Big Database Research Articles

The financial benefits of health engagement programs to life insurers

Life insurance companies, as equity stakeholders in policyholders’ lives, have incentives to mitigate their health risks. I introduce a framework that enables life insurers to evaluate the financial viability of developing and implementing health engagement programs. By leveraging a proprietary big database of health and mortality information from a large U.S. life insurer, I use machine learning techniques to quantify the benefits and use a rational addiction model to calculate the costs associated with these programs. The estimated net benefit available to the life insurer from the smoking cessation program is USD 87 million and the aggregate benefit from including other chronic conditions is USD 872 million. I explore the broader application of this framework in a general health policy context.

Правовые особенности определения баз данных для их свободного использования

The article examines the normative concept of databases, as well as their differentiation from other materials. Particular attention is paid to the controversy about the attribution to big data databases, the consequences of such an action and the prospects for the development of civil legislation. The author assesses the legal definition of databases with regard to its completeness and compliance with modern legal realities. Besides, the author touches upon the importance of determining the content of databases, criteria for their free use by citizens for non-commercial purposes. It is suggested to expand the possibilities of free use of databases beyond personal purposes. An attempt is made to determine the balance between copyright protection and society's need for free use of databases, especially in the context of digitalization, when copying and use of significant amounts of data is becoming more and more accessible. The study emphasizes the importance of enabling society to use information from databases to facilitate search and access to knowledge, especially when such databases become the main sources of information in a particular field.

Cleaning method for abnormal energy big data based on sparse self-coding

In order to reduce the interference of abnormal energy big data and improve the accuracy of anomaly cleaning and detection, this paper proposes a method for cleaning abnormal energy big database based on sparse self-coding. Firstly, the abnormal data detection method based on multi-criteria evaluation is used to analyze the spectral feature distribution of abnormal energy big data. By using chaotic time series reconstruction model, robust local weighted regression analysis and sparse self-coding method, the feature decomposition of time series of energy data is realized. Then, according to the periodicity of the original sequence, the original sequence is segmitated adaptively to represent the morphological characteristics of the abnormal energy data sub-sequences driven by dynamic carbon emissions, and the anomaly index of each sub-sequence is obtained by AFCM algorithm. Secondly, an energy anomaly evaluation model based on LOF value is established. Finally, the output decision function of data cleaning is constructed to realize the abnormal energy big data cleaning. The test results show that the error detection rate of this method is 0.24%, the missing detection rate is 0.27%, the cleaning rate can reach 99.49%, and the cleaning time is less than 2s.

Data-driven multiscale geomechanical modeling of unconventional shale gas reservoirs: a case study of Duvernay Formation, Alberta, West Canadian Basin

The Duvernay Formation in Canada is one of the major oil and gas source formations in the Western Canadian Sedimentary Basin, located at its deepest point. While it demonstrates promising development potential, challenges arise in the urgent need for integration of geology and engineering models, as well as in optimizing sweet spots, particularly as infill wells and pads become central operational objectives for the shale gas field. A lack of the geomechanical understanding of shale gas reservoirs presents a significant obstacle in addressing these challenges. To overcome this, we implemented data acquisition and prepared historical models and profiles, resulting in an extended high-resolution geological and reservoir property model with a fine grid system. Subsequently, a 3D full-field multi-scale geomechanical model was constructed for the main district by integrating seismic data (100 m), geological structures (km), routine logs (m), core data (cm), and borehole imaging (0.25 m), following a well-designed workflow. The predicted fracturability index (brittleness) ranges from 0.6 to 0.78, and a lower horizontal stress difference (STDIFF) is anticipated in the target formation, Upper Duvernay_D, making it a favorable candidate for hydraulic fracturing treatment. Post-analysis of the multi-disciplinary models and various data types provides guidelines for establishing a specific big database, which serves as the foundation for production performance analysis and aggregate sweet spot analysis. Fourteen geological and geomechanical candidate parameters are selected for the subsequent sweet spot analysis. This study highlights the effectiveness of multi-scale geomechanical modeling as a tool for the integration of multi-disciplinary data sources, providing a bridge between geological understanding and future field development decisions. The workflows also offer a data-driven framework for selecting parameters for sweet spot analysis and production dynamic analysis.

A Study on Historical Big Data Analysis of Surface Ecological Damage in the Coal Mining Area of Lvliang City Based on Two Mining Modes

Coal mining inevitably causes damage to the surface ecological environment. In response to the characteristics of multiple factors, wide scope, and long time series of surface ecological environment damage in coal mining subsidence areas, how to integrate multiple data sources and monitoring methods to achieve monitoring and trend analysis of ecological damage throughout the entire mining cycle still needs to be studied. In addition, the 110 mining method provides an innovative method for underground coal mining to reduce its surface ecological damage, but the differences in surface damage between the two mining modes and the effectiveness of the 110 method in realizing surface ecological damage-reducing mining still need to be studied in depth. Therefore, this study takes the surface ecological damage in the mining area of Lvliang City, a typical resource city in Shanxi Province, China, as the object. It establishes a four-in-one “Satellite–UAV–Ground–Underground” information monitoring method, proposes a historical big data evolution analysis method, constructs three spatial scales of historical big databases, clarifies the current situation and development trend of damage in coal mining areas in Lvliang City and explores the differences in surface ecological environment damage characteristics in coal mining areas based on the 121 and 110 mining methods. The results show that: (1) The existing coal mining subsidence area in Lvliang City is as high as 92,191.47 hectares, and it is expected to continue to increase to 130,739.55 hectares in the future 2035, with a growth rate of 41.812%, which realizes the goals of mapping the current situation, retracing the history and predicting the future of the ecological damage of the surface in Lvliang City. (2) The surface NDVI of the 110 working face is restored to the average level of the mining area faster than that of the 121 working face. The surface crack width, step displacement, length, distribution density, and surface settlement height of the 110 working face are all smaller than those of the 121 working face. It has been verified that the unique top-cutting and swelling filling effect of the 110 methods can effectively reduce the ecological damage caused by coal mining subsidence. And its widespread application can effectively realize the ecological environmental protection of the coal mine area and contribute to the high-quality development of the coal industry in Lvliang City.

258. SHORT-TERM OUTCOMES AFTER ESOPHAGECTOMY ANALYZED USING NATIONAL CLINICAL DATABASE IN JAPAN

Abstract Esophagectomy for esophageal cancer is a highly invasive procedure in gastrointestinal surgery. The National Clinical Database (NCD) of Japan starting data collection in 2011 has accumulated real world data of esophagectomy, one of the main nine gastroenterological surgeries for which detailed input items have been required for the evaluation of surgical mortality and postoperative complications. We herein reviewed the short-term outcomes after esophagectomy analyzed by Japanese nationwide big database. Many studies on esophagectomy, including investigations of quality indicators, feasibility of minimally invasive surgery, and various patients’ conditions, have been conducted using NCD. The validation of certification systems by the Japanese Society of Gastroenterological Surgery (JSGS) and the Japan Esophageal Society (JES) revealed that the institute’s certification had greater influence on short-term surgical outcomes than the operating surgeon’s certification. Minimally invasive esophagectomy (MIE) can replace open esophagectomy (OE). MIE performed with no preoperative treatment was associated with significantly less incidence of any pulmonary morbidities, prolonged ventilation 48hours, unplanned intubation, surgical site infection, and sepsis. The total surgery-related mortality rates of MIE and OE were 1.7% and 2.4%, respectively (P < 0.001). The operative mortality in the elderly group with residual tumors increased to more than twice than that in the non-elderly groups (15.9 vs. 5.5 or 6.5%) and was much higher than that in elderly patients without residual tumors (15.9 vs. 4.6%). Careful selection is important for the treatment of elderly patients with highly advanced tumors to avoid unfavorable short-term outcomes. Thus, the NCD has contributed to quality assessment and improvement of surgery by feedback of accurate data in Japan.

CONCEPTUAL METAPHORS OF ARTIFICIAL INTELLIGENCE AND AI DEVELOPMENT IN THE GUARDIAN NEWSPAPER

The whirlwind advent of ChatGPT in 2022 has marked a new age of artificial intelligence (AI), the general name for the technology that combines computer technology, big data bases and machines. This AI technology quickly makes its presence felt with hundreds of popular programs and chatbots such as the portrait-making AI diffusion art and the thesis-writing ChatGPT. This paper investigates the conceptual metaphors representing AI and AI development in The Guardian, a UK-based newspaper, to figure out how this technology and its growth have been introduced to ordinary people via mass media. Employing the Conceptual Metaphor Theory proposed by Lakoff & Johnson (1980), this study found three AI-related conceptual metaphors, namely, AI IS A HUMAN BEING, AI IS AN ANIMAL and AI IS A NATURAL FORCE, which are realized by more than 100 linguistic expressions across 33 news articles. Also, this research found five conceptual metaphors related to AI development, namely AI DEVELOPMENT IS WAR, AI DEVELOPMENT IS A RACE, AI DEVELOPMENT IS A CONVERSATION, AI DEVELOPMENT IS A DANCE, AI DEVELOPMENT IS A GAME and these metaphors are manifested by approximately 40 linguistic expressions. This paper discusses the way that these metaphors could influence the way people and technology companies think about AI and AI development.

Outcome research on esophagectomy analyzed using nationwide databases in Japan: evidences generated from real-world data.

Esophagectomy for esophageal cancer is a highly invasive gastrointestinal surgical procedure. The National Clinical Database (NCD) of Japan, initiated in 2011, has compiled real-world data on esophagectomy, one of nine major gastroenterological surgeries. This review examines outcomes after esophagectomy analyzed using the Japanese big databases. Certification systems by the Japanese Society of Gastroenterological Surgery (JSGS) and the Japan Esophageal Society (JES) have shown that institutional certification has a greater impact on short-term surgical outcomes than surgeon certification. Minimally invasive esophagectomy has emerged as a viable alternative to open esophagectomy, although careful patient selection is crucial, especially for elderly patients with advanced tumors. The NCD has significantly contributed to the assessment and enhancement of surgical quality and short-term outcomes, while studies based on Comprehensive Registry of Esophageal Cancer in Japan (CRECJ) have provided data on patient characteristics, treatments, and long-term outcomes. The JES has conducted various questionnaire-based retrospective clinical reviews in collaboration with authorized institutions certified by JES. The Diagnosis Procedure Combination (DPC) database provides administrative claims data including itemized prices for surgical, pharmaceutical, laboratory, and other inpatient services. Analyzing these nationwide databases can offer precise insights into surgical quality for esophageal cancer, potentially leading to improved treatment outcomes.

Artificial intelligence enabled smart design and manufacturing of advanced materials: The endless Frontier in AI+ era

AbstractFuture‐oriented Science & Technology (S&T) Strategies trigger the innovative developments of advanced materials, providing an envision to the significant progress of leading‐/cutting‐edge science, engineering, and technologies for the next few decades. Motivated by Made in China 2025 and New Material Power Strategy by 2035, several key viewpoints about automated research workflows for accelerated discovery and smart manufacturing of advanced materials in terms of AI for Science and main respective of big data, database, standards, and ecosystems are discussed. Referring to classical toolkits at various spatial and temporal scales, AI‐based toolkits and AI‐enabled computations for material design are compared, highlighting the dominant role of the AI agent paradigm. Our recent developed ProME platform together with its functions is introduced briefly. A case study of AI agent assistant welding is presented, which is consisted of the large language model, auto‐coding via AI agent, image processing, image mosaic, and machine learning for welding defect detection. Finally, more duties are called to educate the next generation workforce with creative minds and skills. It is believed that the transformation of knowledge‐enabled data‐driven integrated computational material engineering era to AI+ era promotes the transformation of smart design and manufacturing paradigm from “designing the materials” to “designing with materials.”

A sensitive system based on radon amplification at soil-air interface: Aiming to advance earthquake precursor research

Radon, a natural radioactive gas, serves as a valuable tracer in geophysical research and atmospheric science such as detecting stress induced signal in bedrock. However, the conventional radon monitoring methods often lack the sensitivity required to accurately capture such signals. This limitation, coupled with interference from meteorological effects, poses challenges in distinguishing genuine stress-induced signals. In this study, we propose a novel approach utilizing radon concentration gradients at the soil-air interface to enhance sensitivity and detect stress induced radon signals more effectively. Drawing from pressure diffusion models, we demonstrate how seismic stress accumulation in bedrock alters radon profiles in the sub-soil, providing insights into the mechanisms underlying stress-induced radon variations. Building upon this theoretical framework, we introduce the “Bhabha Radon Observatory for Seismic Application (BhaROSA)," a remote sensing, solar-powered radon observatory designed for widespread deployment and continuous unattended monitoring for big database generation. Field experiments comparing BhaROSA's performance to conventional soil probe techniques validate and confirm the superior sensitivity in line with theoretical predictions. This innovative approach holds promise for improving our understanding of stress dynamics in bedrock and has potential applications in various geophysical and atmospheric science such as earthquake precursory research, geo-genic radon potential and risk assessment. To progress, we propose international alliance and application of deep learning to a big database of precursor signals, which may lead to more informed conclusions on earthquake predictability-an enduring and unsolved challenge for humanity.

Seventy Years of Dose-response Models: From the Target Theory to the Use of Big Databases Involving Cell Survival and DNA Repair.

Radiobiological data, whether obtained at the clinical, biological or molecular level has significantly contributed to a better description and prediction of the individual dose-response to ionizing radiation and a better estimation of the radiation-induced risks. Particularly, over the last seventy years, the amount of radiobiological data has considerably increased, and permitted the mathematical formulas describing dose-response to become less empirical. A better understanding of the basic radiobiological mechanisms has also contributed to establish quantitative inter-correlations between clinical, biological and molecular biomarkers, refining again the mathematical models of description. Today, big data approaches and, more recently, artificial intelligence may finally complete and secure this long process of thinking from the multi-scale description of radiation-induced events to their prediction. Here, we reviewed the major dose-response models applied in radiobiology for quantifying molecular and cellular radiosensitivity and aimed to explain their evolution: Specifically, we highlighted the advances concerning the target theory with the cell survival models and the progressive introduction of the DNA repair process in the mathematical models. Furthermore, we described how the technological advances have changed the description of DNA double-strand break (DSB) repair kinetics by introducing the important notion of DSB recognition, independent of that of DSB repair. Initially developed separately, target theory on one hand and, DSB recognition and repair, on the other hand may be now fused into a unified model involving the cascade of phosphorylations mediated by the ATM kinase in response to any genotoxic stress.

Role Of Artificial Intelligence in Big Database Management

This exploration article digs into the significant effect of (artificial intelligence) AI, on big Data base management in the field of computer science by utilizing the broad abilities of artificial intelligence AI technologies, this study investigates different strategies to improve the oversight and administration of huge information bases. Through an exhaustive assessment of existing academic works, itemized contextual investigations, and master experiences, the article disentangles the mind-boggling combination of Ai’s integration and its important impacts on database management rehearses. Covering an extensive variety of artificial intelligence aspects, for example, AI, regular language handling, and profound realizing, this academic request plans to explain the complex job of computer-based intelligence in the domain of big database management. As a critical commitment to scholarly talk, this research article fills in as an original work, planning the groundbreaking excursion of Ai transformative into the center of data base management system, introducing another time of development and productivity in the computerized scene.

BF-BigGraph: An efficient subgraph isomorphism approach using machine learning for big graph databases

Graph databases are flexible NoSQL databases used to efficiently store and query complex and big data. One of the most difficult problems in graph databases is the problem of subgraph isomorphism, which involves finding a matching pattern in a given graph. Subgraph isomorphism algorithms generally encounter problems in the efficient processing of complex queries based on a lack of pruning methods and the use of a matching order. In this study, we present a new subgraph isomorphism approach based on the best-first search design strategy and name it BF-BigGraph. Our approach includes a machine learning technique to efficiently find the best matching order for various complex queries. The parameters we used in our approach as heuristics to improve the performance of complex queries on graph-based NoSQL databases are database volatility, database size, type of query, and the size of the query. We utilized the Random Forest machine learning method to narrow candidate nodes to a higher level of search and effectively reduce the search space for efficient querying and retrieval. We compared BF-BigGraph with state-of-the-art approaches, namely BB-Graph, Neo4j’s Cypher, DualIso, GraphQL, TurboIso, and VF3 using publicly available databases including undirected graphs; WorldCup, Pokec, Youtube, and a big graph database of a real demographic application (a population database) with approximately 70 million nodes of a big directed graph. The performance results of our approach for different types of complex queries on all these databases are significantly better in terms of computation time and required memory than other competing approaches in the literature.

New data reduction algorithms based on the fusion of instance and feature selection

This paper presents two novel data reduction algorithms that combine instance selection and feature selection methods to simultaneously reduce the number of records and features. These algorithms are rigorously defined and thoroughly explained. They are applied to reduce twelve data sets selected from a commonly available database repository and single representative of big database. Their reduction efficiency is tested using three well-known machine learning models: classification and regression tree, convolutional neural network, and support vector machines. The results are extensively analyzed, comparing the classification performance of the models on the reduced data sets with that on the original ones. The computational complexity of algorithms is theoretically estimated and practically verified. The obtained performance results are compared with those available in the literature. It is shown that implementing the proposed data reduction algorithms consistently improves model accuracy in the vast majority of classification scenarios.

A Performance Analysis of CassandraDB and ScyllaDB

Abstract: Big Data encompasses vast amounts of data, reaching exabytes or even zettabytes. In the current landscape, large databases play a crucial role, particularly in generating substantial data for daily analyses of social media and multimedia content. The enormity of Big Data poses challenges, given its extensive volume of structured, semi-structured, and unstructured data, making traditional database systems and software techniques insufficient. Big Data is frequently defined by its 9 V’s: velocity, variety, volume, veracity, validity, variability, volatility, visualization, and value. This complexity highlights the need for a simple information management strategy that integrates various new data types alongside traditional data. The significance of Big Data databases is emphasized by the daily generation of millions of terabytes of data from sources like social media posts and multimedia. This study aims to evaluate the performance of two Wide-Column Store Big Data database systems, Apache CassandraDB and ScyllaDB, using the CassandraStress Benchmarking Tool. Key metrics such as total operation time, operation rate, partition rate, row rate, and maximum latency will be assessed as the number of records and operations increase. The development of these databases, motivated by diverse industry requirements, emphasizes their adaptation to specific needs. The research methodology outlines the tool used to compare these WideColumn Store databases on various parameters, contributing valuable insights into their performance in real-world scenarios

Preparing for the Worst: Management and Predictive Factors of Open Conversion During Minimally Invasive Renal Tumor Surgery (UroCCR-135 Study)

Background and objectiveData regarding open conversion (OC) during minimally invasive surgery (MIS) for renal tumors are reported from big databases, without precise description of the reason and management of OC. The objective of this study was to describe the rate, reasons, and perioperative outcomes of OC in a cohort of patients who underwent MIS for renal tumor initially. The secondary objective was to find the factors associated with OC. MethodsBetween 2008 and 2022, of the 8566 patients included in the UroCCR project prospective database (NCT03293563), who underwent laparoscopic or robot-assisted minimally invasive partial (MIPN) or radical (MIRN) nephrectomy, 163 experienced OC. Each center was contacted to enlighten the context of OC: “emergency OC” implied an immediate life-threatening situation not reasonably manageable with MIS, otherwise “elective OC”. To evaluate the predictive factors of OC, a 2:1 paired cohort on the UroCCR database was used. Key findings and limitationsThe incidence rate of OC was 1.9% for all cases of MIS, 2.9% for MIRN, and 1.4% for MIPN. OC procedures were mostly elective (82.2%). The main reason for OC was a failure to progress due to anatomical difficulties (42.9%). Five patients (3.1%) died within 90 d after surgery. Increased body mass index (BMI; odds ratio [OR]: 1.05, 95% confidence interval [CI]: 1.01–1.09, p = 0.009) and cT stage (OR: 2.22, 95% CI: 1.24–4.25, p = 0.008) were independent predictive factors of OC. Conclusions and clinical implicationsIn MIS for renal tumors, OC was a rare event (1.9%), caused by various situations, leading to impaired perioperative outcomes. Emergency OC occurred once every 300 procedures. Increased BMI and cT stage were independent predictive factors of OC. Patient summaryThe incidence rate of open conversion (OC) in minimally invasive surgery for renal tumors is low. Only 20% of OC procedures occur in case of emergency, and others are caused by various situations. Increased body mass index and cT stage were independent predictive factors of OC.

Development and Implementation of a Dynamically Updated Big Data Intelligence Platform Using Electronic Medical Records for Secondary Hypertension.

The accurate identification and diagnosis of secondary hypertension is critical,especially while cardiovascular heart disease continues to be the leading cause of death. To develop a big data intelligence platform for secondary hypertension using electronic medical records to contribute to future basic and clinical research. Using hospital data, the platform, named Hypertension DATAbase at Urumchi (UHDATA), included patients diagnosed with hypertension at the People's Hospital of Xinjiang Uygur Autonomous Region since December 2004. The electronic data acquisition system, the database synchronization technology, and data warehouse technology (extract-transform-load, ETL) for the scientific research big data platform were used to synchronize and extract the data from each business system in the hospital. Standard data elements were established for the platform, including demographic and medical information. To facilitate the research, the database was also linked to the sample database system, which includes blood samples, urine specimens, and tissue specimens. From December 17, 2004, to August 31, 2022, a total of 295,297 hypertensive patients were added to the platform, with 53.76% being males, with a mean age of 59 years, and 14% with secondary hypertension. However, 75,802 patients visited the Hypertension Center at our hospital, with 43% (32,595 patients) being successfully diagnosed with secondary hypertension. The database contains 1458 elements, with an average fill rate of 90%. The database can continuously include the data for new hypertensive patients and add new data for existing hypertensive patients, including post-discharge follow-up information, and the database updates every 2 weeks. Presently, some studies that are based on the platform have been published. Using computer information technology, we developed and implemented a big database of dynamically updating electronic medical records for patients with hypertension, which is helpful in promoting future research on secondary hypertension.

Materials genome engineering accelerates the research and development of organic and perovskite photovoltaics

AbstractThe emerging photovoltaic (PV) technologies, such as organic and perovskite PVs, have the characteristics of complex compositions and processing, resulting in a large multidimensional parameter space for the development and optimization of the technologies. Traditional manual methods are time‐consuming and labor‐intensive in screening and optimizing material properties. Materials genome engineering (MGE) advances an innovative approach that combines efficient experimentation, big database and artificial intelligence (AI) algorithms to accelerate materials research and development. High‐throughput (HT) research platforms perform multidimensional experimental tasks rapidly, providing a large amount of reliable and consistent data for the creation of materials databases. Therefore, the development of novel experimental methods combining HT and AI can accelerate materials design and application, which is beneficial for establishing material‐processing‐property relationships and overcoming bottlenecks in the development of emerging PV technologies. This review introduces the key technologies involved in MGE and overviews the accelerating role of MGE in the field of organic and perovskite PVs.

Taylor jellyfish search algorithm enabled feature selection and deep learning for big data classification using mapreduce framework

ABSTRACT Nowadays, the increasing growth of technology contributes various organisations to storing a huge amount of records using web servers in the form of big data. The big database possesses a vast amount of relevant and irrelevant features, and the less significant features are difficult to remove which makes the process time-consuming and more complex. Hence, this paper proposes an optimisation algorithm, named the Taylor Jellyfish Search Optimisation (TaylorJSO) for selecting the relevant features. Here, an optimisation-based deep learning approach called Taylor Jellyfish Search Optimisation-Deep Maxout Network (TaylorJSO-DMN) is introduced to classify big data more effectively by selecting optimal features. In this model, the input database is preprocessed using quantile normalisation and uses the oversampling of imbalanced classes for data augmentation. In the MapReduce framework, the selection of optimal features is performed in the mapper phase, whereas in the reducer phase, the DMN is used for the classification of big data. The DMN is trained using the introduced TaylorJSO algorithm to improve the classification performance of big data classification. Moreover, the TaylorJSO-DMN model attained greater performance with a maximum of 92.5% accuracy, 91.6% sensitivity, and 92.9% specificity than other traditional approaches.

The systemic immune-inflammation index is significantly associated with the severity of silicosis: a 9-year retrospective study in Beijing.

Silicosis shows an increasing trend with the development of new industries. However, the potential biomarkers for predicting the disease severity are lacking. A novel inflammatory marker, the systemic immune-inflammation Index (SII), has not been studied in silicosis. In this retrospective study, we used data from a big database platform of a tertiary general hospital in Beijing, which was established based on the electronic medical records of the hospital. The clinical data of adult patients diagnosed with silicosis at the Department of Occupational Medicine and Toxicology from 2013 to 2022 were collected. The data extracted from the database were in de-identified form. Only patients with a first diagnosis of silicosis and without conditions that might affect the parameters of routine blood tests were included in the analysis. Analyses were performed to assess the relationship between SII and the advanced stage of silicosis. A total of 246 participants were included in the study. Most of the patients were exposed to silica particles during excavation and digging (n = 149, 60.6%). SII level was significantly higher in patients with advanced stages of silicosis. A multivariate logistic regression analysis revealed that a higher SII level was associated with the advanced stage of silicosis [odds ratio (OR) = 1.002; 95% confidence interval (CI): 1.000-1.003, p < 0.001] after adjusting for all covariates. The best cutoff value of SII was 444.1. The results of the subgroup analysis also showed a significant correlation between SII level over 444.1 and the advanced stage of silicosis in groups stratified by gender, history of smoking, and duration of silica exposure. Moreover, our results showed a significant but weak negative correlation between the level of SII and some lung function parameters in silicosis. Higher SII is associated with the advanced stage of silicosis and impaired lung function. More long-term, large-scale studies are needed to confirm these findings.