Deep Learning Method for Precise Landmark Identification and Structural Assessment of Whole-Spine Radiographs.

This study measured parameters automatically by marking the point for measuring each parameter on whole-spine radiographs. Between January 2020 and December 2021, 1017 sequential lateral whole-spine radiographs were retrospectively obtained. Of these, 819 and 198 were used for training and testing the performance of the landmark detection model, respectively. To objectively evaluate the program's performance, 690 whole-spine radiographs from four other institutions were used for external validation. The combined dataset comprised radiographs from 857 female and 850 male patients (average age 42.2 ± 27.3 years; range 20-85 years). The landmark localizer showed the highest accuracy in identifying cervical landmarks (median error 1.5-2.4 mm), followed by lumbosacral landmarks (median error 2.1-3.0 mm). However, thoracic landmarks displayed larger localization errors (median 2.4-4.3 mm), indicating slightly reduced precision compared with the cervical and lumbosacral regions. The agreement between the deep learning model and two experts was good to excellent, with intraclass correlation coefficient values >0.88. The deep learning model also performed well on the external validation set. There were no statistical differences between datasets in all parameters, suggesting that the performance of the artificial intelligence model created was excellent. The proposed automatic alignment analysis system identified anatomical landmarks and positions of the spine with high precision and generated various radiograph imaging parameters that had a good correlation with manual measurements.

Utilizing VO2 as a Hole Injection Layer for Efficient Charge Injection in Quantum Dot Light-Emitting Diodes Enables High Device Performance.

Quantum dot light-emitting diodes (QLEDs) are promising devices for display applications. Polyethylenedioxythiophene:polystyrene sulfonate (PEDOT:PSS) is a common hole injection layer (HIL) material in optoelectronic devices because of its high conductivity and high work function. Nevertheless, PEDOT:PSS-based QLEDs have a high energy barrier for hole injection, which results in low device efficiency. Therefore, a new strategy is needed to improve the device efficiency. Herein, we have demonstrated a bilayer-HIL using VO2 and a PEDOT:PSS-based QLED that exhibits an 18% external quantum efficiency (EQE), 78 cd/A current efficiency (CE), and 25,771 cd/m2 maximum luminance. In contrast, the PEDOT:PSS-based QLED exhibits an EQE of 13%, CE of 54 cd/A, and maximum luminance of 14,817 cd/m2. An increase in EQE was attributed to a reduction in the energy barrier between indium tin oxide (ITO) and PEDOT:PSS, caused by the insertion of a VO2 HIL. Therefore, our results could demonstrate that using a bilayer-HIL is effective in increasing the EQE in QLEDs.

Synthesis and Densification of Nano-Sized W Powders Prepared by Hydrogen Reduction of Ball-Milled WO3 Powders.

The synthesis and consolidation of nano-sized W powders are attempted with the combination process of hydrogen reduction of ball-milled WO3 powder and spark plasma sintering. The reduction behavior of WO3 is analyzed by temperature-programmed reduction. The reaction peaks for reduction of WO3 are observed in the temperature range of 590-782 °C. XRD and TEM analysis reveals that oxide powder is changed to metallic W with an average particle size of 100 nm by hydrogen reduction at 900 °C for 1 h. The densified specimen by spark plasma sintering at 1700 °C under an applied pressure of 50 MPa using nano-sized W powder shows increased relative density compared with that using micron-sized W powder. The results suggested that the W bulk with increased relative density fine microstructure can be fabricated by spark plasma sintering of hydrogen-reduced WO3 powder, more effectively.