A knowledge-interpretable multi-task learning framework for automated thyroid nodule diagnosis in ultrasound videos.

Ultrasound has become the most widely used modality for thyroid nodule diagnosis, due to its portability, real-time feedback, lack of toxicity, and low cost. Recently, the computer-aided diagnosis (CAD) of thyroid nodules has attracted significant attention. However, most existing techniques can only be applied to either static images with prominent features (manually selected from scanning videos) or rely on 'black boxes' that cannot provide interpretable results. In this study, we develop a user-friendly framework for the automated diagnosis of thyroid nodules in ultrasound videos, by simulating the typical diagnostic workflow used by radiologists. This process consists of two orderly part-to-whole tasks. The first interprets the characteristics of each image using prior knowledge, to obtain corresponding frame-wise TI-RADS scores. Associated embedded representations not only provide diagnostic information for radiologists but also reduce computational costs. The second task models temporal contextual information in an embedding vector sequence and selectively enhances important information to distinguish benign and malignant thyroid nodules, thereby improving the efficiency and generalizability of the proposed framework. Experimental results demonstrated this approach outperformed other state-of-the-art video classification methods. In addition to assisting radiologists in understanding model predictions, these CAD results could further ease diagnostic workloads and improve patient care.

CacheTrack-YOLO: Real-Time Detection and Tracking for Thyroid Nodules and Surrounding Tissues in Ultrasound Videos.

To accurately detect and track the thyroid nodules in a video is a crucial step in the thyroid screening for identification of benign and malignant nodules in computer-aided diagnosis (CAD) systems. Most existing methods just perform excellent on static frames selected manually from ultrasound videos. However, manual acquisition is labor-intensive work. To make the thyroid screening process in a more natural way with less labor operations, we develop a well-designed framework suitable for practical applications for thyroid nodule detection in ultrasound videos. Particularly, in order to make full use of the characteristics of thyroid videos, we propose a novel post-processing approach, called Cache-Track, which exploits the contextual relation among video frames to propagate the detection results into adjacent frames to refine the detection results. Additionally, our method can not only detect and count thyroid nodules, but also track and monitor surrounding tissues, which can greatly reduce the labor work and achieve computer-aided diagnosis. Experimental results show that our method performs better in balancing accuracy and speed.

Low-loss coupling from single-mode solid-core fibers to anti-resonant hollow-core fibers by fiber tapering technique.

We demonstrate here for the first time, to the best of our knowledge, an effective method to achieve low-loss light coupling from solid-core fibers to anti-resonant hollow-core fibers (AR-HCFs) by fiber tapering technique. We establish the coupling models by beam propagation method (BPM), and the simulation results show that the coupling efficiency can be optimized by choosing a proper waist diameter of the tapered solid-core fiber. Two types of AR-HCFs have been tested experimentally, and the maximum light coupling efficiency is ∼91.4% at 1.06 µm and ∼90.2% at 1.57 µm for the ice-cream AR-HCF, and ∼83.7% at 1.57 µm for the node-less AR-HCF. This work provides a feasible low-loss light coupling scheme for AR-HCFs, which is very useful for implementing all fiber systems.

The Fabrication of Flow Field Plates for Direct Methanol Fuel Cell Using Lithography and Radio Frequency Sputtering.

This study uses lithography to etch flow fields on a single side of a printed circuit board (PCB) and combines a flow field plate with a collector plate to make innovative anode flow field plates and cathode flow field plates for a direct methanol fuel cell (DMFC). TiO2 thin film is also sputtered on the anode flow field plate using radio frequency (RF) sputtering. The experimental results show that the prepared DMFC has a better maximum power density of 11.928 mW/cm2. Furthermore, when a TiO2 thin film is sputtered on the flow field plate of the assembled DMFC, the maximum power density is 14.426 mW/cm2, which is actually 21% more than that for a DMFC with no TiO2 thin film coated on the flow field plate.