Search

In nuclear power plants, the loss-of-coolant accident (LOCA) stands out as the most prevalent and consequential incident. Accurate breach size diagnosis is crucial for the mitigation of LOCAs, and identifying the cause of an accident can prevent catastrophic consequences. Traditional methods mostly focus on combining model algorithms and utilize intricate composite model neural network architectures. However, it is crucial to investigate whether greater complexity necessarily leads to better performance. In addition, the consideration of the impact of dataset construction and data preprocessing on model performance is also needed for model building. This paper proposes a framework named DeepLOCA-Lattice to experiment with different preprocessing approaches to fundamental deep learning models for a comprehensive analysis of the diagnosis of LOCA breach size. The DeepLOCA-Lattice involves data preprocessing via the lattice algorithm and equal-interval partitioning and deep-learning-based models, including the multi-layer perceptron (MLP), recurrent neural networks (RNNs), convolutional neural networks (CNNs), and the transformer model in LOCA breach size diagnosis. After conducting rigorous ablation experiments, we have discovered that even rudimentary foundational models can achieve accuracy rates that exceed 90%. This is a significant improvement when compared to the previous models, which yield an accuracy rate of lower than 50%. The results interestingly demonstrate the superior performance and efficacy of the fundamental deep learning model, with an effective dataset construction approach. It elucidates the presence of a complex interplay among diagnostic scales, sliding window size, and sliding stride. Furthermore, our investigation reveals that the model attains its highest accuracy within the discussed range when utilizing a smaller sliding stride size and a longer sliding window length. This study could furnish valuable insights for constructing models for LOCA breach size estimation. [ABSTRACT FROM AUTHOR]

Purpose: Patient-tailored minimal residual disease (MRD) monitoring based on circulating tumor DNA (ctDNA) sequencing of leukemia-specific mutations enables early detection of relapse for pre-emptive treatment, but its utilization in pediatric acute myelogenous leukemia (AML) is scarce. Thus, we aim to examine the role of ctDNA as a prognostic biomarker in monitoring response to the treatment of pediatric AML., Experimental Design: A prospective longitudinal study with 50 children with AML was launched, and sequential bone marrow (BM) and matched plasma samples were collected. The concordance of mutations by next-generation sequencing-based BM-DNA and ctDNA was evaluated. In addition, progression-free survival (PFS) and overall survival (OS) were estimated., Results: In 195 sample pairs from 50 patients, the concordance of leukemia-specific mutations between ctDNA and BM-DNA was 92.8%. Patients with undetectable ctDNA were linked to improved OS and PFS versus detectable ctDNA in the last sampling (both P < 0.001). Patients who cleared their ctDNA post three cycles of treatment had similar PFS compared with persistently negative ctDNA (P = 0.728). In addition, patients with >3 log reduction but without clearance in ctDNA were associated with an improved PFS as were patients with ctDNA clearance (P = 0.564)., Conclusions: Thus, ctDNA-based MRD monitoring appears to be a promising option to complement the overall assessment of pediatric patients with AML, wherein patients with continuous ctDNA negativity have the option for treatment de-escalation in subsequent therapy. Importantly, patients with >3 log reduction but without clearance in ctDNA may not require an aggressive treatment plan due to improved survival, but this needs further study to delineate., (©2024 American Association for Cancer Research.)

Objectives: To investigate the clinical features of juvenile myelomonocytic leukemia (JMML) and their association with prognosis., Methods: Clinical and prognosis data were collected from the children with JMML who were admitted from January 2008 to December 2016, and the influencing factors for prognosis were analyzed., Results: A total of 63 children with JMML were included, with a median age of onset of 25 months and a male/female ratio of 3.2∶1. JMML genetic testing was performed for 54 children, and PTPN11 mutation was the most common mutation and was observed in 23 children (43%), among whom 19 had PTPN11 mutation alone and 4 had compound PTPN11 mutation, followed by NRAS mutation observed in 14 children (26%), among whom 12 had NRAS mutation alone and 2 had compound NRAS mutation. The 5-year overall survival (OS) rate was only 22%±10% in these children with JMML. Of the 63 children, 13 (21%) underwent hematopoietic stem cell transplantation (HSCT). The HSCT group had a significantly higher 5-year OS rate than the non-HSCT group (46%±14% vs 29%±7%, P <0.05). There was no significant difference in the 5-year OS rate between the children without PTPN11 gene mutation and those with PTPN11 gene mutation (30%±14% vs 27%±10%, P >0.05). The Cox proportional-hazards regression model analysis showed that platelet count <40×10 9 /L at diagnosis was an influencing factor for 5-year OS rate in children with JMML ( P <0.05)., Conclusions: The PTPN11 gene was the most common mutant gene in JMML. Platelet count at diagnosis is associated with the prognosis in children with JMML. HSCT can improve the prognosis of children with JMML.