Large-scale Data Research Articles

Next-visit prediction and prevention of hypertension using large-scale routine health checkup data

This paper proposes the use of machine learning models to predict one’s risk of having hypertension in the future using their routine health checkup data of their current and past visits to a health checkup center. The large-scale and high-dimensional dataset used in this study comes from MJ Health Research Foundation in Taiwan. The training data for models is separated into 5 folds and used to train 5 models in a 5-fold cross validation manner. While predicting the results for the test set, the voted result of 5 models is used as the final prediction. Experimental results show that our models achieve 69.59% of precision, 77.90% of recall, and 73.51% of F1-score, which outperforms a baseline using only the blood pressure of visitors’ last visits. Experiments also show that a visitor who performs a health checkup more often can be predicted better, and models trained with selected important factors achieve better results than those trained with Framingham risk score. We also demonstrate the possibility of using our models to suggest visitors for weight control by adding virtual visits that assume their body weight can be reduced in the near future to model input. Experimental results show that around 5.48% of the people who are with high Body Mass Index of the true positive cases are rejudged as negative, and a rising trend appears when adding more virtual visits, which may be used to suggest visitors that controlling their body weight for a longer time lead to lower probability of having hypertension in the future.

Design of judicial public opinion supervision and intelligent decision-making model based on Bi-LSTM

Fuzzy preference modeling in intelligent decision support systems aims to improve the efficiency and accuracy of decision-making processes by incorporating fuzzy logic and preference modeling techniques. While network public opinion (NPO) has the potential to drive judicial reform and progress, it also poses challenges to the independence of the judiciary due to the negative impact of malicious public opinion. To tackle this issue within the context of intelligent decision support systems, this study provides an insightful overview of current NPO monitoring technologies. Recognizing the complexities associated with handling large-scale NPO data and mitigating significant interference, a novel judicial domain NPO monitoring model is proposed, which centers around semantic feature analysis. This model takes into account time series characteristics, binary semantic fitting, and public sentiment intensity. Notably, it leverages a bidirectional long short-term memory (Bi-LSTM) network (S-Bi-LSTM) to construct a judicial domain semantic similarity calculation model. The semantic similarity values between sentences are obtained through the utilization of a fully connected layer. Empirical evaluations demonstrate the remarkable performance of the proposed model, achieving an accuracy rate of 85.9% and an F1 value of 87.1 on the test set, surpassing existing sentence semantic similarity models. Ultimately, the proposed model significantly enhances the monitoring capabilities of judicial authorities over NPO, thereby alleviating the burden on public relations faced by judicial institutions and fostering a more equitable execution of judicial power.

Study of the application of UAV and LiDAR to topographic survey of mountainous forest areas – Case study of Win 3 Wind Power Plant Project, Huong Hoa District, Quang Tri Province, Vietnam

Abstract Unmanned Aerial Vehicle (UAV) technology in conjunction with aerial photography and aerial Light Detection and Ranging (LiDAR) was used for topographic surveying of the Win 3 Wind Power Project area, a complex mountainous terrain situated in the Huong Hoa district of Quang Tri province, Vietnam with the primary objective to produce a map with a scale of 1:1,000 with a contour interval of 1.0 m, accompanied by 3D point clouds, a Digital Elevation Model (DEM), and an orthomosaic. Employing the UAV LiDAR system of DJI Matrice 300 RTK, we conduct ed survey flights, achieving a high point density laser scanning of 229 points/m2 and capturing ultra-high-resolution imagery at 4.22 cm/pixel. The integration of aerial LiDAR and photographic data results in a rich and detailed information repository regarding the surveyed terrain demonstrating that UAV LiDAR technology is a robust tool for geographic data collection, particularly in challenging terrain conditions in Vietnam. We also evaluated the accuracy of the created topographic map, achieving results that surpass the technical requirements and quality standards of the project. This underscores the potential and effectiveness of UAV LiDAR technology in the field of large-scale geographic data collection and mapping in Vietnam.

A comprehensive study of common and rare genetic variants in spermatogenesis-related loci identifies new risk factors for idiopathic severe spermatogenic failure

Abstract STUDY QUESTION Can genome-wide genotyping data be analysed using a hypothesis-driven approach to enhance the understanding of the genetic basis of severe spermatogenic failure (SPGF) in male infertility? SUMMARY ANSWER Our findings revealed a significant association between SPGF and the SHOC1 gene, and identified three novel genes (PCSK4, AP3B1, and DLK1) along with 32 potentially pathogenic rare variants in 30 genes that contribute to this condition. WHAT IS KNOWN ALREADY SPGF is a major cause of male infertility, often with an unknown aetiology. SPGF can be due to either multifactorial causes, including both common genetic variants in multiple genes and environmental factors, or highly damaging rare variants. Next-generation sequencing (NGS) methods are useful for identifying rare mutations that explain monogenic forms of SPGF. Genome-wide association studies (GWAS) have become essential approaches for deciphering the intricate genetic landscape of complex diseases, offering a cost-effective and rapid means to genotype millions of genetic variants. Novel methods have demonstrated that GWAS datasets can be used to infer rare coding variants that are causal for male infertility phenotypes. However, this approach has not been previously applied to characterize the genetic component of a whole case-control cohort. STUDY DESIGN, SIZE, DURATION We employed a hypothesis-driven approach focusing on all genetic variation identified, using a GWAS platform and subsequent genotype imputation, encompassing over 6 million polymorphisms and a total of 1571 SPGF patients and 2431 controls. Both common (minor allele frequency, MAF > 0.01) and rare (MAF < 0.01) variants were investigated within a total of 1797 loci with a reported role in spermatogenesis. This gene panel was meticulously assembled through comprehensive searches in the literature and various databases focused on male infertility genetics. PARTICIPANTS/MATERIALS, SETTING, METHODS This study involved a European cohort using previously and newly generated data. Our analysis consisted of three independent methods: (1) variant-wise association analyses using logistic regression models, (2) gene-wise association analyses using combined multivariate and collapsing (CMC) burden tests, and (3) identification and characterisation of highly damaging rare coding variants showing homozygosity only in SPGF patients. MAIN RESULTS AND THE ROLE OF CHANCE The variant-wise analyses revealed an association between SPGF and SHOC1-rs12347237 (p = 4.15E-06, OR = 2.66), which was likely explained by an altered binding affinity of key transcription factors in regulatory regions and the disruptive effect of coding variants within the gene. Three additional genes (PCSK4, AP3B1, and DLK1) were identified as novel relevant players in human male infertility using the gene-wise burden test approach (p < 5.56E-04). Furthermore, we linked a total of 32 potentially pathogenic and recessive coding variants of the selected genes to 35 different cases. LARGE SCALE DATA Publicly available via GWAS catalog (Accession number: GCST90239721). LIMITATIONS, REASONS FOR CAUTION The analysis of low-frequency variants presents challenges in achieving sufficient statistical power to detect genetic associations. Consequently, independent studies with larger sample sizes are essential to replicate our results. Additionally, the specific roles of the identified variants in the pathogenic mechanisms of SPGF should be assessed through functional experiments. WIDER IMPLICATIONS OF THE FINDINGS Our findings highlight the benefit of using GWAS genotyping to screen for both common and rare variants potentially implicated in idiopathic cases of SPGF, whether due to complex or monogenic causes. The discovery of novel genetic risk factors for SPGF and the elucidation of the underlying genetic causes provide new perspectives for personalized medicine and reproductive counselling. STUDY FUNDING/COMPETING INTEREST(S) This work was supported by the Spanish Ministry of Science and Innovation through the Spanish National Plan for Scientific and Technical Research and Innovation (PID2020-120157RB-I00) and the Andalusian Government through the research projects of “Plan Andaluz de Investigación, Desarrollo e Innovación (PAIDI 2020)” (ref. PY20_00212) and “Proyectos de Investigación aplicada FEDER-UGR 2023” (ref. C-CTS-273-UGR23). SGM was funded by the previously mentioned projects (ref. PY20_00212 and PID2020-120157RB-I00). AGJ was funded by MCIN/AEI/10.13039/501100011033 and FSE “El FSE invierte en tu futuro” (grant ref. FPU20/02926). IPATIMUP integrates the i3S Research Unit, which is partially supported by the Portuguese Foundation for Science and Technology (FCT), financed by the European Social Funds (COMPETE-FEDER) and National Funds (projects PEstC/SAU/LA0003/2013 and POCI-01-0145-FEDER-007274). SS is supported by FCT funds (10.54499/DL57/2016/CP1363/CT0019), ToxOmics-Centre for Toxicogenomics and Human Health, Genetics, Oncology and Human Toxicology, and is also partially supported by the Portuguese Foundation for Science and Technology (UIDP/00009/2020 and UIDB/00009/2020). SLarriba received support from Instituto de Salud Carlos III (grant: DTS18/00101], co-funded by FEDER funds/European Regional Development Fund (ERDF)-a way to build Europe-), and from “Generalitat de Catalunya” (grant 2021SGR052). SLarriba is also sponsored by the “Researchers Consolidation Program” from the SNS-Dpt. Salut Generalitat de Catalunya (Exp. CES09/020). All authors declare no conflict of interest related to this study.

Abstract B065: FabricaTM: A large-scale data simulation platform isolates tumor signal from cell-free DNA and improves tissue of origin prediction accuracy

Abstract Cell-free DNA (cfDNA) liquid biopsy is a promising non-invasive method for disease detection, but data availability and the complexity of cfDNA composition pose barriers to model development. Cancer prediction models trained on cfDNA data from clinically collected blood samples often struggle to isolate tumor-derived signals from confounding factors, and tissue of origin (TOO) models suffer from small sample sizes of rarer tumor types. To address these challenges, we developed FabricaTM, a platform to generate diverse, large-scale, high-fidelity simulated cfDNA datasets for robust and accurate machine learning (ML) model training. A key advantage of FabricaTM is its ability to enhance ML models by matching confounding variables between non-cancer and cancer samples. We first synthesized age-balanced non-cancer samples, then simulated tumor DNA shedding into circulation from 502 reference biopsy samples across 16 indications, optimizing for final tumor content distributions ranging from 0.001 to 14.2%. All reference non-cancer and biopsy samples consisted of aligned bisulfite sequencing reads from a custom target hybrid capture panel. This approach expanded a non-cancer dataset of 177 samples to 1,212 simulated samples and generated 6,400 simulated cancer cfDNA samples matching the non-cancers in age distribution. We then assessed this dataset for equivalence with clinical data and ability to improve cancer prediction. Non-linear dimensionality reduction and clustering of methylation signal showed simulated samples cluster indistinguishably from 2,290 samples (1,149 non-cancer, 1,141 treatment-naïve cancer) from the CORE-HH clinical study (NCT05435066). A binary cancer prediction model trained on the FabricaTM-generated dataset and evaluated on the CORE-HH dataset showed reduced reliance on age-associated signal (R2=0.15) and increased tuning towards tumor content, with minor cost to sensitivity (<4% at 95% specificity), relative to a model trained and cross-validated on the CORE-HH data (R2=0.60). To evaluate the benefit of these data for TOO prediction, we trained multiclass TOO models using FabricaTM's tumor biopsy reference data with and without the addition of our simulated data for 10 target indication groupings and benchmarked their performance on our clinical dataset. Training with both data types yielded a 10% increase in balanced accuracy. Moreover, peak performance was achieved at different training sample numbers per indication, illustrating FabricaTM's potential to estimate sample size requirements during study design. Our findings suggest FabricaTM-generated data can augment limited datasets to train ML models to identify tumor-specific biomarkers obscured by technical and demographic biases in real-world data. The age-balancing approach is readily applied to other confounders to help models learn true disease signatures. The platform is also adaptable to other diseases and biofluids (e.g., Alzheimer's disease, urine), allowing for extension to diagnostic and screening development beyond cancer and blood across the care spectrum. Citation Format: Kade Pettie, Shiva Farashahi, Jackson Killian, Dorna Kashef, Josh Hubbell, Feras Hantash, Jocelyn Charlton, Kieran Chacko. FabricaTM: A large-scale data simulation platform isolates tumor signal from cell-free DNA and improves tissue of origin prediction accuracy [abstract]. In: Proceedings of the AACR Special Conference: Liquid Biopsy: From Discovery to Clinical Implementation; 2024 Nov 13-16; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(21_Suppl):Abstract nr B065.

Abstract 4123872: Unsupervised clustering analysis of large-scale proteomics data identifies molecular phenogroups of heart failure in a population-based cohort

Background: Heart failure (HF) is a heterogeneous syndrome with high mortality. Previous work characterizing HF subgroups primarily relied on clinical data to cluster patients, an approach constrained by the boundaries of documented clinical features. Clustering based on biomarker data, such as proteomics, may provide insights into underlying molecular processes. However, few studies have used this approach, and all used targeted cardiovascular proteomics panels and a restricted HF ejection fraction group. Hypothesis: We used unsupervised clustering analysis to characterize molecular phenogroups in a population-based HF cohort using a large-scale array of circulating plasma proteins. We hypothesized that distinct proteomic-defined phenogroups would exhibit differences in clinical characteristics, mortality, and protein expression. Methods: We measured 7151 SOMAmers (Slow off-rate modified aptamers) targeting human proteins via the SomaLogic 7K SomaScan assay in 1351 HF patients. After feature selection, 831 SOMAmers were used in k-means clustering of patients via R package “ConsensusClusterPlus.” Phenogroups were compared for clinical characteristics, all-cause mortality, and protein expression. Results: Three phenogroups were identified, exhibiting different clinical characteristics ( Table 1 ) and survival probabilities (5-year survival probability in phenogroup 1: 65% [61%, 68%], phenogroup 2: 45% [40%, 51%], and phenogroup 3: 26% [22%, 30%]). In covariate-adjusted Cox models, membership in phenogroups 2 and 3 was significantly associated with increased rate of mortality compared to phenogroup 1. Phenogroups did not differ by ejection fraction or New York Heart Association class. Pathway overrepresentation analysis of differentially expressed SOMAmers identified pathways associated with inflammatory response and immune cell trafficking. Conclusions: Unsupervised clustering analysis using large-scale proteomics data identified three HF phenogroups with different proteomic signatures, clinical characteristics, and outcomes. Future work is necessary to elucidate how molecular differences contribute to heterogeneity in HF to improve risk stratification beyond current recommended clinical guidelines.

Abstract 4128369: Inpatient Mortality After Catheter Ablation for Atrial Fibrillation Among Heart Failure Patients: A 6-Year Nationwide Analysis

Background: Significant advancements have been made in the safety of catheter ablation for atrial fibrillation (AF) over the past decade. However, large-scale data regarding inpatient mortality among heart failure (HF) patients (HF with reduced ejection fraction [HFrEF] and HF with preserved ejection fraction [HFpEF]) who undergo AF ablation are lacking. Aim: To compare mortality rates and trends between patients with HF (HF group) and without HF (non-HF group) following AF catheter ablation. Methods: The National Inpatient Sample was utilized from January 2015 to December 2020 to identify the study population using the International Classification of Diseases, Ninth and Tenth Revisions, Clinical Modification. The primary outcome was in-hospital mortality for patients with and without HF undergoing ablation for AF, and the secondary outcome was mortality in patients with HFrEF compared with those with HFpEF. Results: We evaluated 142,860 patients hospitalized for AF who underwent catheter ablation. Of these, 42.2% (n=60,334) had HF (29% HFrEF; 71% HFpEF). The HF cohort was older (69.7 years) and predominantly male (61.3%) compared with non-HF (66.7 years, P<.001 and 55.7%, P<0.00, respectively). Overall mortality in the HF group was 3-fold higher (1.6% vs 0.5%, P<.001). Within the HF group, mortality in HFrEF was 1.8 times higher than in HFpEF (2.4% vs 1.3%, P<.001). Over the study period, mortality rates declined in the HF cohort (Fig. 1A), both in HFrEF (from 4.1% to 2.2%, P=.012) and HFpEF (from 2.7% to 1.3%, P=.006) (Fig 1B). In non-HF, mortality rates remained relatively unchanged during the study period (from 1.4% to 1.6%, P=.344). Conclusion: Overall mortality following AF ablation is relatively low. Despite the higher mortality observed in patients with HF, the overall rate declined in both HFrEF and HFpEF over the study period. This reduction in mortality may reflect improvements in HF treatment and ablation techniques. Further research is needed to elaborate on the factors contributing to these trends and to develop targeted interventions to reduce mortality in this high-risk population.

Abstract 4115541: Association of Glucagon-like Peptide 1 Receptor Agonist Use with Incident Atrial Fibrillation

Background: Glucagon-like peptide 1 receptor agonists (GLP-1RAs) have shown efficacy for reducing cardiovascular morbidity and mortality in patients with diabetes mellitus (DM), but conflicting evidence exists regarding their impact on cardiac arrhythmias. Whereas some studies suggested a possible association between GLP-1RA use and arrhythmogenesis, other studies suggested decreased association with atrial fibrillation (AF). Overall, large-scale real-world data evaluating risk of AF in association with GLP-1RAs are lacking. Objective: To compare risk of incident AF after initiation of GLP1-RA vs Dipeptidyl peptidase-4 inhibitors (DPP4i), as active comparators, in patients with DM. Methods: This retrospective propensity score-matched cohort study was conducted within the Veterans Health Administration from fiscal years 2006 to 2021. The study included U.S. veterans aged 35 years or older who initiated either GLP1-RA or DPP4i prescriptions. The primary outcome assessed was a composite outcome of atrial fibrillation (AF), defined as a diagnosis of AF/flutter based on administrative codes or undergoing an AF procedure. Results: Out of 116,235 GLP1-RA users and 217,668 DPP4i users, we propensity score-matched 80,948 pairs, on 88 characteristics, including demographics, comorbidities, vital signs, cardiovascular risk score, healthcare utilization, DM control, laboratory investigation, and medications. Composite outcome of AF was similar in GLP-1RA group (4.1%) and DPP4i group (4.3%); odds ratio (OR) 0.96, 95% confidence interval (CI) 0.92-1.01. Secondary analyses stratified by medication use duration showed no significant differences in composite AF risk (P>0.05). Notably, individuals achieving weight loss of 2%, 5%, or 10% of baseline body weight had significantly lower AF incidence, whereas no significant differences were observed in those with no weight loss or weight gain (OR 0.99, 95% CI 0.91-1.07). Conclusions: Use of GLP1-RA in patients with DM was not associated with a decreased risk of AF when compared with the use of DPP4i. In the subgroup analysis, significantly lower AF risk was seen in the GLP-1 RA group who achieved weight loss vs. DPP4i and this highlights weight loss as a potential modifier of cardiovascular outcomes in DM.

Abstract 4129397: Cardiovascular Risk Models Using Large-Scale Physical Examination Data

Background: Physical examination (PE) data offer more potential key risk factors that can enhance the development of risk models. Additionally, the latest CVD risk model, PREVENT, developed from an American population, necessitates validation within China. Objectives: This study aimed to develop CVD risk models based on large-scale PE records and compare them with PREVENT model in Chinese adults. Methods: Data were collected from a China national database (2016-2024), including 929682 participants aged 30-79 years without previous CVD (Table I). The variables covered demographics, comorbidities, and PE items, resulting in 59 features as input. The endpoint events were defined as total CVD and its subtypes, ASCVD and heart failure (HF). Sex-specific survival analysis was developed based on age-scale, utilizing Cox Proportional Hazards and Random Survival Forests (RSF). Feature importance was assessed using the average minimal depth in RSF. The performance was evaluated by the Harrell C-statistic in the validation set. Results: The results showed that our RSF model achieved the best performance in total CVD, with the highest C-index of 0.82 (Table II). The top 15 predictive risk factors for each gender are presented in Figure 1. Additionally, validation of the PREVENT model in our Chinese population dataset showed its great generalizability, particularly for HF events. Conclusions: Overall, this research underscores the importance of leveraging large-scale PE data and advanced modeling techniques to improve CVD risk prediction. Beyond traditional risk factors, greater emphasis should be placed on more demographic characteristics, comorbidities, urinalysis, and ECG features.

Abstract 4131412: Effects of Angiotensin Converting Enzyme Inhibition And Afterload Reduction in Single Ventricle Patients Prior To Fontan: Not What You Might Think

Introduction: Pts with single ventricle (SV) after bidirectional Glenn (BDG) are placed on angiotensin converting enzyme inhibitors (ACEI) to reduce afterload, improve ventricular performance and allow for decreased systemic to pulmonary collateral flow (APC). No large scale data exists to support ACEI use in BDG pts to accomplish this. We hypothesized that ACEI may not perform as predicted. Goals: To determine if ACEI use in BDG pts decreases APC and improves ventricular function. Methods: Single center retrospective study of all BDG pts who underwent cardiac magnetic resonance (CMR) from 2010 to present. Demographics, medication, ventricular function and flow data were collected. Pts included were >1 year old and on ACEI for >3 months prior to CMR. Wilcoxon and Fisher Exact tests were used. Significance P<0.05 Results: Of the 320 BDG pts (median 2.8 years old) studied, 137 (43%) were treated with ACEI. See table for all data. There was no difference in age, sex, medications (except for ACEI) or body surface area (BSA) between both groups. APC as a % of aortic flow and APC/BSA were not different between those treated with ACEI compared to those who were not, whether right (RV) or left ventricle (LV) dominant. There was no correlation of APC with time on ACEI or ACEI dosage. Ejection fraction (EF) was lower and end-diastolic volume (EDV) was higher in BDG on ACEI for the entire group and RVs. Although LV EDV was higher with those on ACEI, LV EF was the same for those on ACEI and those who weren’t (graph). APC as a % of aortic flow and APC/BSA positively correlated with EDV (rho 0.25-0.34, P<.001). There was no difference in atrioventricular valve regurgitation between those on ACEI and those who were not (median 5 vs 6% respectively). As a surrogate for cerebral blood flow (CBF), superior vena caval (SVC) flow was lower for those on ACEI compared to those who were not. Conclusion: BDG pts treated with ACEI did not demonstrate decreased APC or improved ventricular performance. There is a suggestion that ventricular performance may be worse on ACEI. CBF is decreased in BDG on ACEI using SVC flow as a surrogate. This suggests that ACEI use in SV pts at the BDG stage is not indicated and should be discouraged.

Machine learning-based crop type detection and vegetation mon-itoring using Sentinel-1 SAR and Landsat 8

This study explores the use of machine learning for crop type detection and vegetation monitoring in arid and semi-arid regions, specifically in Biskra and Khenchela, Algeria. Traditional field-based agricultural monitoring is labor-intensive and limited in spatial coverage, prompting the need for more efficient and scalable methods. The research integrates Synthetic Aperture Radar (SAR) data from Sentinel-1 and optical imagery from Landsat 8 to enhance the classification of different crop types and track vegetation dynamics. Key metrics such as radar backscatter and Normalized Difference Vegetation Index (NDVI) are employed to distinguish between irrigated and rainfed crops. The Random Forest algorithm is utilized for classification, with training data derived from field surveys, high-resolution satellite imagery, and existing land cover maps. The Google Earth Engine (GEE) platform facilitates large-scale data processing, providing real-time agricultural analysis. Over the 12-month study period (January to December 2020), the integrated approach demonstrated high accuracy, achieving over 85% precision in classifying key crops such as date palms, cereals, and vegetables. Results indicate that SAR and NDVI data provide complementary insights into vegetation health, phenology, and agricultural practices in different climatic zones. This methodology offers a promising solution for improving agricultural monitoring and resource management in arid and semi-arid regions, enhancing food security and optimizing water usage. The study’s findings also underscore the value of satellite-based remote sensing for large-scale, real-time agricultural analysis.

A INTELIGÊNCIA ARTIFICAL COMO FERRAMENTA DE EFICIÊNCIA NA ADMINISTRAÇÃO PÚBLICA

The article “Artificial Intelligence as a Tool for Efficiency in Public Administration” explores the growing role of AI in optimizing government operations and improving the delivery of public services. Focusing on AI’s capabilities to process and analyze large volumes of data, the article discusses how this technology can be used to enhance efficiency, transparency, and responsiveness in government services. The text highlights several key applications of AI in public administration, including large-scale data analysis and processing, automation of bureaucratic processes, and customization of services to citizens. These technologies, equipped with machine learning and natural language processing, enable better resource management, faster citizen service, and more efficient operations less susceptible to human errors. Beyond the benefits, the article also addresses the challenges associated with implementing AI in the public sector, such as technical, ethical, and privacy issues. Concerns about algorithmic bias, data quality, and integration with legacy systems are discussed in detail. The article emphasizes the need for robust policies and adequate training for public employees as essential for successful AI integration. Finally, the discussion concludes with the view that while AI has the potential to significantly transform public administration for the better, a careful and regulated approach is vital. The success of this integration depends not only on technological advancements but also on a commitment to ethical practices and the development of public policies that guide its responsible use.

BIOM-20. EVALUATING THE POTENTIAL OF PSMA TARGETING IN CNS TUMORS: INSIGHTS FROM LARGE-SCALE TRANSCRIPTOME PROFILING

Abstract PURPOSE Prostate-specific membrane antigen (PSMA) is a well-established target in prostate cancer therapy that has shown potential as a theranostic target across non-central nervous system (CNS) and CNS tumor types. We aimed to investigate the pan-tissue expression pattern of the PSMA-encoding gene FOLH1 to assess whether transcriptome profiling can inform tumor theranostics. EXPERIMENTAL DESIGN We assessed FOLH1 expression from the Open Pediatric Cancer Project (OpenPedCan, n = 2132 specimens), The Cancer Genome Atlas (TCGA, n = 10411 specimens), and the Genotype Tissue Expression Project (GTEx, n = 17382 specimens) in relation to published reports of PSMA radionuclide uptake. RESULTS When comparing FOLH1 expression across tumor versus normal tissues, we found that non-CNS tumors exhibiting elevated expression of at least two-fold (FDR < 0.05) were reported to have successful PSMA radionuclide uptake in contrast to tumors with less than a two-fold elevation or with lower expression of FOLH1 relative to normal. Notably, CNS tumors universally exhibited lower expression of FOLH1 relative to normal brain tissue, but we observed considerable variation in the expression of blood-tumor barrier (BTB) components associated with reports of BTB integrity and uptake of PSMA radiotracers. CONCLUSIONS Large-scale transcriptomics data may help guide the application of PSMA-based radionuclide therapies in non-CNS tumors, but care should be taken to account for BTB effects in CNS tumors when assessing the potential for radionuclide success. Our analysis demonstrated that FOLH1 showed a lack of tumor-specific expression for both adult and pediatric CNS tumors when compared to normal brain tissue, suggesting that PSMA is not a desirable target in this disease type. This study highlights the utility of utilizing large-scale molecular data to guide the selection of theranostic targets.

Pseudo-Labeling and Time-Series Data Analysis Model for Device Status Diagnostics in Smart Agriculture

This study proposes an automated data-labeling model that combines a pseudo-labeling algorithm with waveform segmentation based on Long Short-Term Memory (LSTM) to effectively label time-series data in smart agriculture. This model aims to address the inefficiency of manual labeling for large-scale data generated by agricultural systems, enhancing the performance and scalability of predictive models. Our proposed method leverages key features of time-series data to automatically generate labels for new data, thereby improving model accuracy and streamlining data processing. By automating the labeling process, we reduce dependence on manual labeling, which is often labor-intensive and prone to errors in large datasets. This approach enables the efficient preparation of labeled data for applications such as anomaly detection, pattern recognition, and predictive modeling in smart agriculture. Experimental results demonstrate that the automated labeling model achieves 89% accuracy in agricultural environments and reduces data processing time by 30%. Future research will focus on extending the model’s applicability to diverse agricultural settings, enhancing generalization performance, and improving real-time processing capabilities, thereby advancing intelligent and sustainable smart agriculture systems.

Novel application of metabolic imaging of early embryos using a light-sheet on-a-chip device: a proof-of-concept study

Abstract STUDY QUESTION Is it feasible to safely determine metabolic imaging signatures of nicotinamide adenine dinucleotide [NAD(P)H] associated auto-fluorescence in early embryos using a light-sheet on-a-chip approach? SUMMARY ANSWER We developed an optofluidic device capable of obtaining high-resolution 3D images of the NAD(P)H autofluorescence of live mouse embryos using a light-sheet on-a-chip device as a proof-of-concept. WHAT IS KNOWN ALREADY Selecting the most suitable embryos for implantation and subsequent healthy live birth is crucial to the success rate of assisted reproduction and offspring health. Besides morphological evaluation using optical microscopy, a promising alternative is the non-invasive imaging of live embryos to establish metabolic activity performance. Indeed, in recent years, metabolic imaging has been investigated using highly advanced microscopy technologies such as fluorescence-lifetime imaging and hyperspectral microscopy. STUDY DESIGN, SIZE, DURATION The potential safety of the system was investigated by assessing the development and viability of live embryos after embryo culture for 67 h post metabolic imaging at the two-cell embryo stage (n = 115), including a control for culture conditions and sham controls (system non-illuminated). Embryo quality of developed blastocysts was assessed by immunocytochemistry to quantify trophectoderm and inner mass cells (n = 75). Furthermore, inhibition of metabolic activity (FK866 inhibitor) during embryo culture was also assessed (n = 18). PARTICIPANTS/MATERIALS, SETTING, METHODS The microstructures were fabricated following a standard UV-photolithography process integrating light-sheet fluorescence microscopy into a microfluidic system, including on-chip micro-lenses to generate a light-sheet at the centre of a microchannel. Super-ovulated F1 (CBA/C57Bl6) mice were used to produce two-cell embryos and embryo culture experiments. Blastocyst formation rates and embryo quality (immunocytochemistry) were compared between the study groups. A convolutional neural network (ResNet 34) model using metabolic images was also trained. MAIN RESULTS AND THE ROLE OF CHANCE The optofluidic device was capable of obtaining high-resolution 3D images of live mouse embryos that can be linked to their metabolic activity. The system’s design allowed continuous tracking of the embryo location, including high control displacement through the light-sheet and fast imaging of the embryos (<2 s), while keeping a low dose of light exposure (16 J · cm−2 and 8 J · cm−2). Optimum settings for keeping sample viability showed that a modest light dosage was capable of obtaining 30 times higher signal-noise-ratio images than images obtained with a confocal system (P < 0.00001; t-test). The results showed no significant differences between the control, illuminated and non-illuminated embryos (sham control) for embryo development as well as embryo quality at the blastocyst stage (P > 0.05; Yate’s chi-squared test). Additionally, embryos with inhibited metabolic activity showed a decreased blastocyst formation rate of 22.2% compared to controls, as well as a 47% reduction in metabolic activity measured by metabolic imaging (P < 0.0001; t-test). This indicates that the optofluidic device was capable of producing metabolic images of live embryos by measuring NAD(P)H autofluorescence, allowing a novel and affordable approach. The obtained metabolic images of two-cell embryos predicted blastocyst formation with an AUC of 0.974. LARGE SCALE DATA N/A. LIMITATIONS, REASONS FOR CAUTION The study was conducted using a mouse model focused on early embryo development assessing illumination at the two-cell stage. Further safety studies are required to assess the safety and use of 405 nm light at the blastocyst stage by investigating any potential negative impact on live birth rates, offspring health, aneuploidy rates, mutational load, changes in gene expression, and/or effects on epigenome stability in newborns. WIDER IMPLICATIONS OF THE FINDINGS This light-sheet on-a-chip approach is novel and after rigorous safety studies and a roadmap for technology development, potential future applications could be developed for ART. The overall cost-efficient fabrication of the device will facilitate scalability and integration into future devices if full-safety application is demonstrated. STUDY FUNDING/COMPETING INTEREST(S) This work was partially supported by an Ideas Grant (no 2004126) from the National Health and Medical Research Council (NHMRC), by the Education Program in Reproduction and Development (EPRD), Department Obstetrics and Gynaecology, Monash University, and by the Department of Mechanical and Aerospace Engineering, Faculty of Engineering, Monash University. The authors E.V-O, R.N., V.J.C., A.N., and F.H. have applied for a patent on the topic of this technology (PCT/AU2023/051132). The remaining authors have nothing to disclose.

Analysis methods for large-scale neuronal recordings.

Simultaneous recordings from hundreds or thousands of neurons are becoming routine because of innovations in instrumentation, molecular tools, and data processing software. Such recordings can be analyzed with data science methods, but it is not immediately clear what methods to use or how to adapt them for neuroscience applications. We review, categorize, and illustrate diverse analysis methods for neural population recordings and describe how these methods have been used to make progress on longstanding questions in neuroscience. We review a variety of approaches, ranging from the mathematically simple to the complex, from exploratory to hypothesis-driven, and from recently developed to more established methods. We also illustrate some of the common statistical pitfalls in analyzing large-scale neural data.

Machine Learning-Based Financial Big Data Analysis and Forecasting: From Preprocessing to Deep Learning Models

Abstract: Machine learning technology has revolutionized the financial sector by allowing faster and more accurate analysis and forecasting of large-scale financial data. This paper focuses on how machine learning ( ML ), especially deep learning models, can help to deal with high dimensional, noisy, and non-stationary financial data. Essential methodologies such as data preprocessing, feature engineering, and dimensionality reduction are imperative for preparing the raw financial data to ML algorithms. Methods such as outlier detection, normalisation for preprocessing, and feature (variable) selection for dimensionality reduction improve the models accuracy and efficiency. The paper also examines how deep learning models, such as Recurrent Neural Networks ( RNN ) and Long Short-Term Memory ( LSTM ) networks, can overcome the issues of autoregressive integrated moving average ( ARIMA ) models for financial time series prediction. An in-depth comparison of the machine learning models, ranging from supervised to unsupervised methods, is also provided to discuss their pros and cons in the financial domain, including popular applications such as credit scoring, fraud detection, and market risk prediction. The study finally concludes by discussing how optimisation methods such as hyperparameter tuning and cross-validation are imperative for ML models in complex financial scenarios to ensure their generalisation capability and avoid overfitting.

Employing Xception convolutional neural network through high-precision MRI analysis for brain tumor diagnosis

The classification of brain tumors from medical imaging is pivotal for accurate medical diagnosis but remains challenging due to the intricate morphologies of tumors and the precision required. Existing methodologies, including manual MRI evaluations and computer-assisted systems, primarily utilize conventional machine learning and pre-trained deep learning models. These systems often suffer from overfitting due to modest medical imaging datasets and exhibit limited generalizability on unseen data, alongside substantial computational demands that hinder real-time application. To enhance diagnostic accuracy and reliability, this research introduces an advanced model utilizing the Xception architecture, enriched with additional batch normalization and dropout layers to mitigate overfitting. This model is further refined by leveraging large-scale data through transfer learning and employing a customized dense layer setup tailored to effectively distinguish between meningioma, glioma, and pituitary tumor categories. This hybrid method not only capitalizes on the strengths of pre-trained network features but also adapts specific training to a targeted dataset, thereby improving the generalization capacity of the model across different imaging conditions. Demonstrating an important improvement in diagnostic performance, the proposed model achieves a classification accuracy of 98.039% on the test dataset, with precision and recall rates above 96% for all categories. These results underscore the possibility of the model as a reliable diagnostic tool in clinical settings, significantly surpassing existing diagnostic protocols for brain tumors.

Performance Optimization in Distributed SQL Environments : A Comprehensive Analysis of Presto Query Engine

This comprehensive article examines performance optimization techniques for Presto, a distributed SQL query engine widely adopted for large-scale data analytics. Through systematic analysis of both theoretical frameworks and empirical evidence, we present a multifaceted approach to enhancing query performance and resource utilization. The article encompasses critical aspects including memory management strategies, parallel processing optimization, and storage connector configurations across diverse deployment scenarios. Our investigation reveals that strategic implementation of query planning algorithms, coupled with fine-tuned JVM configurations, can yield performance improvements of up to 40% in complex analytical workloads. The article also introduces a novel framework for workload-specific optimization patterns, validated through extensive testing across various data scales and query complexities. Through detailed case studies of large-scale deployments, we demonstrate how combined optimization techniques can significantly reduce query latency while maintaining system stability. Furthermore, we present empirical evidence supporting the effectiveness of adaptive resource allocation strategies in mixed-workload environments. These findings contribute to the growing body of knowledge in distributed query processing optimization and provide practical guidelines for organizations seeking to enhance their Presto deployments.

Comparison Distributed and Parallel Machine Learning: Evidence from Models, Principles and Application Scenarios

Abstract. Contemporarily, the demand for processing large-scale data has been rapidly increasing, prompting continuous advancements in the field of machine learning. This study examines the state of parallel and distributed machine learning at present, both of which aim to enhance computational efficiency when dealing with large datasets and demonstrate promising applications across various domains. As the scale of data escalates, traditional single-machine learning methods are becoming increasingly inadequate, which lead to the emergence of parallel and distributed machine learning. These approaches enable substantial computations to be performed more efficiently through the collaboration of multi-core CPUs or multiple computing nodes. This research conducts an in-depth analysis of the inherent challenges associated with these methods, including data transmission latency, synchronization requirements, and user privacy concerns. Ultimately, this research emphasizes the tremendous potential of both methods for future applications, a potential that is bolstered by ongoing advancements in hardware and algorithm optimization. These results provide valuable insights for practitioners in the field and offers guidance for future research directions.