Fully Bayesian analysis of the relevance vector machine classification for imbalanced data problem

Relevance Vector Machine(RVM)is a supervised learning algorithm extended from Support Vector Machine based on the Bayesian sparsity model.Relevance Vector Machine classification suffers from theoretical limitations and computational inefficiency mainly because there is no closed-form solution for the posterior of the weight parameters.We propose two advanced Bayesian approaches for RVM classification,namely the Enhanced RVM and the Reinforced RVM,to perfect the theoretic framework of RVM and extend the algorithm to the imbalanced data problem,which has an arresting skew in data size between classes.First,the Enhanced RVM conducts a strict Bayesian sampling process instead of the approximation method in the original one to remedy its theoretic limitations,especially the nonconvergence of the iterations.Secondly,we conjecture that the hierarchical prior makes the Reinforced RVM achieve consistent estimations of the quantities of interest compared with the non-consistent estimations of the original RVM.Consistency is necessary for RVM classification since it makes the model more stable and localises the relevant vectors more accurately in the imbalanced data problem.The two-level prior also renders the Reinforced one competitive in the imbalanced data problem by building the inner connection of parameter dimensions and alloting a more vital relevance to the small class data weight parameter.The theoretic proofs and several numeric studies demonstrate the merits of our two proposed algorithms.

Buoyance-driven hydrocarbon accumulation depth and its implication for unconventional resource prediction

The discovery of unconventional hydrocarbon resources since the late 20th century changed geologists’understanding of hydrocarbon migration and accumulations and provides a solution to energy shortage.In 2016,unconventional oil production in the USA accounted for 41%of the total oil production;and unconventional natural gas production in China accounted for 35%of total gas production,showing strong growth momentum of unconventional hydrocarbons explorations.Unconventional hydrocarbons generally coexist with conventional petroleum resources;they sometimes distribute in a separate system,not coexisting with a conventional system.Identification and prediction of unconventional resources and their potentials are prominent challenges for geologists.This study analyzed the results of 12,237 drilling wells in six representative petroliferous basins in China and studied the correlations and differences between conventional and unconventional hydrocarbons by comparing their geological features.Migration and accumulation of conventional hydrocarbon are caused dominantly by buoyance.Wepropose a concept of buoyance-driven hydrocarbon accumulation depth to describe the deepest hydrocarbon accumulation depth driven dominantly by buoyance;beyond this depth the buoyance becomes unimportant for hydrocarbon accumulation.We found that the buoyance-driven hydrocarbon accumulation depth in petroliferous basins controls the different oil/gas reservoirs distribution and resource potentials.Hydrocarbon migration and accumulations above this depth is dominated by buoyancy,forming conventional reservoirs in traps with high porosity and permeability,while hydrocarbon migration and accumulation below this depth is dominated by non-buoyancy forces(mainly refers to capillary force,hydrocarbon volumeexpansion force,etc.),forming unconventional reservoirs in tight layers.The buoyance-driven hydrocarbon accumulation depths in six basins in China range from 1200mto 4200 m,which become shallowerwith increasing geothermal gradient,decreasing particle size of sandstone reservoir layers,or an uplift in the whole petroliferous basin.The predicted unconventional resource potential belowthe buoyance-driven hydrocarbon accumulation depth in six basins in China is more than 15.71×10^(9) t oil equivalent,among them 4.71×10^(9) t reserves have been proved.Worldwide,94%of 52,926 oil and gas reservoirs in 1186 basins are conventional reservoirs and only 6%of them are unconventional reservoirs.These 94%conventional reservoirs show promising exploration prospects in the deep area below buoyance-driven hydrocarbon accumulation depth.

A unified model for the formation and distribution of both conventional and unconventional hydrocarbon reservoirs

The discovery and large-scale exploration of unconventional oil/gas resources since 1980s have been considered as the most important advancement in the history of petroleum geology;that has not only changed the balance of supply and demand in the global energy market,but also improved our understanding of the formation mechanisms and distribution characteristics of oil/gas reservoirs.However,what is the difference of conventional and unconventional resources and why they always related to each other in petroliferous basins is not clear.As the differences and correlations between unconventional and conventional resources are complex challenging issues and very critical for resources assessment and hydrocarbon exploration,this paper focused on studying the relationship of formations and distributions among different oil/gas reservoirs.Drilling results of 12,237 exploratory wells in 6 representative petroliferous basins of China and distribution characteristics for 52,926 oil/gas accumulations over the world were applied to clarify the formation conditions and genetic relations of different oil/gas reservoirs in a petroliferous basin,and then to establish a unified model to address the differences and correlations of conventional and unconventional reservoirs.In this model,conventional reservoirs formed in free hydrocarbon dynamic field with high porosity and permeability located above the boundary of hydrocarbon buoyancy-driven accumulation depth limit.Unconventional tight reservoirs formed in confined hydrocarbon dynamic field with low porosity and permeability located between hydrocarbon buoyancy-driven accumulation depth limit and hydrocarbon accumulation depth limit.Shale oil/gas reservoirs formed in the bound hydrocarbon dynamic field with low porosity and ultra-low permeability within the source rock layers.More than 75%of proved reserves around the world are discovered in the free hydrocarbon dynamic field,which is estimated to contain only 10%of originally generated hydrocarbons.Most of undiscovered resources distributed in the confined hydrocarbon dynamic field and the bound hydrocarbon dynamic field,which contains 90%of original generated hydrocarbons,implying a reasonable and promising area for future hydrocarbon explorations.The buried depths of hydrocarbon dynamic fields become shallow with the increase of heat flow,and the remaining oil/gas resources mainly exist in the deep area of“cold basin”with low geothermal gradient.Lithology changing in the hydrocarbon dynamic field causes local anomalies in the oil/gas dynamic mechanism,leading to the local formation of unconventional hydrocarbon reservoirs in the free hydrocarbon dynamic field or the occurrence of oil/gas enrichment sweet points with high porosity and permeability in the confined hydrocarbon dynamic field.The tectonic movements destroy the medium conditions and oil/gas components,which leads to the transformation of conventional oil/gas reservoirs formed in free hydrocarbon dynamic field to unconventional ones or unconventional ones formed in confined and bound hydrocarbon dynamic fields to conventional ones.

Propofol inhibits cells migration and invasion via HOTAIR/miR-93/HIF-1α-mediated lactate secretion in colon cancer

Propofol,a common intravenous anesthetic used in clinical,has been shown to regulate cells proliferation,inflammation,angiogenesis and metastasis as well as exerting anti-cancer effect in several different types of cancer.However,the functions and mechanisms of propofol in lactate secretion and invasion of cancer cells remain unknown.The primary aim of this study was to investigate the role of propofol in cells migration and invasion of colon cancer(CRC).Scratches assay,Transwell assay were used to detect colon cells migration and invasion.Realtime Quantitative Polymerase Chain Reaction(RT-qPCR),Western blotting were utilized to detect the expression of related molecules.Dual-luciferase assay was performed to identify transcriptional activity of target genes,RNA pulldown assay was used to confirm the interactions of RNAs.Results indicated that propofol inhibited cell migration and invasion through suppressing lactic acid production in CRC.Moreover,propofol reduced lactic acid production by down-regulating the expression of hypoxia inducible factor-1α(HIF-1α)and monocarboxylate transporter 4(MCT4).In addition,propofol inhibited the expression of HIF-1αand cells migration/invasion via suppressing the interaction of HOX transcript antisense RNA(HOTAIR)with miR-93.The present study indicated that propofol inhibited colon cancer cell migration and invasion by reducing lactic acid production and the expression of HIF-1αand MCT4 via HOTAIR/miR-93 axis.These data may provide potential metabolic targets for CRC treatment.

A multimodal integration pipeline for accurate diagnosis, pathogen identification, and prognosis prediction of pulmonary infections

Pulmonary infections pose formidable challenges in clinical settings with high mortality rates across all age groups worldwide.Accurate diagnosis and early intervention are crucial to improve patient outcomes.Artificial intelligence(AI)has the capability to mine imaging features specific to different pathogens and fuse multimodal features to reach a synergistic diagnosis,enabling more precise investigation and individualized clinical management.In this study,we successfully developed a multimodal integration(MMI)pipeline to differentiate among bacterial,fungal,and viral pneumonia and pulmonary tuberculosis based on a real-world dataset of 24,107 patients.The area under the curve(AUC)of the MMI system comprising clinical text and computed tomography(CT)image scans yielded 0.910(95%confidence interval[CI]:0.904–0.916)and 0.887(95%CI:0.867–0.909)in the internal and external testing datasets respectively,which were comparable to those of experienced physicians.Furthermore,the MMI system was utilized to rapidly differentiate between viral subtypes with a mean AUC of 0.822(95%CI:0.805–0.837)and bacterial subtypes with a mean AUC of 0.803(95%CI:0.775–0.830).Here,the MMI system harbors the potential to guide tailored medication recommendations,thus mitigating the risk of antibiotic misuse.Additionally,the integration of multimodal factors in the AI-driven system also provided an evident advantage in predicting risks of developing critical illness,contributing to more informed clinical decision-making.To revolutionize medical care,embracing multimodal AI tools in pulmonary infections will pave the way to further facilitate early intervention and precise management in the foreseeable future.

Recent advances in the synthesis of benzo[b]thiophene fused polycyclic derivatives:Strategies and reactions

Benzo[b]thiophene fused compounds with a unique active heterocyclic skeleton have wide applications in the fields of medicinal chemistry,organic synthesis,and organic functional materials,which resulted in rapid development of many efficient methods for the construction of benzo[b]thiophenefused heterocycles in recent years.Among these methods,the domino reaction of benzo[b]thiophene derivatives is a practical and powerful synthetic route to access benzo[b]thiophene-fused heterocycles by virtue of the particularity of sulfur atom.This review summarizes the latest developments in the construction of benzo[b]thiophene-fused heterocycles by ring formation at the C2-C3-position of benzo[b]thiophene derivatives in the past decade.Additionally,this review is divided into four parts according to the four kinds of benzo[b]thiophene derivatives used,including thioaurone,thioisatin,substituted benzo[b]thiophene,and azadiene.