Search

Abstract Accurately predicting the state of health (SOH) of lithium-ion batteries is fundamental in estimating their remaining lifespan. Various parameters such as voltage, current, and temperature significantly influence the battery’s SOH. However, existing data-driven methods necessitate substantial data from the target domain for training, which hampers the assessment of lithium-ion battery health at the initial stage. To address these challenges, this paper introduces the multi-head attention-time convolution network (MHAT-TCN), amalgamating multi-head attention learning with random block dropout techniques. Additionally, it employs grey relational analysis (GRA) to select health indicators (HIs) highly correlated with battery capacity, thereby enhancing the accuracy of the model training. Employing leave-one-out crossvalidation (LOOCV), the MHAT-TCN network is pre-trained using data from batteries of the same model to facilitate comprehensive prediction of the target battery throughout its operational period. Results demonstrate that the MHAT-TCN network trained on HIs outperforms other models, enabling precise predictions across the entire operational period.

Abstract Serious blunt chest trauma usually induces hemothorax, pneumothorax, and rib fractures. More studies have claimed that early video-assisted thoracoscopic surgery with surgical stabilization of rib fractures (SSRF) results in a good prognosis in patients with major trauma. This study aimed to verify the outcomes in patients with chest trauma whether SSRF was performed. Consecutive patients who were treated in a medical center in Taiwan, for traumatic events between January 2015 and June 2020, were retrospectively reviewed. This study focused on patients with major trauma and thoracic injuries, and they were divided into groups based on whether they received SSRF. We used electrical impedance tomography (EIT) to evaluate the change of ventilation conditions. Different scores used for the evaluation of trauma severity were also compared in this study. Among the 8396 patients who were included, 1529 (18.21%) had major trauma with injury severity score > 16 and were admitted to the intensive care unit initially. A total of 596 patients with chest trauma were admitted, of whom 519 (87%) survived. Younger age and a lower trauma score (including injury severity scale, new injury severity score, trauma and injury severity score, and revised trauma score) account for better survival rates. Moreover, 74 patients received SSRF. They had a shorter intensive care unit (ICU) stay (5.24, p = 0.045) and better performance in electrical impedance tomography (23.46, p

Abstract The traditional lecture-based learning (LBL) method is facing great challenges due to its low efficiency and single proceeding form. We designed a PRI-E learning mode that combined and modified problem-based, case-based, and evidence-based learning with a step-by-step approach. We evaluated the practical learning outcomes of using the PRI-E mode by comparing it with traditional lecture-based learning in oral and maxillofacial oncology education. “PRI-E” consists of the first letters of the English words Passion, Research, Innovation, and Education, and it means “the best Education”. This prospective randomized controlled trial included 40 participants. We evenly divided the participants into the PRI-E (n = 20) and LBL group (n = 20) based on the entrance test scores. The same staff group designed and then taught the learning content with different group measures. The evaluation included the final test scores and questionnaire assessments. Without affecting the examination results, the PRI-E teaching method was more satisfactory and popular with participants in terms of ability development and classroom participation. Enacting the PRI-E teaching method required more time, but this did not affect its popularity among the participants. Compared with the LBL learning mode, the PRI-E learning mode was more organized and efficient in oral and maxillofacial oncology education without affecting academic performance. This model has a high degree of satisfaction, which is conducive to training students' comprehensive ability.

Accurately predicting the state of health (SOH) of lithium-ion batteries is fundamental in estimating their remaining lifespan. Various parameters such as voltage, current, and temperature significantly influence the battery's SOH. However, existing data-driven methods necessitate substantial data from the target domain for training, which hampers the assessment of lithium-ion battery health at the initial stage. To address these challenges, this paper introduces the multi-head attention-time convolution network (MHAT-TCN), amalgamating multi-head attention learning with random block dropout techniques. Additionally, it employs grey relational analysis (GRA) to select health indicators (HIs) highly correlated with battery capacity, thereby enhancing the accuracy of the model training. Employing leave-one-out crossvalidation (LOOCV), the MHAT-TCN network is pre-trained using data from batteries of the same model to facilitate comprehensive prediction of the target battery throughout its operational period. Results demonstrate that the MHAT-TCN network trained on HIs outperforms other models, enabling precise predictions across the entire operational period. [ABSTRACT FROM AUTHOR]

The traditional lecture-based learning (LBL) method is facing great challenges due to its low efficiency and single proceeding form. We designed a PRI-E learning mode that combined and modified problem-based, case-based, and evidence-based learning with a step-by-step approach. We evaluated the practical learning outcomes of using the PRI-E mode by comparing it with traditional lecture-based learning in oral and maxillofacial oncology education. "PRI-E" consists of the first letters of the English words Passion, Research, Innovation, and Education, and it means "the best Education". This prospective randomized controlled trial included 40 participants. We evenly divided the participants into the PRI-E (n = 20) and LBL group (n = 20) based on the entrance test scores. The same staff group designed and then taught the learning content with different group measures. The evaluation included the final test scores and questionnaire assessments. Without affecting the examination results, the PRI-E teaching method was more satisfactory and popular with participants in terms of ability development and classroom participation. Enacting the PRI-E teaching method required more time, but this did not affect its popularity among the participants. Compared with the LBL learning mode, the PRI-E learning mode was more organized and efficient in oral and maxillofacial oncology education without affecting academic performance. This model has a high degree of satisfaction, which is conducive to training students' comprehensive ability., (© 2024. The Author(s).)