Low-dose computed tomography (LDCT) is critical for minimizing radiation exposure, but it often leads to increased noise and reduced image quality. Traditional denoising methods, such as iterative optimization or supervised learning, often fail to preserve image quality. To address these challenges, we introduce PPORLD-EDNetLDCT, a reinforcement learning-based (RL) approach with Encoder–Decoder for LDCT. Our method utilizes a dynamic RL-based approach in which the Proximal Policy Optimization (PPO) algorithm is employed to optimize the denoising policy during training, guided by image quality feedback in a custom gym environment, while inference is performed using the trained fixed-parameter encoder–decoder model. The experimental results on the low dose CT image and projection dataset demonstrate that the proposed PPORLD-EDNetLDCT model outperforms traditional denoising techniques and other DL-based methods, achieving a peak signal-to-noise ratio of 41.87, a structural similarity index measure of 0.9814 and a root mean squared error of 0.00236. Moreover, in NIH-AAPM-Mayo Clinic Low Dose CT Challenge dataset our method achieved a PSNR of 41.52, SSIM of 0.9723 and RMSE of 0.0051. Furthermore, we validated the quality of denoising using a classification task in the COVID-19 LDCT dataset, where the images processed by our method improved the classification accuracy to 94%, achieving 4% higher accuracy compared to denoising without RL-based denoising.
Estimates of the frequency of hydroclimatic extremes from stream gauging, meteorological observations and satellite imagery are not sufficiently robust for estimation of event frequencies because of the shortness of the records. Accordingly, a 1400-year record of extreme hydroclimatic events, both floods and coast-crossing cyclones (TCs), is presented for the Top End of monsoonal Australia. From this record, event frequencies have been estimated, along with peak flows for some floods, and the relationships of the events to climate states explored. The paleorecords documented in this paper are based on radiometrically dated sedimentary archives from beach ridges and ch & eacute;nier ridges for TCs and mainly from paleoflood deposits for large floods. Large floods in the region are a result of either TCs or intense tropical lows that commonly begin as TCs. This hydrological connection is apparent in episodes of co-occurring TCs and floods, both in the recent past and over the entire 1400-year record. Both the El Ni & ntilde;o-Southern Oscillation (ENSO) and the Interdecadal Pacific Oscillation (IPO) have influenced the occurrence of the extreme events. Flood frequency has varied through time with evidence for long periods with few events. The Modern Warm Period from ca 1850 CE is a time of high-flood and TC frequency, consistent with an increase in high stream flows from an independent study of the Katherine/Daly River and rainfall in monsoonal Australia more widely. The results presented here should be part of flood mitigation strategies, and assessments of the role of climate change.
Developing a positive STEM identity is crucial for motivating students to engage with STEM subjects and pursue related careers. Challenging stereotypes about STEM can enhance student motivation to engage with STEM. Moreover, factors such as prior STEM experiences, STEM-related media engagement, STEM discussions with family and friends, and family support shape a student’s STEM identity. This cross-sectional survey with regression and group comparisons investigated the impact of these factors on three dimensions of STEM identity: interest in STEM, STEM recognition, and STEM self-efficacy. STEM stereotype and identity questionnaires and questions about students’ prior experiences were administered to 268 middle-school students (138 female, 130 males, 123 English as an Additional Language (EAL) from three Australian schools, and responses analysed to explore these influences. STEM interest, recognition and self-efficacy were significantly correlated with engaging with science books and media in primary school for most students. However, there were negative correlations between out-of-school STEM experiences and STEM interest and recognition for non-EAL female students. Trait-based stereotypes had a negative effect on female students’ STEM self-efficacy, while gender stereotypes favouring males had a positive effect on males’ STEM interest. This study contributes to the STEM literature by extending the STEM identity model of Dou and Cian (2022) to include experiential factors. The findings suggest the importance of providing engaging books and media about STEM and designing STEM experiences that address diverse student interests without reinforcing stereotypical beliefs about traits required for success in STEM to develop greater STEM interest, self-efficacy, and recognition among students.
Glass fiber polymer-reinforced (GFRP) composite profiles offer advantages such as corrosion resistance and a favorable strength-to-weight ratio, but their limited ductility and poor fire resistance hinder broader structural use. This study examines the thermal and compressive behavior of carbon fiber-reinforced polymer (CFRP)-confined, geopolymer concrete-filled pultruded GFRP square tubes under elevated temperatures. A total of 90 specimens were prepared using geopolymer concrete with three different compressive strengths (average strengths of 59.8, 68.3, and 89.6 MPa), controlled by sodium hydroxide molarity: 4 M (geopolymer concrete with 4 M sodium hydroxide, denoted as GC4), 8 M (GC8), and 12 M (GC12). Specimens were externally wrapped with CFRP wraps and exposed to temperatures ranging from 25 degrees C to 350 degrees C. Results show that CFRP confinement significantly enhanced compressive capacity, particularly in lower-strength cores, with average strength gains of 87%, 63%, and 27% for GC4, GC8, and GC12 specimens, respectively. Interestingly, elevated temperatures improved strength further, with peak load increases of up to 25% at 350 degrees C. Average ductility indices decreased with increasing concrete strength, ranging from 1.42 (GC4) to 1.30 (GC12). One-way ANOVA revealed that temperature accounted for 82%, 79%, and 63% of variance in compressive capacity for GC4, GC8, and GC12 groups, respectively. These findings highlight the effectiveness of CFRP confinement and the potential of sustainable geopolymer-filled GFRP systems in fire-prone structural applications.
