Historically, producing and curating relevant and representative microscopic cross-section images of wood species is limited to highly experienced and trained anatomists. This manuscript demonstrates the feasibility of generating synthetic microscopic cross-sections of hardwood species. We leveraged a publicly available dataset of 119 hardwood species to train a style-based generative adversarial network (GAN). The proposed GAN generated anatomically accurate cross-section images with remarkable fidelity to actual data. Quantitative metrics corroborated the capacity of the generative model in capturing complex wood structure by resulting in a Fréchet inception distance score of 17.38. Image diversity was calculated using the Structural Similarity Index Measure (SSIM). The SSIM results confirmed that the GAN approach can successfully synthesize diverse images. To confirm the usefulness and realism of the GAN generated images, eight professional wood anatomists in two experience levels participated in a visual Turing test and correctly identified fake and actual images at rates of 48.3 and 43.7%, respectively, with no statistical difference when compared to random guess. The generative model can synthesize realistic, diverse, and meaningful high-resolution microscope cross-section images that are virtually indistinguishable from real images. Furthermore, the framework presented may be suitable for improving current deep learning models, helping understand potential breeding between species, and may be used as an educational tool.

RESUMEN

Deep interactive evolution (DeepIE) combines the capacity of interactive evolutionary computation (IEC) to capture a user's preference with the domain-specific robustness of a trained generative adversarial network (GAN) generator, allowing the user to control the GAN output through evolutionary exploration of the latent space. However, the traditional GAN latent space presents feature entanglement, which limits the practicability of possible applications of DeepIE. In this paper, we implement DeepIE within a style-based generator from a StyleGAN model trained on the WikiArt dataset and propose StyleIE, a variation of DeepIE that takes advantage of the secondary disentangled latent space in the style-based generator. We performed two AB/BA crossover user tests that compared the performance of DeepIE against StyleIE for art generation. Self-rated evaluations of the performance were collected through a questionnaire. Findings from the tests suggest that StyleIE and DeepIE perform equally in tasks with open-ended goals with relaxed constraints, but StyleIE performs better in close-ended and more constrained tasks.

RESUMEN

INTRODUCTION: This paper explores the potential of the StyleGAN model as an high-resolution image generator for synthetic medical images. The possibility to generate sample patient images of different modalities can be helpful for training deep learning algorithms as e.g. a data augmentation technique. METHODS: The StyleGAN model was trained on Computed Tomography (CT) and T2- weighted Magnetic Resonance (MR) images from 100 patients with pelvic malignancies. The resulting model was investigated with regards to three features: Image Modality, Sex, and Longitudinal Slice Position. Further, the style transfer feature of the StyleGAN was used to move images between the modalities. The root-mean-squard error (RMSE) and the Mean Absolute Error (MAE) were used to quantify errors for MR and CT, respectively. RESULTS: We demonstrate how these features can be transformed by manipulating the latent style vectors, and attempt to quantify how the errors change as we move through the latent style space. The best results were achieved by using the style transfer feature of the StyleGAN (58.7 HU MAE for MR to CT and 0.339 RMSE for CT to MR). Slices below and above an initial central slice can be predicted with an error below 75 HU MAE and 0.3 RMSE within 4cm for CT and MR, respectively. DISCUSSION: The StyleGAN is a promising model to use for generating synthetic medical images for MR and CT modalities as well as for 3D volumes.

Asunto(s)