Transactions on Intelligent Computing Systems - Articles

Laurent, A.

The mining of extremely thick coal seams causes strong disturbances to the floor strata. When a strong aquifer exists beneath the coal seam and the effective aquiclude is weak, the risk of floor water inrush increases significantly. In recent years, microseismic (MS) monitoring technology has been widely applied, and using MS monitoring to assess floor water inrush risk has proven to be a reliable and effective approach. To address the high-risk problem of floor water inrush during the extraction of extremely thick coal seams, this study employed MS monitoring to investigate the evolution characteristics of floor MS activity and its associated water inrush risk. The MS response characteristics induced by thick seam mining were analyzed, and a floor water inrush risk evaluation model was established based on the Analytic Hierarchy Process–Entropy Weight Method (AHP–EWM). The evaluation results were consistent with field monitoring data. The results indicate that most floor MS events in the 61607 working face of the Longwanggou Coal Mine were concentrated in the Taiyuan Formation. The MS monitoring indicators, such as the extent of the advanced influence zone, floor failure depth, and total daily released energy, showed strong correlations with observed data including water inflow and mine pressure. By summarizing the key controlling factors of floor water-inrush risk, this study departs from traditional static geological parameter indicators and instead focuses on dynamic indicators derived from the activity, spatiotemporal characteristics, and intensity characteristics of floor MS events. Integrating multiple influential factors of floor water inrush, a quantitative evaluation system for floor water-inrush hazard in extremely thick coal seam mining was established using the AHP–EWM combined subjective–objective weighting method, which shows good consistency with field monitoring results. This study provides theoretical and data support for analyzing the dynamic process and influencing factors of floor damage and fracture evolution under thick seam mining. It offers a new approach for predicting and forecasting water inrush risk, enriches the application of MS monitoring in mine floor water hazard prevention, and holds significant practical importance for promoting safe coal mine production.

Im, H., Jang, H.

Background Preemptive kidney transplantation (PKT) is associated with reduced dialysis-related morbidity and improved graft survival, which has led to an increase in the prevalence of PKT. The distinct clinical pathways of PKT and non-preemptive kidney transplantation (NPKT) patients may affect their ability to perform self-care, a key factor in post-transplant recovery. However, there is ongoing controversy regarding which group demonstrates better self-care abilities, highlighting the need to explore the factors influencing self-care in each group. Objectives This study aims to identify and compare predictors of self-care in PKT and NPKT groups. Methods We collected data from 209 KT recipients, consisting of 101 PKT and 108 NPKT patients, using self-administered questionnaires. These questionnaires assessed general and disease-specific characteristics, stress, social support, and self-care behaviors. Data were analyzed using descriptive statistics, t-tests, ANOVA, and multiple regression analysis. Results The study identified significant predictors of self-care among both PKT and NPKT patients. For PKT patients, marital status (β = .19, p = .033) and family support (β = .28, p = .006) are key predictors of self-care, with those having spousal support and strong family networks reporting better self-care levels. Conversely, in NPKT patients, significant predictors of self-care included marital status (β = .31, p = .001), employment status (β = .29, p = .007), post-transplantation duration (less than 36 month) (β = −.22, p = .015), post-transplantation duration (36–72 month) (β = −.33, p = .001), and stress levels (β = −.20, p = .028). Conclusions The study provides valuable insights into the predictors of self-care based on preoperative dialysis status, illustrating distinct predictors between the growing population of PKT and NPKT patients. These findings emphasize the need for personalized nursing strategies to enhance post-transplant self-care, tailored to individual patient characteristics.

Rieke, N., Hancox, J., Li, W., Milletarì, F., Roth, H., Albarqouni, S., Bakas, S., Galtier, M., Landman, B., Maier-Hein, K., Ourselin, S., Sheller, M., Summers, R., Trask, A., Xu, D., Baust, M., Cardoso, M.

Data-driven machine learning (ML) has emerged as a promising approach for building accurate and robust statistical models from medical data, which is collected in huge volumes by modern healthcare systems. Existing medical data is not fully exploited by ML primarily because it sits in data silos and privacy concerns restrict access to this data. However, without access to sufficient data, ML will be prevented from reaching its full potential and, ultimately, from making the transition from research to clinical practice. This paper considers key factors contributing to this issue, explores how federated learning (FL) may provide a solution for the future of digital health and highlights the challenges and considerations that need to be addressed.

Holmberg, B.

Bellini, D.

Schäfer, M.

Natural Language Processing (NLP) is one of the most captivating applications of Deep Learning. In this survey, we consider how the Data Augmentation training strategy can aid in its development. We begin with the major motifs of Data Augmentation summarized into strengthening local decision boundaries, brute force training, causality and counterfactual examples, and the distinction between meaning and form. We follow these motifs with a concrete list of augmentation frameworks that have been developed for text data. Deep Learning generally struggles with the measurement of generalization and characterization of overfitting. We highlight studies that cover how augmentations can construct test sets for generalization. NLP is at an early stage in applying Data Augmentation compared to Computer Vision. We highlight the key differences and promising ideas that have yet to be tested in NLP. For the sake of practical implementation, we describe tools that facilitate Data Augmentation such as the use of consistency regularization, controllers, and offline and online augmentation pipelines, to preview a few. Finally, we discuss interesting topics around Data Augmentation in NLP such as task-specific augmentations, the use of prior knowledge in self-supervised learning versus Data Augmentation, intersections with transfer and multi-task learning, and ideas for AI-GAs (AI-Generating Algorithms). We hope this paper inspires further research interest in Text Data Augmentation.

Ji, H., Shi, X., Wang, S., Wang, S., Yan, D., Tong, L., Zhong, Q., Feng, M., Hao, S., Li, W., Wang, X.

Intravenous administration of vesicants or irritants frequently leads to chemical phlebitis. We aimed to establish a chemical phlebitis rat model by injection vinorelbine via dorsal pedal vein and evaluate the general symptoms as well as the pathological features of lesion venous tissues. Rats in 6 groups with each 8 were injected 0.1 mL saline or vinorelbine (1–5 mg/mL) via the dorsalis pedis vein in hind paw, respectively. The edema and pain were assessed daily by swelling rates (water displacement) and weight-bearing ratio (bipedal balance test), respectively. Phlebitis severity was evaluated by phlebitis scale. Venous tissues harvested on day 7 were underwent histopathological and ultrastructural evaluation. Vinorelbine administration elicited dose-dependent inflammatory responses, with erythema, edema, pain, palpable cord formation, and purulent drainage. Edema and pain peaked on day 3, as evidenced by increased swelling rates and reduced weight-bearing ratios. Persistent symptomatology was observed in the 3–5 mg/mL groups through day 7. Histopathological analysis demonstrated venous wall thickening characterized by intimal hyperplasia and collagen deposition, along with vascular lumen stenosis accompanied by thrombosis. Furthermore, disruption of the intimal lining was observed, with shrunken endothelial cells interspersed among extracellular matrix. These pathological changes also exhibited a dose-dependent exacerbation with increasing concentrations of vinorelbine. Dorsal pedal vein injection of 0.1 mL vinorelbine (3–5 mg/mL) successfully induced chemical phlebitis in rats, mimicking human clinical presentations. Furthermore, vinorelbine-induced endothelial damage resulted in distinct venous lesions, characterized by vascular wall thickening with intimal hyperplasia, luminal stenosis and thrombus.

Haddad, N.

Intravenous administration of vesicants or irritants frequently leads to chemical phlebitis. We aimed to establish a chemical phlebitis rat model by injection vinorelbine via dorsal pedal vein and evaluate the general symptoms as well as the pathological features of lesion venous tissues. Rats in 6 groups with each 8 were injected 0.1 mL saline or vinorelbine (1–5 mg/mL) via the dorsalis pedis vein in hind paw, respectively. The edema and pain were assessed daily by swelling rates (water displacement) and weight-bearing ratio (bipedal balance test), respectively. Phlebitis severity was evaluated by phlebitis scale. Venous tissues harvested on day 7 were underwent histopathological and ultrastructural evaluation. Vinorelbine administration elicited dose-dependent inflammatory responses, with erythema, edema, pain, palpable cord formation, and purulent drainage. Edema and pain peaked on day 3, as evidenced by increased swelling rates and reduced weight-bearing ratios. Persistent symptomatology was observed in the 3–5 mg/mL groups through day 7. Histopathological analysis demonstrated venous wall thickening characterized by intimal hyperplasia and collagen deposition, along with vascular lumen stenosis accompanied by thrombosis. Furthermore, disruption of the intimal lining was observed, with shrunken endothelial cells interspersed among extracellular matrix. These pathological changes also exhibited a dose-dependent exacerbation with increasing concentrations of vinorelbine. Dorsal pedal vein injection of 0.1 mL vinorelbine (3–5 mg/mL) successfully induced chemical phlebitis in rats, mimicking human clinical presentations. Furthermore, vinorelbine-induced endothelial damage resulted in distinct venous lesions, characterized by vascular wall thickening with intimal hyperplasia, luminal stenosis and thrombus.

Shimizu, I., Ida, T., Ozawa, Y., Fukuchi, S., Anbo, H.

Background Intrinsically disordered proteins (IDPs) are proteins that contain intrinsically disordered regions (IDRs), which lack stable three-dimensional structures under physiological conditions. These regions are known to play crucial roles in many biological processes. While IDRs can be predicted from their amino acid sequences, and several accurate IDR prediction programs have been developed, such programs often require substantial computational resources, including long execution times, large databases for homology searches, and advanced computer architectures. Since DNA sequence data continues to grow rapidly, particularly at a genomic scale, there is an increasing need for fast and accurate IDR prediction programs that demand fewer computational resources. Methods In this study, we developed DARUMA (Disorder order clAssifier by Rapid and User-friendly MAchine), an IDR prediction program designed for speed and ease of use. DARUMA uses a one-dimensional convolutional neural network (1D-CNN) that processes the physicochemical properties of amino acid residues instead of relying on sequence profiles. DARUMA employs a simple neural network that predicts IDRs using the output of 1D-CNN as input features. To ensure easy installation on users’ systems, DARUMA was written entirely in Python using standard and NumPy libraries. Results DARUMA achieves fast performance by avoiding iterative homology searches while delivering accuracy comparable to the latest predictors that use sequence profiles. In addition to the advantage of execution time, DARUMA requires no additional homology search programs and operates using standard Python libraries, making it easy to install and run on users’ own environments without the need for specialized computational resources. DARUMA is available at https://antepontem.org/daruma/, which also provides the stand-alone distribution.

Wiśniewski, T., Mbeki, K., Lindgren, S.

Accelerographs are essential instruments for quantifying strong ground motion, serving as the foundation of modern earthquake engineering. In Peru, the first accelerographic station was installed in Lima in 1944; since then, various institutions have promoted the expansion of the national network. However, this network’s spatial coverage and instrumentation remain insufficient to properly characterize strong motion and support seismic risk reduction policies. In this context, the KUYUY accelerograph is presented as a low-cost, low-noise device equipped with real-time telemetry and high-performance MEMS sensors. Its interoperability with the Intelligent Automatic Processing System (SIPA) enables real-time monitoring and automated signal analysis for seismic microzonation studies and rapid damage assessment, contributing to seismic risk reduction in Peru. The validation process included static gravity calibration, field comparison with a reference accelerograph, and an initial deployment in Lima and Yurimaguas. The results demonstrate the proposed accelerograph’s linear response, temporal stability, and amplitude consistency with respect to high-end instruments, with differences below 5–10%.