RESUMEN
In coal mining areas, surface subsidence poses significant risks to human life and property. Fortunately, surface subsidence caused by coal mining can be monitored and predicted by using various methods, e.g., probability integral method and deep learning (DL) methods. Although DL methods show promise in predicting subsidence, they often lack accuracy due to insufficient consideration of spatial correlation and temporal nonlinearity. Considering this issue, we propose a novel DL-based approach for predicting mining surface subsidence. Our method employs K-means clustering to partition spatial data, allowing the application of a gate recurrent unit (GRU) model to capture nonlinear relationships in subsidence time series within each partition. Optimization using snake optimization (SO) further enhances model accuracy globally. Validation shows our method outperforms traditional Long Short-Term Memory (LSTM) and GRU models, achieving 99.1% of sample pixels with less than 8 mm absolute error.
RESUMEN
Backgrounds and aims: Immunotherapies have formed an entirely new treatment paradigm for hepatocellular carcinoma (HCC). Tertiary lymphoid structure (TLS) has been associated with good response to immunotherapy in most solid tumors. Nonetheless, the role of TLS in human HCC remains controversial, and recent studies suggest that their functional heterogeneity may relate to different locations within the tumor. Exploring factors that influence the formation of TLS in HCC may provide more useful insights. However, factors affecting the presence of TLSs are still unclear. The human gut microbiota can regulate the host immune system and is associated with the efficacy of immunotherapy but, in HCC, whether the gut microbiota is related to the presence of TLS still lacks sufficient evidence. Methods: We performed pathological examinations of tumor and para-tumor tissue sections. Based on the location of TLS in tissues, all patients were divided into intratumoral TLS (It-TLS) group and desertic TLS (De-TLS) group. According to the grouping results, we statistically analyzed the clinical, biological, and pathological features; preoperative gut microbiota data; and postoperative pathological features of patients. Results: In a retrospective study cohort of 60 cases from a single center, differential microbiota analysis showed that compared with the De-TLS group, the abundance of Lachnoclostridium, Hungatella, Blautia, Fusobacterium, and Clostridium was increased in the It-TLS group. Among them, the enrichment of Lachnoclostridium was the most significant and was unrelated to the clinical, biological, and pathological features of the patients. It can be seen that the difference in abundance levels of microbiota is related to the presence of TLS. Conclusion: Our findings prove the enrichment of Lachnoclostridium-dominated gut microbiota is associated with the presence of It-TLS in HCC patients.