Cervical cancer is a worldwide public health problem with a high rate of illness and mortality among women. In this study, we proposed a novel framework based on Faster RCNN-FPN architecture for the detection of abnormal cervical cells in cytology images from a cancer screening test. We extended the Faster RCNN-FPN model by infusing deformable convolution layers into the feature pyramid network (FPN) to improve scalability. Furthermore, we introduced a global contextual aware module alongside the Region Proposal Network (RPN) to enhance the spatial correlation between the background and the foreground. Extensive experimentations with the proposed deformable and global context aware (DGCA) RCNN were carried out using the cervical image dataset of “Digital Human Body” Vision Challenge from the Alibaba Cloud TianChi Company. Performance evaluation based on the mean average precision (mAP) and receiver operating characteristic (ROC) curve has demonstrated considerable advantages of the proposed framework. Particularly, when combined with tagging of the negative image samples using traditional computer-vision techniques, 6–9% increase in mAP has been achieved. The proposed DGCA-RCNN model has potential to become a clinically useful AI tool for automated detection of cervical cancer cells in whole slide images of Pap smear.
Tunable terahertz absorbers composed of periodically cross-shaped graphene arrays with the ability to achieve near-unity absorbance are proposed and studied. Our results demonstrate that the bandwidth of absorption rate above 90% can reach up to 1.13 terahertz by use of a single layer of cross-shaped graphene arrays. By simply stacking the double layer cross-shaped graphene with careful design, the working bandwidth can be broadened compared with the single-layer graphene-based absorber. The proposed absorbers have the properties of being polarization insensitive and having large angle tolerance, and the tunability of the Fermi level in graphene allows us to realize tunable terahertz absorbers with potential interest in integrated terahertz optoelectronic devices.
In this paper we propose a graphene-based metasurface structure that can exhibit tunable electromagnetically-induced-transparency-like (EIT) spectral response at mid-infrared frequencies. The metasurface structure is composed of two subwavelength mono-layer graphene nano-disks coupled with a mono-layer graphene nano-strip. We show that the coupling of the nano-disks’ dipole resonance with the quadrupole resonance of the nano-strip can create two split resonances with a transparency window in between at any desired center frequency within a wide frequency range. We show that such an EIT-like response can also be dynamically shifted in frequency by adjusting the Fermi-level of the graphene through external voltage control, which provides convenient post-fabrication tunability. In addition, the performance of such a metastructure for sensing the refractive index of the surrounding medium is analyzed. The simulation results show that its sensitivity can reach 3016.7 nm/(RIU) with a FOM exceeding 12.0. Lastly, we present an analysis of the slow light characteristics of the proposed device, where the group index can reach as large as 200. Our design provides a new miniaturized sensing platform that can facilitate the development of biochemical molecules testing, etc.
Spoof surface plasmon polaritons based notch filter for ultra-wideband microwave waveguide is proposed. Owing to subwavelength confinement, such a filter has advantage in the structure size without sacrificing the performance. The spoof SPP based notch is introduced to suppress the WLAN and satellite communication interference simultaneously. Both the cutoff frequency and the notch frequency are sensitive to the structure parameters, and the cutoff frequency can reach 20 GHz. An adiabatic transition relying on gradient hole-size and flaring ground is designed to effectively couple energy into spoof SPP waveguide. The result shows its cutoff frequency of 17.4 GHz with the insertion loss better than 3dB during the whole pass-band, while having more than 20 dB rejections at 5.36GHz and 9.32GHz with 10dB fractional bandwidth 1.07% and 0.74% respectively to avoid the existing WLAN and satellite communication signals. Due to planar structures proposed here, it is easy to integrate in the microwave integrated systems, which can play an important role in the microwave communication circuit and system.
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.