ABSTRACT
Background: Low-dose computed tomography (LDCT) is a diagnostic imaging technique designed to minimize radiation exposure to the patient. However, this reduction in radiation may compromise computed tomography (CT) image quality, adversely impacting clinical diagnoses. Various advanced LDCT methods have emerged to mitigate this challenge, relying on well-matched LDCT and normal-dose CT (NDCT) image pairs for training. Nevertheless, these methods often face difficulties in distinguishing image details from nonuniformly distributed noise, limiting their denoising efficacy. Additionally, acquiring suitably paired datasets in the medical domain poses challenges, further constraining their applicability. Hence, the objective of this study was to develop an innovative denoising framework for LDCT images employing unpaired data. Methods: In this paper, we propose a LDCT denoising network (DNCNN) that alleviates the need for aligning LDCT and NDCT images. Our approach employs generative adversarial networks (GANs) to learn and model the noise present in LDCT images, establishing a mapping from the pseudo-LDCT to the actual NDCT domain without the need for paired CT images. Results: Within the domain of weakly supervised methods, our proposed model exhibited superior objective metrics on the simulated dataset when compared to CycleGAN and selective kernel-based cycle-consistent GAN (SKFCycleGAN): the peak signal-to-noise ratio (PSNR) was 43.9441, the structural similarity index measure (SSIM) was 0.9660, and the visual information fidelity (VIF) was 0.7707. In the clinical dataset, we conducted a visual effect analysis by observing various tissues through different observation windows. Our proposed method achieved a no-reference structural sharpness (NRSS) value of 0.6171, which was closest to that of the NDCT images (NRSS =0.6049), demonstrating its superiority over other denoising techniques in preserving details, maintaining structural integrity, and enhancing edge contrast. Conclusions: Through extensive experiments on both simulated and clinical datasets, we demonstrated the superior efficacy of our proposed method in terms of denoising quality and quantity. Our method exhibits superiority over both supervised techniques, including block-matching and 3D filtering (BM3D), residual encoder-decoder convolutional neural network (RED-CNN), and Wasserstein generative adversarial network-VGG (WGAN-VGG), and over weakly supervised approaches, including CycleGAN and SKFCycleGAN.
ABSTRACT
In recent years, score-based diffusion models have emerged as effective tools for estimating score functions from empirical data distributions, particularly in integrating implicit priors with inverse problems like CT reconstruction. However, score-based diffusion models are rarely explored in challenging tasks such as metal artifact reduction (MAR). In this paper, we introduce the BiConstraints Diffusion Model for Metal Artifact Reduction (BCDMAR), an innovative approach that enhances iterative reconstruction with a conditional diffusion model for MAR. This method employs a metal artifact degradation operator in place of the traditional metal-excluded projection operator in the data-fidelity term, thereby preserving structure details around metal regions. However, scorebased diffusion models tend to be susceptible to grayscale shifts and unreliable structures, making it challenging to reach an optimal solution. To address this, we utilize a precorrected image as a prior constraint, guiding the generation of the score-based diffusion model. By iteratively applying the score-based diffusion model and the data-fidelity step in each sampling iteration, BCDMAR effectively maintains reliable tissue representation around metal regions and produces highly consistent structures in non-metal regions. Through extensive experiments focused on metal artifact reduction tasks, BCDMAR demonstrates superior performance over other state-of-the-art unsupervised and supervised methods, both quantitatively and in terms of visual results.
ABSTRACT
Digital complete denture fabrication has advanced with the integration of computer-aided design and computer-aided manufacturing (CAD-CAM), intraoral scanning, and 3-dimensional printing. A method of fabricating a stackable CAD-CAM custom record tray for complete dentures is introduced. The method combines a custom tray, record base, and occlusion rim in a single piece.
ABSTRACT
Computed tomography (CT) is widely used in clinical medicine, and low-dose CT (LDCT) has become popular to reduce potential patient harm during CT acquisition. However, LDCT aggravates the problem of noise and artifacts in CT images, increasing diagnosis difficulty. Through deep learning, denoising CT images by artificial neural network has aroused great interest for medical imaging and has been hugely successful. We propose a framework to achieve excellent LDCT noise reduction using independent operation search cells, inspired by neural architecture search, and introduce the Laplacian to further improve image quality. Employing patch-based training, the proposed method can effectively eliminate CT image noise while retaining the original structures and details, hence significantly improving diagnosis efficiency and promoting LDCT clinical applications.
Subject(s)
Neural Networks, Computer , Tomography, X-Ray Computed , Humans , Signal-To-Noise Ratio , Radiation Dosage , Tomography, X-Ray Computed/methods , ArtifactsABSTRACT
In this work, a novel semisupervised framework is proposed to tackle the small-sample problem of dental-based human identification (DHI), achieving enhanced performance via a "classifying while generating" paradigm. A generative adversarial network (GAN), called the DHI-GAN, is presented to implement this idea, in which an extra classifier is also dedicatedly proposed to achieve an efficient training procedure. Considering the complex specificities of this problem, except for the noise input of the generator, an identity embedding-guided architecture is proposed to retain informative features for each individual. A parallel spatial and channel fusion attention block is innovatively designed to encourage the model to learn discriminative and informative features by focusing on different regional details and abstract concepts. The attention block is also widely applied to the overall classifier to learn identity-dependent information. A loss combination of the ArcFace and focal loss is utilized to address the small-sample problem. Two parameters are proposed to control the generated samples that are fed into the classifier during the optimization procedure. The proposed DHI-GAN framework is finally validated on a real-world dataset, and the experimental results demonstrate that it outperforms other baselines, achieving a 92.5% top-one accuracy rate. Most importantly, the proposed GAN-based semisupervised training strategy is able to reduce the required number of training samples (individuals) and can also be incorporated into other classification models. Our code will be available at https://github.com/sculyi/MedicalImages/.
Subject(s)
Forensic Anthropology , Neural Networks, Computer , Humans , Learning , Concept FormationABSTRACT
When accidents occur, panoramic dental images play a significant role in identifying unknown bodies. In recent years, deep neural networks have been applied to address this task. However, while tooth contours are significant in classical methods, few studies using deep learning methods devise an architecture specifically to introduce tooth contours into their models. Since fine-grained image identification aims to distinguish subordinate categories by specific parts, we devise a fine-grained human identification model that leverages the distribution of tooth masks to distinguish different individuals with local and subtle differences in their teeth. First, a bilateral branched architecture is designed, of which one branch was designed as the image feature extractor, while the other was the mask feature extractor. In this step, the mask feature interacts with the extracted image feature to perform elementwise reweighting. Additionally, an improved attention mechanism was used to make our model concentrate more on informative positions. Furthermore, we improved the ArcFace loss by adding a learnable parameter to increase the loss of those hard samples, thereby exploiting the potential of our loss function. Our model was tested on a large dataset consisting of 23,715 panoramic X-ray dental images with tooth masks from 10,113 patients, achieving an average rank-1 accuracy of 88.62% and rank-10 accuracy of 96.16%.
ABSTRACT
INTRODUCTION: The purpose of this in vitro study was to evaluate the accuracy and precision of desktop 3D printers when fabricating stents for guided endodontics. METHODS: A stent was designed using planning software for guided endodontic access on a typodont model. Four different 3D printers were used to fabricate an identical stent, one per printer. Each stent was then used to gain access to the artificial endodontic canal on a typodont tooth and was repeated 10 times per stent by the same operator. Each of the accessed typodont teeth were scanned by a reference scanner and then imported into the inspection software. Inspection software used a best-fit alignment to automatically calculate absolute deviation at the base and tip of the bur. RESULTS: The mean distances between the planned and actual positions of the bur were low, ranging from 0.31 to 0.68 mm. Statistically significant differences were found among the 4 groups (F3,36 = 10.67, P < .05). Post hoc comparison revealed that Group Form2 significantly varied from Groups Form3 and Carbon (P < .05 and P < .05, respectively). Group Form3 obtained the most accurate and most precise axial deviations both coronally and apically. CONCLUSIONS: All of the printers tested produced stents for guided access that allowed for a high level of accuracy in obtaining access to the artificial endodontic canal, which would justify the trial of cost-effective 3D printers for guided endodontic access and necessitates further clinical research on teeth with pulp canal obliteration.
Subject(s)
Dental Cavity Preparation , Dental Pulp Cavity , Printing, Three-Dimensional , Cone-Beam Computed Tomography , Cost-Benefit Analysis , Dental Cavity Preparation/economics , Dental Cavity Preparation/methods , Dental Prosthesis Design , Dental Pulp Cavity/surgery , Endodontics/economics , Printing, Three-Dimensional/economics , Software , StentsABSTRACT
Human identification plays a significant role in the investigations of disasters and criminal cases. Human identification could be achieved quickly and efficiently via 3D sphenoid sinus models by customized convolutional neural networks. In this retrospective study, a deep learning neural network was proposed to achieve human identification of 1475 noncontrast thin-slice CT scans. A total of 732 patients were retrieved and studied (82% for model training and 18% for testing). By establishing an individual recognition framework, the anonymous sphenoid sinus model was matched and cross-tested, and the performance of the framework also was evaluated on the test set using the recognition rate, ROC curve and identification speed. Finally, manual matching was performed based on the framework results in the test set. Out of a total of 732 subjects (mean age 46.45 years ± 14.92 (SD); 349 women), 600 subjects were trained, and 132 subjects were tested. The present automatic human identification has achieved Rank 1 and Rank 5 accuracy values of 93.94% and 99.24%, respectively, in the test set. In addition, all the identifications were completed within 55 s, which manifested the inference speed of the test set. We used the comparison results of the MVSS-Net to exclude sphenoid sinus models with low similarity and carried out traditional visual comparisons of the CT anatomical aspects of the sphenoid sinus of 132 individuals with an accuracy of 100%. The customized deep learning framework achieves reliable and fast human identification based on a 3D sphenoid sinus and can assist forensic radiologists in human identification accuracy.
Subject(s)
Deep Learning , Sphenoid Sinus , Female , Forensic Anthropology , Humans , Middle Aged , Retrospective Studies , Skull , Sphenoid Sinus/diagnostic imagingABSTRACT
Computed tomography (CT) has become a powerful tool for medical diagnosis. However, minimizing X-ray radiation risk for the patient poses significant challenges to obtain suitable low dose CT images. Although various low-dose CT methods using deep learning techniques have produced impressive results, convolutional neural network based methods focus more on local information and hence are very limited for non-local information extraction. This paper proposes ERA-WGAT, a residual autoencoder incorporating an edge enhancement module that performs convolution with eight types of learnable operators providing rich edge information and a window-based graph attention convolutional network that combines static and dynamic attention modules to explore non-local self-similarity. We use the compound loss function that combines MSE loss and multi-scale perceptual loss to mitigate the over-smoothing problem. Compared with current low-dose CT denoising methods, ERA-WGAT confirmed superior noise suppression and perceived image quality.
ABSTRACT
Forensic odontology is regarded as an important branch of forensics dealing with human identification based on dental identification. This paper proposes a novel method that uses deep convolution neural networks to assist in human identification by automatically and accurately matching 2-D panoramic dental X-ray images. Designed as a top-down architecture, the network incorporates an improved channel attention module and a learnable connected module to better extract features for matching. By integrating associated features among all channel maps, the channel attention module can selectively emphasize interdependent channel information, which contributes to more precise recognition results. The learnable connected module not only connects different layers in a feed-forward fashion but also searches the optimal connections for each connected layer, resulting in automatically and adaptively learning the connections among layers. Extensive experiments demonstrate that our method can achieve new state-of-the-art performance in human identification using dental images. Specifically, the method is tested on a dataset including 1,168 dental panoramic images of 503 different subjects, and its dental image recognition accuracy for human identification reaches 87.21% rank-1 accuracy and 95.34% rank-5 accuracy. Code has been released on Github. (https://github.com/cclaiyc/TIdentify).
Subject(s)
Forensic Anthropology , Neural Networks, Computer , HumansABSTRACT
Human identification is an important task in mass disaster and criminal investigations. Although several automatic dental identification systems have been proposed, accurate and fast identification from panoramic dental radiographs (PDRs) remains a challenging issue. In this study, an automatic human identification system (DENT-net) was developed using the customized convolutional neural network (CNN). The DENT-net was trained on 15,369 PDRs from 6300 individuals. The PDRs were preprocessed by affine transformation and histogram equalization. The DENT-net took 128 × 128 × 7 square patches as input, including the whole PDR and six details extracted from the PDR. Using the DENT-net, the feature extraction took around 10 milliseconds per image and the running time for retrieval was 33.03 milliseconds in a 2000-individual database, promising an application on larger databases. The visualization of CNN showed that the teeth, maxilla, and mandible all contributed to human identification. The DENT-net achieved Rank-1 accuracy of 85.16% and Rank-5 accuracy of 97.74% for human identification. The present results demonstrated that human identification can be achieved from PDRs by CNN with high accuracy and speed. The present system can be used without any special equipment or knowledge to generate the candidate images. While the final decision should be made by human specialists in practice. It is expected to aid human identification in mass disaster and criminal investigation.
Subject(s)
Electronic Data Processing , Forensic Dentistry/methods , Neural Networks, Computer , Radiography, Panoramic , Datasets as Topic , Humans , Image Processing, Computer-Assisted/methodsABSTRACT
STATEMENT OF PROBLEM: Estimating the width of the maxillary anterior teeth when creating an esthetic smile can be challenging. Valid metrics to assist in this process are needed. PURPOSE: The purpose of this systematic review was to evaluate the validity of interalar distance and inner canthal distance with the golden proportion, golden mean, and recurring esthetic dental proportion in predicting intercanine distance and the combined width of central incisors to potentially provide a guide for tooth restoration. MATERIAL AND METHODS: A literature search was conducted using PubMed, Medline, Google Scholar, EMBASE, CNKI, Web of Science, and the Cochrane Collaboration, identifying English- and non-English-language articles reporting on interalar width, inner canthal width, and maxillary anterior tooth width. Additional studies were identified by searching reference lists of the articles identified. Only studies that fulfilled inclusion criteria were included. Two examiners independently performed the literature search and data extraction. Using a meta-analysis software program, data extracted from each selected study were statistically combined using the random-effects model. Weighted mean differences, 95% confidence intervals, and heterogeneity were calculated for each measurement. RESULTS: The search strategy resulted in a total of 282 articles, but only 41 articles fulfilling the inclusion criteria were included in the meta-analysis. The interalar distance was found to be significantly larger than intercanine distance, and the inner canthal distance was found to be substantially smaller than the intercanine distance. When predicting the central incisors combined width by interalar distance, both the golden proportion and golden mean predicted value were larger than the combined width of the central incisors. Only the recurring esthetic dental proportion (70%) predicted value showed no significant difference from the combined width of central incisors. When predicting the central incisors combined width by inner canthal distance, the golden proportion predicted value was larger than the combined width of central incisors, whereas both the golden mean and recurring esthetic dental proportion (70%) predicted value were found to be significantly smaller than the combined width of central incisors. CONCLUSIONS: By analyzing the data from the literature, only the recurring esthetic dental proportion (70%) with interalar distance could be an accurate method for predicting the combined width of central incisors. Neither interalar distance nor inner canthal distance could directly be used to predict the intercanine distance.
Subject(s)
Cuspid , Incisor , Esthetics, Dental , Maxilla , OdontometryABSTRACT
Compressive sensing (CS) has proved effective for tomographic reconstruction from sparsely collected data or under-sampled measurements, which are practically important for few-view computed tomography (CT), tomosynthesis, interior tomography, and so on. To perform sparse-data CT, the iterative reconstruction commonly uses regularizers in the CS framework. Currently, how to choose the parameters adaptively for regularization is a major open problem. In this paper, inspired by the idea of machine learning especially deep learning, we unfold the state-of-the-art "fields of experts"-based iterative reconstruction scheme up to a number of iterations for data-driven training, construct a learned experts' assessment-based reconstruction network (LEARN) for sparse-data CT, and demonstrate the feasibility and merits of our LEARN network. The experimental results with our proposed LEARN network produces a superior performance with the well-known Mayo Clinic low-dose challenge data set relative to the several state-of-the-art methods, in terms of artifact reduction, feature preservation, and computational speed. This is consistent to our insight that because all the regularization terms and parameters used in the iterative reconstruction are now learned from the training data, our LEARN network utilizes application-oriented knowledge more effectively and recovers underlying images more favorably than competing algorithms. Also, the number of layers in the LEARN network is only 50, reducing the computational complexity of typical iterative algorithms by orders of magnitude.
Subject(s)
Deep Learning , Image Processing, Computer-Assisted/methods , Tomography, X-Ray Computed/methods , Algorithms , Databases, Factual , Humans , Radiography, Abdominal , Radiography, ThoracicABSTRACT
Since absorption of X-ray radiation has the possibility of inducing cancerous, genetic and other diseases to patients, researches usually attempt to reduce the radiation dose. However, reduction of the radiation dose associated with CT scans will unavoidably increase the severity of noise and artifacts, which can seriously affect diagnostic confidence. Due to the outstanding performance of deep neural networks in image processing, in this paper, we proposed a Stacked Competitive Network (SCN) approach to noise reduction, which stacks several successive Competitive Blocks (CB). The carefully handcrafted design of the competitive blocks was inspired by the idea of multi-scale processing and improvement the network's capacity. Qualitative and quantitative evaluations demonstrate the competitive performance of the proposed method in noise suppression, structural preservation, and lesion detection.
Subject(s)
Signal-To-Noise Ratio , Tomography, X-Ray Computed/methods , Dose-Response Relationship, Radiation , Humans , Image Processing, Computer-Assisted , Neural Networks, ComputerABSTRACT
Given the potential risk of X-ray radiation to the patient, low-dose CT has attracted a considerable interest in the medical imaging field. Currently, the main stream low-dose CT methods include vendor-specific sinogram domain filtration and iterative reconstruction algorithms, but they need to access raw data, whose formats are not transparent to most users. Due to the difficulty of modeling the statistical characteristics in the image domain, the existing methods for directly processing reconstructed images cannot eliminate image noise very well while keeping structural details. Inspired by the idea of deep learning, here we combine the autoencoder, deconvolution network, and shortcut connections into the residual encoder-decoder convolutional neural network (RED-CNN) for low-dose CT imaging. After patch-based training, the proposed RED-CNN achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases. Especially, our method has been favorably evaluated in terms of noise suppression, structural preservation, and lesion detection.
Subject(s)
Image Processing, Computer-Assisted/methods , Neural Networks, Computer , Tomography, X-Ray Computed/methods , Abdomen/diagnostic imaging , Algorithms , Computer Simulation , Humans , Liver Neoplasms/diagnostic imagingABSTRACT
In order to reduce the potential radiation risk, low-dose CT has attracted an increasing attention. However, simply lowering the radiation dose will significantly degrade the image quality. In this paper, we propose a new noise reduction method for low-dose CT via deep learning without accessing original projection data. A deep convolutional neural network is here used to map low-dose CT images towards its corresponding normal-dose counterparts in a patch-by-patch fashion. Qualitative results demonstrate a great potential of the proposed method on artifact reduction and structure preservation. In terms of the quantitative metrics, the proposed method has showed a substantial improvement on PSNR, RMSE and SSIM than the competing state-of-art methods. Furthermore, the speed of our method is one order of magnitude faster than the iterative reconstruction and patch-based image denoising methods.
ABSTRACT
Chronic periodontitis (CPs) could result in damage of periodontal tissues, loss of teeth and impose troublesome hindrance to restore teeth satisfyingly as well. Functional gene polymorphisms of matrix metalloproteinases, cytokines and cyclooxygenase-2 have been found to play important roles in periodontitis. This study was to investigate the association between MMP-1-1067, MMP-3-1171, MMP-9-1562, IL-2-330, IL-8-251, COX-2-765 polymorphisms, and the susceptibility to CP in a Chinese population. A total of 122 patients with CP were evaluated for MMP-1, MMP-3, MMP-9, IL-2, IL-8 and COX-2 genetic polymorphisms and were compared with 532 healthy control subjects using PCR-RFLP analysis. Clinical periodontal parameters were recorded. Serum levels of MMP-1, MMP-3, MMP-9, IL-2, IL-8 and COX-2 were measured by ELISA. The data were analyzed by chi-square, logistic regression and Mann-Whitney-U-tests and t test. There were significant differences between CP patients and healthy subjects in the genotype distribution and allele frequency of MMP-3-1171, MMP-9-1562, IL-2-330, IL-8-251 and COX-2-765 genetic polymorphisms. Significant difference between patients and controls were also observed for MMP-1-1067 genotype frequency, but not for allele frequency. Differences between rare allele carriage rates of CP and healthy groups regarding all the genetic polymorphisms in our study were significant (p<0.05). Serum levels of all the cytokines were higher in the CP patients compared to healthy subjects. These data show that MMP-1-1067, MMP-3-1171, MMP-9-1562 and IL-8-251 polymorphisms are associated with susceptibility to CP. MMP-1-1067 2G, MMP-3-1171 6A, MMP-9-1562 T and IL-8-251 A allele are associated with decreased susceptibility to CP in Chinese population.