Artificial intelligence-based forecasting model for incinerator in sulfur recovery units to predict SO2 emissions.

Pollutant emissions from chemical plants are a major concern in the context of environmental safety. A reliable emission forecasting model can provide important information for optimizing the process and improving the environmental performance. In this work, forecasting models are developed for the prediction of SO2 emission from a Sulfur Recovery Unit (SRU). Since SRUs incorporate complex chemical reactions, first-principle models are not suitable to predict emission levels based on a given feed condition. Accordingly, artificial intelligence-based models such as standard machine learning (ML) algorithms, multi-layer perceptron (MLP), long short-term memory (LSTM), one-dimensional convolution (1D-CNN), and CNN-LSTM models were tested, and their performance was evaluated. The input features and hyperparameters of the models were optimized to achieve maximum performance. The performance was evaluated in terms of mean squared error (MSE) and mean absolute percentage Error (MAPE) for 1 h, 3 h and 5 h ahead of forecasting. The reported results show that the CNN-LSTM encoder-decoder model outperforms other tested models, with its superiority becoming more pronounced as the forecasting horizon increased from 1 h to 5 h. For the 5-h ahead forecasting, the proposed model showed a MAPE advantage of 17.23%, 4.41%, and 2.83%, respectively over the 1D-CNN, Deep LSTM, and single-layer LSTM models in the larger dataset.

Methane prediction equations including genera of rumen bacteria as predictor variables improve prediction accuracy.

Methane (CH4) emissions from ruminants are of a significant environmental concern, necessitating accurate prediction for emission inventories. Existing models rely solely on dietary and host animal-related data, ignoring the predicting power of rumen microbiota, the source of CH4. To address this limitation, we developed novel CH4 prediction models incorporating rumen microbes as predictors, alongside animal- and feed-related predictors using four statistical/machine learning (ML) methods. These include random forest combined with boosting (RF-B), least absolute shrinkage and selection operator (LASSO), generalized linear mixed model with LASSO (glmmLasso), and smoothly clipped absolute deviation (SCAD) implemented on linear mixed models. With a sheep dataset (218 observations) of both animal data and rumen microbiota data (relative sequence abundance of 330 genera of rumen bacteria, archaea, protozoa, and fungi), we developed linear mixed models to predict CH4 production (g CH4/animal·d, ANIM-B models) and CH4 yield (g CH4/kg of dry matter intake, DMI-B models). We also developed models solely based on animal-related data. Prediction performance was evaluated 200 times with random data splits, while fitting performance was assessed without data splitting. The inclusion of microbial predictors improved the models, as indicated by decreased root mean square prediction error (RMSPE) and mean absolute error (MAE), and increased Lin's concordance correlation coefficient (CCC). Both glmmLasso and SCAD reduced the Akaike information criterion (AIC) and Bayesian information criterion (BIC) for both the ANIM-B and the DMI-B models, while the other two ML methods had mixed outcomes. By balancing prediction performance and fitting performance, we obtained one ANIM-B model (containing 10 genera of bacteria and 3 animal data) fitted using glmmLasso and one DMI-B model (5 genera of bacteria and 1 animal datum) fitted using SCAD. This study highlights the importance of incorporating rumen microbiota data in CH4 prediction models to enhance accuracy and robustness. Additionally, ML methods facilitate the selection of microbial predictors from high-dimensional metataxonomic data of the rumen microbiota without overfitting. Moreover, the identified microbial predictors can serve as biomarkers of CH4 emissions from sheep, providing valuable insights for future research and mitigation strategies.

Full adoption of the most effective strategies to mitigate methane emissions by ruminants can help meet the 1.5 °C target by 2030 but not 2050.

To meet the 1.5 °C target, methane (CH4) from ruminants must be reduced by 11 to 30% by 2030 and 24 to 47% by 2050 compared to 2010 levels. A meta-analysis identified strategies to decrease product-based (PB; CH4 per unit meat or milk) and absolute (ABS) enteric CH4 emissions while maintaining or increasing animal productivity (AP; weight gain or milk yield). Next, the potential of different adoption rates of one PB or one ABS strategy to contribute to the 1.5 °C target was estimated. The database included findings from 430 peer-reviewed studies, which reported 98 mitigation strategies that can be classified into three categories: animal and feed management, diet formulation, and rumen manipulation. A random-effects meta-analysis weighted by inverse variance was carried out. Three PB strategies­namely, increasing feeding level, decreasing grass maturity, and decreasing dietary forage-to-concentrate ratio­decreased CH4 per unit meat or milk by on average 12% and increased AP by a median of 17%. Five ABS strategies­namely CH4 inhibitors, tanniferous forages, electron sinks, oils and fats, and oilseeds­decreased daily methane by on average 21%. Globally, only 100% adoption of the most effective PB and ABS strategies can meet the 1.5 °C target by 2030 but not 2050, because mitigation effects are offset by projected increases in CH4 due to increasing milk and meat demand. Notably, by 2030 and 2050, low- and middle-income countries may not meet their contribution to the 1.5 °C target for this same reason, whereas high-income countries could meet their contributions due to only a minor projected increase in enteric CH4 emissions.

Phase II trial of proton therapy versus photon IMRT for GBM: secondary analysis comparison of progression-free survival between RANO versus clinical assessment.

BACKGROUND: This secondary image analysis of a randomized trial of proton radiotherapy (PT) versus photon intensity-modulated radiotherapy (IMRT) compares tumor progression based on clinical radiological assessment versus Response Assessment in Neuro-Oncology (RANO). METHODS: Eligible patients were enrolled in the randomized trial and had MR imaging at baseline and follow-up beyond 12 weeks from completion of radiotherapy. "Clinical progression" was based on a clinical radiology report of progression and/or change in treatment for progression. RESULTS: Of 90 enrolled patients, 66 were evaluable. Median clinical progression-free survival (PFS) was 10.8 (range: 9.4-14.7) months; 10.8 months IMRT versus 11.2 months PT (P = .14). Median RANO-PFS was 8.2 (range: 6.9, 12): 8.9 months IMRT versus 6.6 months PT (P = .24). RANO-PFS was significantly shorter than clinical PFS overall (P = .001) and for both the IMRT (P = .01) and PT (P = .04) groups. There were 31 (46.3%) discrepant cases of which 17 had RANO progression more than a month prior to clinical progression, and 14 had progression by RANO but not clinical criteria. CONCLUSIONS: Based on this secondary analysis of a trial of PT versus IMRT for glioblastoma, while no difference in PFS was noted relative to treatment technique, RANO criteria identified progression more often and earlier than clinical assessment. This highlights the disconnect between measures of tumor response in clinical trials versus clinical practice. With growing efforts to utilize real-world data and personalized treatment with timely adaptation, there is a growing need to improve the consistency of determining tumor progression within clinical trials and clinical practice.

Prediction of enteric methane production, yield, and intensity in dairy cattle using an intercontinental database.

Enteric methane (CH4 ) production from cattle contributes to global greenhouse gas emissions. Measurement of enteric CH4 is complex, expensive, and impractical at large scales; therefore, models are commonly used to predict CH4 production. However, building robust prediction models requires extensive data from animals under different management systems worldwide. The objectives of this study were to (1) collate a global database of enteric CH4 production from individual lactating dairy cattle; (2) determine the availability of key variables for predicting enteric CH4 production (g/day per cow), yield [g/kg dry matter intake (DMI)], and intensity (g/kg energy corrected milk) and their respective relationships; (3) develop intercontinental and regional models and cross-validate their performance; and (4) assess the trade-off between availability of on-farm inputs and CH4 prediction accuracy. The intercontinental database covered Europe (EU), the United States (US), and Australia (AU). A sequential approach was taken by incrementally adding key variables to develop models with increasing complexity. Methane emissions were predicted by fitting linear mixed models. Within model categories, an intercontinental model with the most available independent variables performed best with root mean square prediction error (RMSPE) as a percentage of mean observed value of 16.6%, 14.7%, and 19.8% for intercontinental, EU, and United States regions, respectively. Less complex models requiring only DMI had predictive ability comparable to complex models. Enteric CH4 production, yield, and intensity prediction models developed on an intercontinental basis had similar performance across regions, however, intercepts and slopes were different with implications for prediction. Revised CH4 emission conversion factors for specific regions are required to improve CH4 production estimates in national inventories. In conclusion, information on DMI is required for good prediction, and other factors such as dietary neutral detergent fiber (NDF) concentration, improve the prediction. For enteric CH4 yield and intensity prediction, information on milk yield and composition is required for better estimation.

Combined screening for open spina bifida at 11-13 weeks using fetal biparietal diameter and maternal serum markers.

OBJECTIVE: Screening at 11-13 weeks with ultrasound biparietal diameter (BPD) can detect half of open spina bifida cases. Maternal serum α-fetoprotein (AFP) levels at 15-19 weeks are increased 3- to 4-fold, in open spina bifida. We assessed whether combined screening using BPD, AFP, and other serum markers at 11-13 weeks would increase detection. STUDY DESIGN: Maternal AFP levels were measured on serum stored at 11-13 weeks in 44 open spina bifida and 182 unaffected pregnancies, and results were expressed in multiples of the median (MoM) for gestational age. All samples had been measured for free ß-human chorionic gonadotropin (ß-hCG) and pregnancy-associated plasma protein (PAPP)-A. A multivariate Gaussian model was used to predict screening performance from the serum data and BPD measurements on 80 cases, including 36 previously published. RESULTS: The median AFP level in cases was 1.201 MoM, significantly higher than in unaffected pregnancies (P < .01, 1 tail). The median free ß-hCG was significantly reduced to 0.820 MoM (P < .02), but the median PAPP-A was similar in cases and controls. Modeling predicted the following: BPD alone would detect 50% of cases for a 5% false-positive rate or 63% for 10%; adding AFP increases detection by 2%; and a combined test with BPD, AFP, and free ß-hCG detects 58% for 5% or 70% for 10%. CONCLUSION: Combining AFP and BPD with free ß-hCG as part of first-trimester aneuploidy screening would also allow early detection about two-thirds of cases with open spina bifida.

Atrial tachycardia induces remodelling of muscarinic receptors and their coupled potassium currents in canine left atrial and pulmonary vein cardiomyocytes.

BACKGROUND AND PURPOSE: Both parasympathetic tone and atrial tachycardia (AT) remodelling of ion channels play important roles in atrial fibrillation (AF) pathophysiology. Different muscarinic cholinergic receptor (mAChR) subtypes (M2, M3, M4) in atrial cardiomyocytes are coupled to distinct K+-currents (called IKM2, IKM3, IKM4, respectively). Pulmonary veins (PVs) are important in AF and differential cholinergic current responses are a potential underlying mechanism. This study investigated AT-induced remodelling of mAChR subtypes and K+-currents in left-atrial (LA) and PV cardiomyocytes. EXPERIMENTAL APPROACH: Receptor expression was assayed by western blot. IKM2, IKM3 and IKM4 were recorded with whole-cell patch-clamp in LA and PV cardiomyocytes of nonpaced control dogs and dogs after 7 days of AT-pacing (400 bpm). KEY RESULTS: Current densities of IKM2, IKM3 and IKM4 were significantly reduced by AT-pacing in LA and PV cardiomyocytes. PV cardiomyocyte current-voltage relations were similar to LA for all three cholinergic currents, both in control and AT remodelling. Membrane-protein expression levels corresponding to M2, M3 and M4 subtypes were decreased significantly (by about 50%) after AT pacing. Agonist concentration-response relations for all three currents were unaffected by AT pacing. CONCLUSIONS AND IMPLICATIONS: AT downregulated all three mAChR-coupled K+-current subtypes, along with corresponding mAChR protein expression. These changes in cholinergic receptor-coupled function may play a role in AF pathophysiology. Cholinergic receptor-coupled K+-currents in PV cardiomyocytes were similar to those in LA under control and AT-pacing conditions, suggesting that differential cholinergic current properties do not explain the role of PVs in AF.