The goal of this paper is to develop a new skin imaging modality which addresses the current clinical need for a non-invasive imaging tool that images the skin over its depth with high resolutions while offering large histopathological-like contrasts between malignant and normal tissues. We demonstrate that by taking advantage of the intrinsic millimeter-wave dielectric contrasts between normal and malignant skin tissues, ultra-high-resolution millimeter-wave imaging (MMWI) can achieve 3-D, high-contrast images of the skin. In this paper, an imaging system with a record-wide bandwidth of 98 GHz is developed using the synthetic ultra-wideband millimeter-wave imaging approach, a new ultra-high-resolution imaging technique recently developed by the authors. The 21 non-melanoma skin cancer (NMSC) specimens are imaged and compared with histopathology for evaluation. A programmable measurement platform is designed to automatically scan the tissues across a rectangular aperture plane. Furthermore, a novel frequency-domain imaging algorithm is developed to process the recorded signals and generate an image of the cancerous tissue. The high correlations achieved between MMWI images and histological images allow for rapid and accurate delineation of NMSC tissues. The millimeter-wave reflectivity values are also found to be statistically significant higher for cancerous areas with respect to normal areas. Since MMWI does not require tissue processing or staining, it can be performed promptly, enabling diagnosis of tumors at an early stage as well as simplify the tumor removal surgery to a single-layer excision procedure.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1109/TMI.2019.2902600 | DOI Listing |
In this paper, we report a three-dimensional synthetic aperture imaging method with pulsed terahertz waves realized by a terahertz time-domain spectrometer. In contrast to synthetic aperture imaging systems operating at microwave or millimeter-wave frequencies where the frequency of the transmitter is scanned in the frequency domain, in our imaging system, all the frequency components are contained in a single terahertz pulse that can be generated and detected by photoconductive antennas. The image algorithm was analyzed theoretically and confirmed numerically using the finite-difference time-domain method.
View Article and Find Full Text PDFMetasurface holography has become a surging and revolutionized field due to its flexible manipulation of amplitude and/or phase, which enhances the quality and capacity of holographic images. However, the current meta-holograms primarily focus on half-space manipulation, posing a challenge in developing simplified meta-hologram structures for spatial multiplexing. To address this situation, what we believe to be a novel 4-bit "Janus" metasurface combined with the weighted Gerchberg-Saxton (WGS) algorithm is proposed to record and reconstruct two distinct images in millimeter wave band.
View Article and Find Full Text PDFNeural Netw
January 2025
School of Automation Engineering, University of Electronic Science and Technology of China, Xiyuan street 2006, Chengdu, 611731, Sichuan, China.
To facilitate penetrating-imaging oriented applications such as nondestructive internal defect detection and localization under obstructed environment, a novel pixel-level information fusion strategy for mmWave and visible images is proposed. More concretely, inspired by both the advancement of deep learning on universal image fusion and the maturity of near-field millimeter wave imaging technology, an effective deep transfer learning strategy is presented to capture the information hidden in visible and millimeter wave images. Furthermore, by implementing fine-tuning strategy and by using an improved bilateral filter, the proposed fusion strategy can robustly exploit the information in both the near-field millimeter wave field and the visual light field.
View Article and Find Full Text PDFSci Rep
October 2024
Guangxi Road and Bridge Engineering Group Co., LTD, Nanning, China.
Sensors (Basel)
July 2024
Center of Digital Innovation, Tongji University, Shanghai 200092, China.
Robust object detection in complex environments, poor visual conditions, and open scenarios presents significant technical challenges in autonomous driving. These challenges necessitate the development of advanced fusion methods for millimeter-wave (mmWave) radar point cloud data and visual images. To address these issues, this paper proposes a radar-camera robust fusion network (RCRFNet), which leverages self-supervised learning and open-set recognition to effectively utilise the complementary information from both sensors.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!