Semi-Supervised Facial Acne Segmentation Using Bidirectional Copy-Paste.

Facial acne is a prevalent dermatological condition regularly observed in the general population. However, it is important to detect acne early as the condition can worsen if not treated. For this purpose, deep-learning-based methods have been proposed to automate detection, but acquiring acne training data is not easy. Therefore, this study proposes a novel deep learning model for facial acne segmentation utilizing a semi-supervised learning method known as bidirectional copy-paste, which synthesizes images by interchanging foreground and background parts between labeled and unlabeled images during the training phase. To overcome the lower performance observed in the labeled image training part compared to the previous methods, a new framework was devised to directly compute the training loss based on labeled images. The effectiveness of the proposed method was evaluated against previous semi-supervised learning methods using images cropped from facial images at acne sites. The proposed method achieved a Dice score of 0.5205 in experiments utilizing only 3% of labels, marking an improvement of 0.0151 to 0.0473 in Dice score over previous methods. The proposed semi-supervised learning approach for facial acne segmentation demonstrated an improvement in performance, offering a novel direction for future acne analysis.

Deep learning-based skin care product recommendation: A focus on cosmetic ingredient analysis and facial skin conditions.

BACKGROUND: Recommendations for cosmetics are gaining popularity, but they are not being made with consideration of the analysis of cosmetic ingredients, which customers consider important when selecting cosmetics. AIMS: This article aims to propose a method for estimating the efficacy of cosmetics based on their ingredients and introduces a system that recommends personalized products for consumers, combined with AI skin analysis. METHODS: We constructed a deep neural network architecture to analyze sequentially arranged cosmetic ingredients in the product and incorporated skin analysis models to get the precise skin status of users from frontal face images. Our recommendation system makes decisions based on the results optimized for the individual. RESULTS: Our cosmetic recommendation system has shown its effectiveness through reliable evaluation metrics, and numerous examples have demonstrated its ability to make reasonable recommendations for various skin problems. CONCLUSION: The result shows that deep learning methods can be used to predict the effects of products based on their cosmetic ingredients and are available for use in personalized cosmetic recommendations.

Facial wrinkle segmentation using weighted deep supervision and semi-automatic labeling.

Facial wrinkles are important indicators of human aging. Recently, a method using deep learning and a semi-automatic labeling was proposed to segment facial wrinkles, which showed much better performance than conventional image-processing-based methods. However, the difficulty of wrinkle segmentation remains challenging due to the thinness of wrinkles and their small proportion in the entire image. Therefore, performance improvement in wrinkle segmentation is still necessary. To address this issue, we propose a novel loss function that takes into account the thickness of wrinkles based on the semi-automatic labeling approach. First, considering the different spatial dimensions of the decoder in the U-Net architecture, we generated weighted wrinkle maps from ground truth. These weighted wrinkle maps were used to calculate the training losses more accurately than the existing deep supervision approach. This new loss computation approach is defined as weighted deep supervision in our study. The proposed method was evaluated using an image dataset obtained from a professional skin analysis device and labeled using semi-automatic labeling. In our experiment, the proposed weighted deep supervision showed higher Jaccard Similarity Index (JSI) performance for wrinkle segmentation compared to conventional deep supervision and traditional image processing methods. Additionally, we conducted experiments on the labeling using a semi-automatic labeling approach, which had not been explored in previous research, and compared it with human labeling. The semi-automatic labeling technology showed more consistent wrinkle labels than human-made labels. Furthermore, to assess the scalability of the proposed method to other domains, we applied it to retinal vessel segmentation. The results demonstrated superior performance of the proposed method compared to existing retinal vessel segmentation approaches. In conclusion, the proposed method offers high performance and can be easily applied to various biomedical domains and U-Net-based architectures. Therefore, the proposed approach will be beneficial for various biomedical imaging approaches. To facilitate this, we have made the source code of the proposed method publicly available at: https://github.com/resemin/WeightedDeepSupervision.

Deep-Learning-Based Morphological Feature Segmentation for Facial Skin Image Analysis.

Facial skin analysis has attracted considerable attention in the skin health domain. The results of facial skin analysis can be used to provide skin care and cosmetic recommendations in aesthetic dermatology. Because of the existence of several skin features, grouping similar features and processing them together can improve skin analysis. In this study, a deep-learning-based method of simultaneous segmentation of wrinkles and pores is proposed. Unlike color-based skin analysis, this method is based on the analysis of the morphological structures of the skin. Although multiclass segmentation is widely used in computer vision, this segmentation was first used in facial skin analysis. The architecture of the model is U-Net, which has an encoder-decoder structure. We added two types of attention schemes to the network to focus on important areas. Attention in deep learning refers to the process by which a neural network focuses on specific parts of its input to improve its performance. Second, a method to enhance the learning capability of positional information is added to the network based on the fact that the locations of wrinkles and pores are fixed. Finally, a novel ground truth generation scheme suitable for the resolution of each skin feature (wrinkle and pore) was proposed. The experimental results revealed that the proposed unified method achieved excellent localization of wrinkles and pores and outperformed both conventional image-processing-based approaches and one of the recent successful deep-learning-based approaches. The proposed method should be expanded to applications such as age estimation and the prediction of potential diseases.

CLIMAR: classified linear interpolation based metal artifact reduction for severe metal artifact reduction in x-ray CT imaging.

In x-ray CT imaging, the existence of metal in the imaging field of view deteriorates the quality of the reconstructed image. This is because rays penetrating dense metal implants are highly corrupted, causing huge inconsistency between projection data. The result appears as strong artifacts such as black and white streaks on the reconstructed image disturbing correct diagnosis. For several decades, there have been various trials to reduce metal artifacts for better image quality. As the computing power of computer processors became more powerful, more complex algorithms with improved performance have been introduced. For instance, the initially developed metal artifact reduction (MAR) algorithms based on simple sinogram interpolation were combined with computationally expensive iterative reconstruction techniques to pursue better image quality. Recently, even machine learning based techniques have been introduced, which require huge amounts of computations for training. In this paper, we introduce an image based novel MAR algorithm in which severe metal artifacts such as black shadings are detected by the proposed method in a straightforward manner based on a linear interpolation. To do that, a new concept of metal artifact classification is devised using linear interpolation in the virtual projection domain. The proposed method reduces severe artifacts very quickly and effectively and has good performance to keep the detailed body structure preserved. Results of qualitative and quantitative comparisons with other representative algorithms such as LIMAR and NMAR support the excellence of the proposed algorithm. Thanks to the nature of reducing artifacts in the image itself and its low computational cost, the proposed algorithm can function as an initial image generator for other MAR algorithms, as well as being integrated in the modalities under limited computation power such as mobile CT scanners.

Motion adaptive patch-based low-rank approach for compressed sensing cardiac cine MRI.

One of the technical challenges in cine magnetic resonance imaging (MRI) is to reduce the acquisition time to enable the high spatio-temporal resolution imaging of a cardiac volume within a short scan time. Recently, compressed sensing approaches have been investigated extensively for highly accelerated cine MRI by exploiting transform domain sparsity using linear transforms such as wavelets, and Fourier. However, in cardiac cine imaging, the cardiac volume changes significantly between frames, and there often exist abrupt pixel value changes along time. In order to effectively sparsify such temporal variations, it is necessary to exploit temporal redundancy along motion trajectories. This paper introduces a novel patch-based reconstruction method to exploit geometric similarities in the spatio-temporal domain. In particular, we use a low rank constraint for similar patches along motion, based on the observation that rank structures are relatively less sensitive to global intensity changes, but make it easier to capture moving edges. A Nash equilibrium formulation with relaxation is employed to guarantee convergence. Experimental results show that the proposed algorithm clearly reconstructs important anatomical structures in cardiac cine image and provides improved image quality compared to existing state-of-the-art methods such as k-t FOCUSS, k-t SLR, and MASTeR.