Spatial prediction of groundwater salinity in multiple aquifers of the Mekong Delta region using explainable machine learning models.

Groundwater salinization is a prevalent issue in coastal regions, yet accurately predicting and understanding its causal factors remains challenging due to the complexity of the groundwater system. Therefore, this study predicted groundwater salinity in multi-layered aquifers spanning the entire Mekong Delta (MD) region using machine learning (ML) models based on an in situ dataset and using three indicators (Cl-, pH, and HCO3-). We applied nine different decision tree-based models and evaluated their prediction performances. The models were trained using 13 input variables: weather (2), hydrogeological conditions (4), water levels (3), groundwater usage (2), and relative distance from water sources (2). Subsequently, by employing model interpretation techniques, we quantified the significance of factors within the model prediction. Performance evaluations of the ML models demonstrated that the Extra Trees model exhibited superior performance and demonstrated generalization capabilities in predicting Cl- concentration, whereas the Bagging and Random Forest models outperformed the other models in predicting pH and HCO3- concentration. The coefficients of determination were determined to be 0.94, 0.67, and 0.78 for Cl-, pH, and HCO3-, respectively Additionally, the model interpretation effectively identified significant factors that depended on the target variables and aquifers. In particular, salinity indicators and aquifers that were strongly influenced by the artificial usage of groundwater were identified. Therefore, our research, which provides accurate spatial predictions and interpretations of groundwater salinity in the MD, has the potential to establish a foundation for formulating effective groundwater management policies to control groundwater salinization.

Modeling freshwater plankton community dynamics with static and dynamic interactions using graph convolution embedded long short-term memory.

Given the frequent association between freshwater plankton and water quality degradation, several predictive models have been devised to understand and estimate their dynamics. However, the significance of biotic and abiotic interactions has been overlooked. In this study, we aimed to address the importance of the interaction term in predicting plankton community dynamics by applying graph convolution embedded long short-term memory networks (GC-LSTM) models, which can incorporate interaction terms as graph signals. Temporal graph series comprising plankton genera or environmental drivers as node features and their relationships for edge features from two distinct water bodies, a reservoir and a river, were utilized to develop these models. To assess the predictability, the performances of the GC-LSTM models on community dynamics were compared those of LSTM and GCN models at various lead times. Moreover, GNNExplainer was used to examine the global and local importance of the nodes and edges for all predictions and specific predictions, respectively. The GC-LSTM models outperformed the LSTM models, consistently showing higher prediction accuracy. Although all the models exhibited performance degradation at longer lead times, the GC-LSTM models consistently demonstrated better performance regarding each graph signal and plankton genus. GNNExplainer yielded interpretable explanations for important genera and interaction pairs among communities, revealing consistent importance patterns across different lead times at both global and local scales. These findings underscore the potential of the proposed modeling approach for forecasting community dynamics and emphasize the critical role of graph signals with interaction terms in plankton communities.

Distribution coefficient prediction using multimodal machine learning based on soil adsorption factors, XRF, and XRD spectrum data.

The distribution coefficient (Kd) plays a crucial role in predicting the migration behavior of radionuclides in the soil environment. However, Kd depends on the complexities of geological and environmental factors, and existing models often do not reflect the unique soil properties. We propose a multimodal technique to predict Kd values for radionuclide adsorption in soils surrounding nuclear facilities in Republic of Korea. We integrated and trained three sub-networks reflecting different data domains: soil adsorption factors for physicochemical conditions, X-ray fluorescence (XRF) data, and X-ray diffraction (XRD) spectra for inherent soil properties. Our multimodal model achieved high performance, with a coefficient of determination (R2) of 0.84 and root mean squared error (RMSE) of 0.89 for natural log-transformed Kd. This is the first study to develop a multimodal model that simultaneously incorporates inherent soil properties and adsorption factors to predict Kd. We investigated influential peaks in XRD spectra and also revealed that pH and calcium oxide (CaO) were significant variables in soil adsorption factors and XRF data, respectively. These results promote the use of a multimodal model to predict Kd values by integrating data from different domains, providing a cost-effective and novel approach to elucidate the mechanisms of radionuclide adsorption in soil.

A hybrid approach to improvement of watershed water quality modeling by coupling process-based and deep learning models.

Watershed water quality modeling to predict changing water quality is an essential tool for devising effective management strategies within watersheds. Process-based models (PBMs) are typically used to simulate water quality modeling. In watershed modeling utilizing PBMs, it is crucial to effectively reflect the actual watershed conditions by appropriately setting the model parameters. However, parameter calibration and validation are time-consuming processes with inherent uncertainties. Addressing these challenges, this research aims to address various challenges encountered in the calibration and validation processes of PBMs. To achieve this, the development of a hybrid model, combining uncalibrated PBMs with data-driven models (DDMs) such as deep learning algorithms is proposed. This hybrid model is intended to enhance watershed modeling by integrating the strengths of both PBMs and DDMs. The hybrid model is constructed by coupling an uncalibrated Soil and Water Assessment Tool (SWAT) with a Long Short-Term Memory (LSTM). SWAT, a representative PBM, is constructed using geographical information and 5-year observed data from the Yeongsan River Watershed. The output variables of the uncalibrated SWAT, such as streamflow, suspended solids (SS), total nitrogen (TN), and total phosphorus (TP), as well as observed precipitation for the day and previous day, are used as training data for the deep learning model to predict the TP load. For the comparison, the conventional SWAT model is calibrated and validated to predict the TP load. The results revealed that TP load simulated by the hybrid model predicted the observed TP better than that predicted by the calibrated SWAT model. Also, the hybrid model reflects seasonal variations in the TP load, including peak events. Remarkably, when applied to other sub-basins without specific training, the hybrid model consistently outperformed the calibrated SWAT model. In conclusion, application of the SWAT-LSTM hybrid model could be a useful tool for decreasing uncertainties in model calibration and improving the overall predictive performance in watershed modeling. PRACTITIONER POINTS: We aimed to enhance process-based models for watershed water-quality modeling. The Soil and Water Assessment Tool-Long Short-Term Memory hybrid model's predicted and total phosphorus (TP) matched the observed TP. It exhibited superior predictive performance when applied to other sub-basins. The hybrid model will overcome the constraints of conventional modeling. It will also enable more effective and efficient modeling.

Suction thrombectomy of distal medium vessel occlusion using microcatheter during mechanical thrombectomy for acute ischemic stroke: A case series.

While mechanical thrombectomy is known to be effective for distal medium vessel occlusion (DMVO) as well as large vessel occlusion, tortuous DMVO are predisposed to vessel injury during stent retriever thrombectomy. Furthermore, getting access to the thrombus may be difficult during suction thrombectomy using a dedicated suction catheter. Most studies describe DMVO treatment using stent retrievers and dedicated suction catheters, but there are limited studies reporting DMVO treated with suction thrombectomy using a microcatheter. Herein, we describe three cases of DMVO treated with suction thrombectomy that was performed using a microcatheter and subsequently showed good results. Therefore, suction thrombectomy using a microcatheter is a viable alternative treatment for tortuous DMVO.

Machine learning-based water quality prediction using octennial in-situ Daphnia magna biological early warning system data.

Biological early warning system (BEWS) has been globally used for surface water quality monitoring. Despite its extensive use, BEWS has exhibited limitations, including difficulties in biological interpretation and low alarm reproducibility. This study addressed these issues by applying machine learning (ML) models to eight years of in-situ BEWS data for Daphnia magna. Six ML models were adopted to predict contamination alarms from Daphnia behavioral parameters. The light gradient boosting machine model demonstrated the most significant improvement in predicting alarms from Daphnia behaviors. Compared with the traditional BEWS alarm index, the ML model enhanced the precision and recall by 29.50% and 43.41%, respectively. The speed distribution index and swimming speed were significant parameters for predicting water quality warnings. The nonlinear relationships between the monitored Daphnia behaviors and water physicochemical water quality parameters (i.e., flow rate, Chlorophyll-a concentration, water temperature, and conductivity) were identified by ML models for simulating Daphnia behavior based on the water contaminants. These findings suggest that ML models have the potential to establish a robust framework for advancing the predictive capabilities of BEWS, providing a promising avenue for real-time and accurate assessment of water quality. Thereby, it can contribute to more proactive and effective water quality management strategies.

Risk factor analyses of contrast leakage and contrast-induced encephalopathy following coil embolization for unruptured intracranial aneurysm.

BACKGROUND: Contrast-induced encephalopathy (CIE) following endovascular interventions is a rare but serious complication. This study aimed to investigate the risk factors of contrast leakage (CL) and CIE in patients who underwent coil embolization of unruptured intracranial aneurysms (UIAs). METHODS: Patients with UIAs who underwent coil embolization at a single tertiary institute between January 2019 and January 2022 were enrolled retrospectively. CL was defined as cortical or subcortical contrast enhancement with effacement of the cortical sulci. CIE was defined as the new onset of neurological deficits associated with CL. Following the procedure, all patients underwent CT scans, and MRI scans were performed on those with symptoms. Patient and procedural risk factors were investigated. RESULTS: In total, 459 patients were analyzed. The median procedure time and contrast dose were 69 min and 96 mL, respectively. CL was evident in 35 patients. In the multivariate analysis, hypertension, large aneurysm, longer procedure time, and greater contrast dose were associated with CL. CIE was diagnosed in 19 patients, and the risk factors included large aneurysm, longer procedure time, and greater contrast dose. The procedure time was predictive of both CL (P<0.001) and CIE (P=0.01). The optimal cut-off value for procedure time was 81.5 min. All CIE patients recovered completely within 8-96 hours. CONCLUSIONS: A large aneurysm and prolonged procedure time may increase the patient's risk of CL and CIE due to increased contrast exposure. Patients who underwent a procedure that exceeded 1.5 hours necessitate post-procedure evaluation and monitoring.

Automatic classification of microplastics and natural organic matter mixtures using a deep learning model.

Several preprocessing procedures are required for the classification of microplastics (MPs) in aquatic systems using spectroscopic analysis. Procedures such as oxidation, which are employed to remove natural organic matter (NOM) from MPs, can be time- and cost-intensive. Furthermore, the identification process is prone to errors due to the subjective judgment of the operators. Therefore, in this study, deep learning (DL) was applied to improve the classification accuracies for mixtures of microplastic and natural organic matter (MP-NOM). A convolutional neural network (CNN)-based DL model with a spatial attention mechanism was adopted to classify substances from their Raman spectra. Subsequently, the classification results were compared with those obtained using conventional Raman spectral library software to evaluate the applicability of the model. Additionally, the crucial spectral band for training the DL model was investigated by applying gradient-weighted class activation mapping (Grad-CAM) as a post-processing technique. The model achieved an accuracy of 99.54%, which is much higher than the 31.44% achieved by the Raman spectral library. The Grad-CAM approach confirmed that the DL model can effectively identify MPs based on their visually prominent peaks in the Raman spectra. Furthermore, by tracking distinctive spectra without relying solely on visually prominent peaks, we can accurately classify MPs with less prominent peaks, which are characterized by a high standard deviation of intensity. These findings demonstrate the potential for automated and objective classification of MPs without the need for NOM preprocessing, indicating a promising direction for future research in microplastic classification.

Distal stenting technique for coil embolization of early branch aneurysms of middle cerebral artery.

BACKGROUND AND PURPOSE: Stent-assisted coil embolization of early branch aneurysms of the middle cerebral artery (MCA) can sometimes be challenging due to the small diameter and acute angle of the branches. This study aimed to analyze the treatment results and report the feasibility and outcomes of the distal stenting technique for these aneurysms. MATERIALS AND METHODS: The distal stenting technique was used for 15 wide-neck MCA aneurysms (females, 10; males, 5; mean age, 65.1 years) originating from the early branch between December 2018 and October 2021. The average sizes of the dome, depth, and neck of the aneurysms were 4.17 mm (range: 2.99-6.21 mm), 2.86 mm (range: 1.82-3.72 mm), and 3.42 mm (range: 2.44-4.32 mm), respectively. RESULTS: The average diameter of the stents was 4.0 mm (3.0 mm, 3; 4.0 mm, 6; 4.5 mm, 6). Stents were successfully deployed in all 15 aneurysms (100%). The average length of the procedure was 69 min (range: 45-117 min). On postoperative angiography, nine (60.0%) aneurysms were completely occluded, four (26.7%) had neck remnants, and two (13.3%) had contrast flow in the sac. During the procedure, vasospasm and thrombus formation occurred in one case each, but no neurological sequelae were observed. On follow-up digital subtraction angiography of the 10 aneurysms after an average of 13.7 months (range: 12-18 months), the branches were well preserved, and neck remnants were noted in four aneurysms (40.0%). No thromboembolic events occurred during the clinical follow-up. CONCLUSION: The distal stenting technique using an open-cell stent can be a good option for coil embolization of early branch MCA aneurysms.

Nutrition Therapy for Patients With Traumatic Brain Injury: A Narrative Review.

Traumatic brain injury (TBI) is a global health and socio-economic problem, resulting in significant disability and mortality. Malnutrition is common in TBI patients and is associated with increased vulnerability to infection, higher morbidity and mortality rates, as well as longer stays in the intensive care unit and hospital. Following TBI, various pathophysiological mechanisms, such as hypermetabolism and hypercatabolism, affect patient outcomes. It is crucial to provide adequate nutrition therapy to prevent secondary brain damage and promote optimal recovery. This review includes a literature review and discusses the challenges encountered in clinical practice regarding nutrition in TBI patients. The focus is on determining energy requirements, timing and methods of nutrition delivery, promoting enteral tolerance, providing enteral nutrition to patients receiving vasopressors, and implementing trophic enteral nutrition. Enhancing our understanding of the current evidence regarding appropriate nutrition practices will contribute to improving overall outcomes for TBI patients.

Recent Updates on Controversies in Decompressive Craniectomy and Cranioplasty: Physiological Effect, Indication, Complication, and Management.

Decompressive craniectomy (DCE) and cranioplasty (CP) are surgical procedures used to manage elevated intracranial pressure (ICP) in various clinical scenarios, including ischemic stroke, hemorrhagic stroke, and traumatic brain injury. The physiological changes following DCE, such as cerebral blood flow, perfusion, brain tissue oxygenation, and autoregulation, are essential for understanding the benefits and limitations of these procedures. A comprehensive literature search was conducted to systematically review the recent updates in DCE and CP, focusing on the fundamentals of DCE for ICP reduction, indications for DCE, optimal sizes and timing for DCE and CP, the syndrome of trephined, and the debate on suboccipital CP. The review highlights the need for further research on hemodynamic and metabolic indicators following DCE, particularly in relation to the pressure reactivity index. It provides recommendations for early CP within three months of controlling increased ICP to facilitate neurological recovery. Additionally, the review emphasizes the importance of considering suboccipital CP in patients with persistent headaches, cerebrospinal fluid leakage, or cerebellar sag after suboccipital craniectomy. A better understanding of the physiological effects, indications, complications, and management strategies for DCE and CP to control elevated ICP will help optimize patient outcomes and improve the overall effectiveness of these procedures.

Optimizing Mannitol Use in Managing Increased Intracranial Pressure: A Comprehensive Review of Recent Research and Clinical Experiences.

Mannitol, derived from mannose sugar, is crucial in treating patients with elevated intracranial pressure (ICP). Its dehydrating properties at the cellular and tissue levels increase plasma osmotic pressure, which is studied for its potential to reduce ICP through osmotic diuresis. While clinical guidelines support mannitol use in these cases, the best approach for its application continues to be debated. Important aspects needing further investigation include: 1) bolus administration versus continuous infusion, 2) ICP-based dosing versus scheduled bolus, 3) identifying the optimal infusion rate, 4) determining the appropriate dosage, 5) establishing fluid replacement plans for urinary loss, and 6) selecting monitoring techniques and thresholds to assess effectiveness and ensure safety. Due to the lack of adequate high-quality prospective research data, a comprehensive review of recent studies and clinical trials is crucial. This assessment aims to bridge the knowledge gap, improve understanding of effective mannitol use in elevated ICP patients, and provide insights for future research. In conclusion, this review aspires to contribute to the ongoing discourse on mannitol application. By integrating the latest findings, this review will offer valuable insights into the function of mannitol in decreasing ICP, thereby informing better therapeutic approaches and enhancing patient outcomes.

Treatment results of anterior choroidal artery aneurysms treated mostly with coil embolization: A single-center experience.

BACKGROUND: Anterior choroidal artery (AchA) aneurysms are usually small in diameter because of the size of the involved artery and are often wide-necked. Coil embolization of AchA aneurysm is thus challenging because of serious risks, such as thromboembolic occlusion of artery and perforation of aneurysm dome. Therefore, aneurysmal neck clipping remains widely performed despite a recent increase in the use of coil embolization for aneurysm treatment. We report the treatment results of AchA aneurysms mostly (92.3%) treated with coil embolization at our institute. METHODS: The database and medical records of patients who underwent coil embolization for AchA aneurysms were retrospectively analyzed. The clinical and imaging results and procedure-related complications were investigated after coil embolization performed between January 2006 and March 2022 at our institute. RESULTS: In total, 96 AchA aneurysms comprising 65 unruptured and 31 ruptured aneurysms, including only 1 ruptured aneurysm (1.0%) re-embolized at postoperative day 192 because of coil compaction, were evaluated. After the initial coil embolization, complete occlusion was attained in 41, residual neck in 45, and residual aneurysm in 10 patients. Follow-up radiological studies after 6-174 months were performed for 80 aneurysms. Complete occlusion was noted in 57 patients, residual neck in 22, and residual aneurysm in 1. The dysarthria experienced by one (1.0%) patient was the only symptomatic procedure-related complication. After coil embolization, neither delayed new rupture nor re-rupture was observed. CONCLUSIONS: The results of this study demonstrate that coil embolization is a safe and effective treatment option for patients with AchA aneurysms.

Distribution and Afferent Effects of Transplanted mESCs on Cochlea in Acute and Chronic Neural Hearing Loss Models.

Hearing loss is a sensory deprivation that can affect the quality of life. Currently, only rehabilitation devices such as hearing aids and cochlear implants are used, without a definitive cure. However, in chronic hearing-deprived patients, in whom secondary auditory neural degeneration is expected, a relatively poor rehabilitation prognosis is projected. Stem cell therapy for cochlear neural structures would be an easier and feasible strategy compared with cochlear sensory cells. Considering the highly developed cochlear implantation technology, improving cochlear neural health has significant medical and social effects. Stem cell delivery to Rosenthal's canal in an acutely damaged mouse model has been performed and showed cell survival and the possibility of differentiation. The results of stem cell transplantation in chronic auditory neural hearing loss should be evaluated because neural stem cell replacement therapy for chronic (long-term) sensorineural hearing loss is a major target in clinics. In the present study, we established a mouse model that mimicked chronic auditory neural hearing loss (secondary degeneration of auditory neurons after loss of sensory input). Then, mouse embryonic stem cells (mESCs) were transplanted into the scala tympani and survival and distribution of transplanted cells were compared between the acute and chronic auditory neural hearing loss models induced by ouabain or kanamycin (KM), respectively. The mESC survival was similar to the acute model, and perilymphatic distribution of cell aggregates was more predominant in the chronic model. Lastly, the effects of mESC transplantation on neural signal transduction observed in the cochlear nucleus (CN) were compared and a statistical increase was observed in the chronic model compared with other models. These results indicated that after transplantation, mESCs survived in the cochlea and increased the neural signaling toward the central auditory pathway, even in the chronic (secondary) hearing loss mouse model.

Complete Genome Sequence of Weissella cibaria Strain BM2, Isolated from Korean Kimchi.

Weissella cibaria appears to have broad-spectrum health benefits. Here, we report the genome sequence of Weissella cibaria strain BM2, which was isolated from homemade kimchi; it consists of one circular chromosome of 2,462,443 bp and one plasmid of 11,067 bp. A total of 2,337 coding sequences were predicted, including 2,117 protein-coding sequences and a G+C content of 45.06%.

Complete Genome Sequence of Bacillus velezensis Strain DKU_NT_04, Isolated from a Traditional Korean Food Made from Soybeans (Cheonggukjang).

In the present work, we report the complete genome sequence of Bacillus velezensis DKU_NT_04, isolated from cheonggukjang, which is a traditional Korean fermented soybean paste. The final genome assembly consists of a 4.328-Mbp chromosome with 4,134 coding sequences and a G+C content of 45.21%.

Complete Genome Sequence of Clostridium butyricum Strain DKU_butyricum 4-1, Isolated from Infant Feces.

Clostridium butyricum is a strictly anaerobic spore-forming bacillus that is commonly present in the gut of humans. We report here the complete genome sequence of Clostridium butyricum strain DKU_butyricum 4-1, isolated from infant feces.

Complete Genome Sequence of Bacillus subtilis Strain DKU_NT_02, Isolated from Traditional Korean Food Using Soybean (Chung-gook-jang) for High-Quality Poly-γ-Glutamic Acid Activity.

The complete genome sequence of Bacillus subtilis strain DKU_NT_02, isolated from traditional Korean food using soybeans (chung-gook-jang), is presented here. This strain was chosen to help identify genetic factors with high-quality poly-γ-glutamic acid (γPGA) activity.

Complete Genome Sequence of Bacillus subtilis Strain DKU_NT_03, Isolated from a Traditional Korean Food Using Soybean (Chung-gook-jang) for High-Quality Nattokinase Activity.

We present here the complete genome sequence of Bacillus subtilis strain DKU_NT_03 isolated from the traditional Korean food chung-gook-jang, which is made from soybeans. This strain was chosen to identify genetic factors with high-quality nattokinase activity.

Mortality and Epidemiology in 256 Cases of Pediatric Traumatic Brain Injury: Korean Neuro-Trauma Data Bank System (KNTDBS) 2010-2014.

OBJECTIVE: Among pediatric injury, brain injury is a leading cause of death and disability. To improve outcomes, many developed countries built neurotrauma databank (NTDB) system but there was not established nationwide coverage NTDB until 2009 and there have been few studies on pediatric traumatic head injury (THI) patients in Korea. Therefore, we analyzed epidemiology and outcome from the big data of pediatric THI. METHODS: We collected data on pediatric patients from 23 university hospitals including 9 regional trauma centers from 2010 to 2014 and analyzed their clinical factors (sex, age, initial Glasgow coma scale, cause and mechanism of head injury, presence of surgery). RESULTS: Among all the 2617 THI patients, total number of pediatric patients was 256. The average age of the subjects was 9.07 (standard deviation±6.3) years old. The male-to female ratio was 1.87 to 1 and male dominance increases with age. The most common cause for trauma were falls and traffic accidents. Age (p=0.007), surgery (p<0.001), mechanism of trauma (p=0.016), subdural hemorrhage (SDH) (p<0.001), diffuse axonal injury (DAI) (p<0.001) were statistically significant associated with severe brain injury. CONCLUSION: Falls were the most common cause of trauma, and age, surgery, mechanism of trauma, SDH, DAI increased with injury severity. There is a critical need for effective fall and traffic accidents prevention strategies for children, and we should give attention to these predicting factors for more effective care.