Bacterial infection risk in work environments has been extensively reported for healthcare workers, while this risk is rarely researched in other occupations. This study aimed to identify occupational environments in Taiwan’s agricultural and healthcare industries with elevated bacterial infection risks by comparing risks for general bacterial infections and pneumonia. Using labour and health insurance claim data from 3.3 million workers (January 2004–December 2020), a retrospective cohort was constructed to estimate occupational infection risks with Cox regression and the Anderson-Gill extension. Significantly elevated hazard ratios were found for workers in vegetable growing, crop cultivation service, mushroom growing, flower growing, and fruit growing, ranging from 1.13 to 1.39 for general bacterial infections and 1.68 to 3.06 for pneumonia infections. In afforestation and the inland fishing industry, pneumonia risk was significantly elevated with, respectively, 1.87 and 1.21. In the healthcare section, especially workers in residential care services and residential care services for elderly stand out regarding their pneumonia risk, with significant hazard ratios of 3.49 and 1.75. The methods used in this study were proven to be effective in identification of occupation environments at risk and can be used in other settings. These findings call for prioritization of bacterial infection prevention by occupation.

Femtosecond oscillators with gigahertz (GHz) repetition rate are appealing sources for spectroscopic applications benefiting from the individually accessible and high-power comb line. The mode mismatch between the potent pump laser diode (LD) and the incredibly small laser cavity, however, limits the average output power of existing GHz Kerr-lens mode-locked (KLM) oscillators to tens of milliwatts. Here, we present a novel method that solves the difficulty and permits high average power LD-pumped KLM oscillators at GHz repetition rate. We propose a numerical simulation method to guide the realization of Kerr-lens mode-locking and comprehend the dynamics of the Kerr-lens mode-locking process. As a proof-of-principle demonstration, an LD-pumped Yb:KGW oscillator with up to 6.17-W average power and 184-fs pulse duration at 1.6-GHz repetition rate is conducted. The simulation had a good agreement with the experimental results. The cost-effective, compact and powerful laser source opens up new possibilities for research and industrial applications.

In most of video coding standards such as high efficiency video coding (HEVC), I-frame assignment is periodic even when the content change is minor, which degrades the coding efficiency. This paper proposes an I-frame assignment method based on Nash bargaining solution (NBS) in game theory to solve this problem. The encoded sequence is divided into several subsequences. Each subsequence is regarded as a game. All group of picture (GOP) in a subsequence is further divided into several sets of GOP. Each set of GOP is regarded as a player and compete for the number of I-frames. The optimal I-frame assignment is determined based on the generalized NBS. Experimental results show the proposed method outperforms HEVC by 5.21% bitrate saving.

This study proposes two multimodal frameworks to classify pathological voice samples by combining acoustic signals and medical records. In the first framework, acoustic signals are transformed into static supervectors via Gaussian mixture models; then, a deep neural network (DNN) combines the supervectors with the medical record and classifies the voice signals. In the second framework, both acoustic features and medical data are processed through first-stage DNNs individually; then, a second-stage DNN combines the outputs of the first-stage DNNs and performs classification. Voice samples were recorded in a specific voice clinic of a tertiary teaching hospital, including three common categories of vocal diseases, i.e. glottic neoplasm, phonotraumatic lesions, and vocal paralysis. Experimental results demonstrated that the proposed framework yields significant accuracy and unweighted average recall (UAR) improvements of 2.02–10.32% and 2.48–17.31%, respectively, compared with systems that use only acoustic signals or medical records. The proposed algorithm also provides higher accuracy and UAR than traditional feature-based and model-based combination methods.

Direct conversion of an amorphous carbon (C) film to capsules by gallium (Ga), and nickel and cobalt (NiCo) alloy particles upon heating is investigated in situ by transmission electron microscopy (TEM). Capsules are catalyzed in an NH3 atmosphere when the temperature is raised to 1050 °C. High resolution TEM reveals that graphene flakes initially nucleate at the surface of the catalysts, then segregate and transform into faceted multi-shell capsules upon continued heating. The solubility of carbon in the NiCo alloy particles can be differentiated from the solubility of carbon in Ga particles by the thickness of the walls. The C/Ga binary phase in nanoparticles is discussed regarding the formation of thin-walled carbon capsules.

A novel T-shaped-gated (T-Gate) polycrystalline silicon thin-film transistor (poly-Si TFT) with vacuum gaps has been proposed and fabricated only with a simple process. The T-Gate structure is formed only by a selective undercut-etching technology of the Mo/Al bi-layers. Then, vacuum gaps are in-situ embedded in this T-Gate structure subsequent to capping the SiH4-based passivation oxide under the vacuum process chamber. Experimental results reveal that the proposed T-Gate poly-Si TFTs have excellent electrical performance, which has higher maximum on-off current ratio of 4.6 e107, and the lower off-state leakage current at VGS = -10 V and VDS = 5V of about 100 times less than that of the conventional one. It is attributed to the additional undoped offset region and the vacuum gap to reduce the maximum electric field at drain junction while ascribed to the sub-gate to maintain the on-current. Therefore, such a T-Gate poly-Si TFT is very suitable for the applications and manufacturing in active matrix liquid crystal displays (AMLCDs) and active matrix organic light emitting diodes (AMOLEDs).