Third molar impacted teeth are a common issue with all ages, possibly causing tooth decay, root resorption, and pain. This study was aimed at developing a computer-assisted detection system based on deep convolutional neural networks for the detection of third molar impacted teeth using different architectures and to evaluate the potential usefulness and accuracy of the proposed solutions on panoramic radiographs. A total of 440 panoramic radiographs from 300 patients were randomly divided. As a two-stage technique, Faster RCNN with ResNet50, AlexNet, and VGG16 as a backbone and one-stage technique YOLOv3 were used. The Faster-RCNN, as a detector, yielded a mAP@0.5 rate of 0.91 with ResNet50 backbone while VGG16 and AlexNet showed slightly lower performances: 0.87 and 0.86, respectively. The other detector, YOLO v3, provided the highest detection efficacy with a mAP@0.5 of 0.96. Recall and precision were 0.93 and 0.88, respectively, which supported its high performance. Considering the findings from different architectures, it was seen that the proposed one-stage detector YOLOv3 had excellent performance for impacted mandibular third molar tooth detection on panoramic radiographs. Promising results showed that diagnostic tools based on state-ofthe-art deep learning models were reliable and robust for clinical decision-making.
Current standards for safe delivery of electrical stimulation to the central nervous system are based on foundational studies which examined post-mortem tissue for histological signs of damage. This set of observations and the subsequently proposed limits to safe stimulation, termed the “Shannon limits,” allow for a simple calculation (using charge per phase and charge density) to determine the intensity of electrical stimulation that can be delivered safely to brain tissue. In the three decades since the Shannon limits were reported, advances in molecular biology have allowed for more nuanced and detailed approaches to be used to expand current understanding of the physiological effects of stimulation. Here, we demonstrate the use of spatial transcriptomics (ST) in an exploratory investigation to assess the biological response to electrical stimulation in the brain. Electrical stimulation was delivered to the rat visual cortex with either acute or chronic electrode implantation procedures. To explore the influence of device type and stimulation parameters, we used carbon fiber ultramicroelectrode arrays (7 μm diameter) and microwire electrode arrays (50 μm diameter) delivering charge and charge density levels selected above and below reported tissue damage thresholds (range: 2–20 nC, 0.1–1 mC/cm2). Spatial transcriptomics was performed using Visium Spatial Gene Expression Slides (10x Genomics, Pleasanton, CA, United States), which enabled simultaneous immunohistochemistry and ST to directly compare traditional histological metrics to transcriptional profiles within each tissue sample. Our data give a first look at unique spatial patterns of gene expression that are related to cellular processes including inflammation, cell cycle progression, and neuronal plasticity. At the acute timepoint, an increase in inflammatory and plasticity related genes was observed surrounding a stimulating electrode compared to a craniotomy control. At the chronic timepoint, an increase in inflammatory and cell cycle progression related genes was observed both in the stimulating vs. non-stimulating microwire electrode comparison and in the stimulating microwire vs. carbon fiber comparison. Using the spatial aspect of this method as well as the within-sample link to traditional metrics of tissue damage, we demonstrate how these data may be analyzed and used to generate new hypotheses and inform safety standards for stimulation in cortex.
Recently, both the progress in some technological fields and multidisciplinary studies conducted with collaboration from different branches of science have impressive effect on clinically tested systems. Studies on development of visual prosthesis, which is based on passing damaged parts of the visual pathway and electrically stimulating nerve cells remaining intact, date back to elicit visual sense in blind patients. Investigation of some research topics, in silico, which should be taken into consideration in the design phase before expensive animal experiments provides great advantages in terms of both financial and time issues. In this study, factors such as heat, electric field distribution for current thresholds, current density which should be discussed in the design phase are simulated using monophasic rectangle pulses depending on various stimulation parameters with developed computational retina model. Change of heat, electric field, current density in points selected from center and periphery of retina tissue are investigated for various stimulation parameters. As a result, it is concluded that distribution of heat and electric field intensity over the periphery retina are much less than center region. Moreover, when larger pulse width is used, change of heat and electric field intensity seems much more in regions from center retina near stimulation electrode. Current density is higher in the sharp ends of the electrode than flat regions. Besides, when the size of stimulation electrode increases, electric field distribution becomes more uniform.
Objectives: Automatically detecting dental conditions using Artificial intelligence (AI) and reporting it visually are now a need for treatment planning and dental health management. This work presents a comprehensive computer-aided detection system to detect dental restorations. Methods: The state of art ten different Deep Learning detection models was used including R-CNN, Faster R-CNN, SSD, YOLOv3 and RetinaNet as detectors. ResNet-50, ResNet-101, XCeption-101, VGG16 and DarkNet53 were integrated as backbone and feature extractor in addition to efficient approaches such Side-Aware Boundary Localization, cascaded structures and simple model frameworks like Libra and Dynamic. Total 684 objects in panoramic radiographs were used to detect available three classes, namely dental restorations, denture and implant. Each model was evaluated by mean average precision (mAP), average recall (AR) and precision-recall curve using Common Objects in Context (COCO) detection evaluation metrics. Results: mAP varied between 0.755 and 0.973 for ten models explored while AR ranges between 0.605 and 0.771. Faster R-CNN RegnetX provided the best detection performance with mAP of 0.973 and AR of 0.771. Area under precision-recall curve was 0.952. Precision-recall curve indicated that errors were mainly dominated by localization confusions. Conclusions: Results showed that the proposed AI-based computer-aided system had great potential with reliable, accurate performance detecting dental restorations, denture and implant in panoramic radiographs. As training models with more data and standardization in reporting, AI-based solutions will be implemented to dental clinics for daily use soon.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.