Extraction Software Research Articles

Research on Fine-Grained Phenotypic Analysis of Temporal Root Systems - Improved YoloV8seg Applied for Fine-Grained Analysis of In Situ Root Temporal Phenotypes.

Root systems are crucial organs for crops to absorb water and nutrients. Conducting phenotypic analysis on roots is of great importance. To date, methods for root system phenotypic analysis have predominantly focused on semantic segmentation, integrating phenotypic extraction software to achieve comprehensive root phenotype analysis. This study demonstrates the feasibility of instance segmentation tasks on in situ root system images. An improved YoloV8n-seg network tailored for detecting elongated roots is proposed, which outperforms the original YoloV8seg in all network performance metrics. Additionally, the post-processing method introduced reduces root identification errors, ensuring a one-to-one correspondence between each root system and its detection box. The experiment yields phenotypic parameters for fine-grained roots, such as fine-grained root length, diameter, and curvature. Compared to traditional parameters like total root length and average root diameter, these detailed phenotypic analyses enable more precise phenotyping and facilitate accurate artificial intervention during crop cultivation.

BUSClean: Open-source software for breast ultrasound image pre-processing and knowledge extraction for medical AI.

Development of artificial intelligence (AI) for medical imaging demands curation and cleaning of large-scale clinical datasets comprising hundreds of thousands of images. Some modalities, such as mammography, contain highly standardized imaging. In contrast, breast ultrasound imaging (BUS) can contain many irregularities not indicated by scan metadata, such as enhanced scan modes, sonographer annotations, or additional views. We present an open-source software solution for automatically processing clinical BUS datasets. The algorithm performs BUS scan filtering (flagging of invalid and non-B-mode scans), cleaning (dual-view scan detection, scan area cropping, and caliper detection), and knowledge extraction (BI-RADS Labeling and Measurement fields) from sonographer annotations. Its modular design enables users to adapt it to new settings. Experiments on an internal testing dataset of 430 clinical BUS images achieve >95% sensitivity and >98% specificity in detecting every type of text annotation, >98% sensitivity and specificity in detecting scans with blood flow highlighting, alternative scan modes, or invalid scans. A case study on a completely external, public dataset of BUS scans found that BUSClean identified text annotations and scans with blood flow highlighting with 88.6% and 90.9% sensitivity and 98.3% and 99.9% specificity, respectively. Adaptation of the lesion caliper detection method to account for a type of caliper specific to the case study demonstrates the intended use of BUSClean in new data distributions and improved performance in lesion caliper detection from 43.3% and 93.3% out-of-the-box to 92.1% and 92.3% sensitivity and specificity, respectively. Source code, example notebooks, and sample data are available at https://github.com/hawaii-ai/bus-cleaning.

Fiducial Reference Measurements for Air Quality Monitoring Using Ground-Based MAX-DOAS Instruments (FRM4DOAS)

The UV–Visible Working Group of the Network for the Detection of Atmospheric Composition Changes (NDACC) focuses on the monitoring of air-quality-related stratospheric and tropospheric trace gases in support of trend analysis, satellite validation and model studies. Tropospheric measurements are based on MAX-DOAS-type instruments that progressively emerged in the years 2010 onward. In the interest of improving the overall consistency of the NDACC MAX-DOAS network and facilitating its further extension to the benefit of satellite validation, the ESA initiated, in late 2016, the FRM4DOAS project, which aimed to set up the first centralised data processing system for MAX-DOAS-type instruments. Developed by a consortium of European scientists with proven expertise in measurements, data extraction algorithms and software design specialities, the system has now reached pre-operational status and has demonstrated its ability to deliver a set of quality-controlled atmospheric composition data products with a latency of one day. The processing system has been designed using a highly modular approach, making it easy to integrate new tools or processing updates. It incorporates advanced algorithms selected by community consensus for the retrieval of total ozone, lower tropospheric and stratospheric NO2 vertical profiles and formaldehyde profiles. The ozone and NO2 products are currently generated from a total of 22 stations and delivered daily to the NDACC rapid delivery (RD) repository, with an additional mirroring to the ESA Validation Data Centre (EVDC). Although it is still operated in a pre-operational/demonstrational mode, FRM4DOAS was already used for several validation and science studies, and it was also deployed in support of field campaigns for the validation of the TROPOMI and GEMS satellite missions. It recently went through a CEOS-FRM self-assessment process aiming at assessing the level of maturity of the service in terms of instrumentation, operations, data sampling, metrology and verification. Based on this evaluation, it falls under class C, which is a good rating but also implies that further improvements are needed to reach full compliance with FRM standards, i.e., class A.

CoFFEE: a codec-based forensic feature extraction and evaluation software for H.264 videos

The forensic analysis of digital videos is becoming increasingly relevant to deal with forensic cases, propaganda, and fake news. The research community has developed numerous forensic tools to address various challenges, such as integrity verification, manipulation detection, and source characterization. Each tool exploits characteristic traces to reconstruct the video life-cycle. Among these traces, a significant source of information is provided by the specific way in which the video has been encoded. While several tools are available to analyze codec-related information for images, a similar approach has been overlooked for videos, since video codecs are extremely complex and involve the analysis of a huge amount of data. In this paper, we present a new tool designed for extracting and parsing a plethora of video compression information from H.264 encoded files, including macroblocks structure, prediction residuals, and motion vectors. We demonstrate how the extracted features can be effectively exploited to address various forensic tasks, such as social network identification, source characterization, and double compression detection. We provide a detailed description of the developed software, which is released free of charge to enable its use by the research community to create new tools for forensic analysis of video files.

Assessment of Age-Related Differences in Lower Leg Muscles Quality Using Radiomic Features of Magnetic Resonance Images.

Sarcopenia, characterised by a decline in muscle mass and strength, affects the health of the elderly, leading to increased falls, hospitalisation, and mortality rates. Muscle quality, reflecting microscopic and macroscopic muscle changes, is a critical determinant of physical function. To utilise radiomic features extracted from magnetic resonance (MR) images to assess age-related changes in muscle quality, a dataset of 24 adults, divided into older (male/female: 6/6, 66-79years) and younger (male/female: 6/6, 21-31years) groups, was used to investigate the radiomics features of the dorsiflexor and plantar flexor muscles of the lower leg that are critical for mobility. MR images were processed using MaZda software for feature extraction. Dimensionality reduction was performed using principal component analysis and recursive feature elimination, followed by classification using machine learning models, such as support vector machine, extreme gradient boosting, and naïve Bayes. A leave-one-out validation test was used to train and test the classifiers, and the area under the receiver operating characteristic curve (AUC) was used to evaluate the classification performance. The analysis revealed that significant differences in radiomic feature distributions were found between age groups, with older adults showing higher complexity and variability in muscle texture. The plantar flexors showed similar or higher AUC than the dorsiflexors in all models. When the dorsiflexor muscles were combined with the plantar flexor muscles, they tended to have a higher AUC than when they were used alone. Radiomic features in lower-leg MR images reflect ageing, especially in the plantar flexor muscles. Radiomic analysis can offer a deeper understanding of age-related muscle quality than traditional muscle mass assessments.

TG Vascutrack: A User-Friendly and Open-Source Software for Automated Extraction of Arterial Diameter and Velocity Profile Data From Vascular Ultrasound Videos.

Existing automated software for vascular ultrasound data extraction lacks free, open-source options suitable for professionals without coding experience. These programs typically include signal-cleaning algorithms, resulting in processed output without access to raw data. To address these needs, we developed TG Vascutrack, an open-source and user-friendly software tailored for non-coder professionals. It features a graphical interface, multiple functionalities, and provides access to raw data. Comparative analysis against validated software and manual extraction revealed minimal biases and standard deviations in diameter and velocity measurements. TG Vascutrack offers a free, promising solution for non-coders needing automated vascular ultrasound data extraction.

Simplified Ftacv Model to Quantify the Electrochemically Active Sitedensity in PGM-Free ORR Catalysts

The development of platinum group metal-free catalysts is considered the most prominent path for reducing thecost of low-temperature fuel cells (LTFC). Despite the great advancement in the field, its further progress iscurrently limited by the ability to understand and mitigate the catalysts’ degradation mechanisms, which up torecent years was limited by the lack of activity descriptors. Recent work showed that this could be solved usingFourier-transformed alternating current voltammetry that enables to deconvolute Faradaic currents arising fromthe redox reaction of the active sites from the capacitive currents, and by that accurately measure the electrochemicallyactive site density of these catalysts in situ fuel cells. However, the analysis of the results can becomplex, requiring simulation software for accurate parameter extraction. Herein, a simplified analysis ofFourier-transformed alternating current voltammetry is presented. This was done by mapping the influence ofvarious variables, such as resistivity, capacitance, and thermodynamic and kinetic dispersion, on the accuracy ofelectrochemically active site density measurements. Under specific, yet realistic, operating conditions, a singleequation with the peak current of the 5th harmonic as the only variable can be used to quantify electrochemicallyactive site density with high accuracy. This approach was demonstrated using molecular catalysts, iron phthalocyanine,as a model molecule, and a commercial, heat-treated catalyst.

A Simple and Systematic Approach to Qualitative Data Extraction From Social Media for Novice Health Care Researchers: Tutorial.

Social media analyses have become increasingly popular among health care researchers. Social media continues to grow its user base and, when analyzed, offers unique insight into health problems. The process of obtaining data for social media analyses varies greatly and involves ethical considerations. Data extraction is often facilitated by software tools, some of which are open source, while others are costly and therefore not accessible to all researchers. The use of software for data extraction is accompanied by additional challenges related to the uniqueness of social media data. Thus, this paper serves as a tutorial for a simple method of extracting social media data that is accessible to novice health care researchers and public health professionals who are interested in pursuing social media research. The discussed methods were used to extract data from Facebook for a study of maternal perspectives on sudden unexpected infant death.

3D surgical planning method for lower jaw osteotomies applied to facial feminization surgery

Our proposed method uses a three-dimensional (3D) measurement approach that focuses mainly on the lower jaw from basal, lateral, and frontal views applied to the volumetric skull model derived from a computed tomography (CT) of the head. Likewise, we discuss the geometrical features and clinical considerations involved in the 3D biomodeling of the surgical osteotomy. The workflow that allowed this virtual planning to be developed was composed of medical imaging processing software, data extraction software from images, and statistical software that allows the creation and generation of curve-fitting (nonlinear regression) graphs from data. Thirty-two (32) anatomical points were positioned, sixteen (16) measurements were taken, and two-dimensional (2D) sketches in three views (frontal, lateral, and inferior) were generated to overlap in a 3D environment, which informed the cutting of the desired bone segments. Implementing a nonlinear regression curve-fitting on the contours of the original jaws allowed optimal planning of the osteotomy. Desired cutting shapes were extrapolated for the front view by third-order equations, while for the side and bottom views, log-normal distribution curves and second-order polynomial curves were used, respectively. The reduction in the mandibular volume was between 6.55 and 10.27 %, with two of the most important measurements related to vertical reduction in the lateral views and the difference to determine gonion reduction.

La felicidad social como fundamento en el diseño de políticas públicas :Una revisión sistemática

This paper examines the literature on social happiness as a basis for designing public policies. The systematic review initiated with a Scopus search using Boolean criteria, covering the period from 2013 to 2023. 49 out of 943 articles were selected through inclusion and exclusion criteria using RAYYAN software for data management and extraction, and the PRISMA model as a guide to ensure the replicability of the research. The descriptive findings reveal that most publications are from Latin America, showing a consistent increase in scientific output over the years, with 2021 being the peak year for related publications. This paper identifies debates on the viability of using happiness as a foundation in public policy design, importance of impact evaluation, and experimental methods. The document categorizes the macroeconomic, institutional and governmental, sociodemographic, psychological, and environmental and sustainability variables that interrelate with happiness, including factors such as income, economic stability, wealth distribution, state intervention, among other relevant factors. The literature reveals gaps, such as the lack of a universally accepted definition of social happiness and a shortage of longitudinal and experimental studies. Recommendations include a multidimensional approach, consideration of cultural and socioeconomic context, use of experimental methodologies, citizen participation, and incorporating happiness into the political agenda. Final reflections underscore transformative potential of integrating subjective well-being into the government agenda and the associated challenges, such as resistance to change and the lack of consensus on how to define and measure social happiness.

Characterizing Canopy Structure Variability in Amazonian Secondary Successions with Full-Waveform Airborne LiDAR

Full-waveform LiDAR (FWF) offers a promising advantage over other technologies to represent the vertical canopy structure of secondary successions in the Amazon region, as the waveform encapsulates the properties of all elements intercepting the emitted beam. In this study, we investigated modifications in the vertical structure of the Amazonian secondary successions across the vegetation gradient from early to advanced stages of vegetation regrowth. The analysis was performed over two distinct climatic regions (Drier and Wetter), designated using the Maximum Cumulative Water Deficit (MCWD). The study area was covered by 309 sample plots distributed along 25 LiDAR transects. The plots were grouped into three successional stages (early—SS1; intermediate—SS2; advanced—SS3). Mature Forest (MF) was used as a reference of comparison. A total of 14 FWF LiDAR metrics from four categories of analysis (Height, Peaks, Understory and Gaussian Decomposition) were extracted using the Waveform LiDAR for Forestry eXtraction (WoLFeX) software (v1.1.1). In addition to examining the variation in these metrics across different successional stages, we calculated their Relative Recovery (RR) with vegetation regrowth, and evaluated their ability to discriminate successional stages using Random Forest (RF). The results showed significant differences in FWF metrics across the successional stages, and within and between sample plots and regions. The Drier region generally exhibited more pronounced differences between successional stages and lower FWF metric values compared to the Wetter region, mainly in the category of height, peaks, and Gaussian decomposition. Furthermore, the Drier region displayed a lower relative recovery of metrics in the early years of succession, compared to the areas of MF, eventually reaching rates akin to those of the Wetter region as succession progressed. Canopy height metrics such as Waveform distance (WD), and Gaussian Decomposition metrics such as Bottom of canopy (BC), Bottom of canopy distance (BCD) and Canopy distance (CD), related to the height of the lower forest stratum, were the most important attributes in discriminating successional stages in both analyzed regions. However, the Drier region exhibited superior discrimination between successional stages, achieving a weighted F1-score of 0.80 compared to 0.73 in the Wetter region. When comparing the metrics from SS in different stages to MF, our findings underscore that secondary forests achieve substantial relative recovery of FWF metrics within the initial 10 years after land abandonment. Regions with potentially slower relative recovery (e.g., Drier regions) may require longer-term planning to ensure success in providing full potential ecosystem services in the Amazon.

Advanced Data Extraction and Analytics Software on AWS: Revolutionizing Insights with Personalized

Abstract: In the current evolving landscape of marketing, a robust analytics platform is required that integrates data from various marketing channels for insightful decision-making. This paper introduces an Integrated Marketing Data Analytics Platform, a comprehensive solution designed to enhance marketing strategies through advanced analytics. The platform amalgamates data from different marketing sources, enabling an integrated view of campaigns, customer behavior, and market trends. Key features include cross-channel analytics, predictive modeling, and real-time reporting. This research aims to empower marketing professionals with a versatile tool that maximizes the impact and outcome of marketing efforts by providing actionable insights and promoting data-driven decision-making in marketing strategy. (Abstract) Keywords: Marketing analytics, Integrated marketing data , Advanced analytics Cross-ch

Rains, Tragedies and Media Coverage: Analysis of Floods In Rio Grande do Sul

Objective: The objective of this study is to present how the tragedy in Rio Grande do Sul is handled by the media, in order to observe whether it is being related to climate change. Theoretical Framework: Environmental disasters have been intensifying with the great advances of human actions, which cause climate change and journalistic structures play a crucial role in shaping the public perception of natural disasters, framing them in the context of climate change. Method: From the use of text mining techniques in reports published in Brazilian news portals about the floods in Rio Grande do Sul, a corpus of 100 texts from Google News was selected, being processed through the Voyant Tools software for extraction and analysis of textual data. Results and Discussion: Media coverage of the tragedy in Rio Grande do Sul highlighted the social consequences and the impact on urban centers. The news simplifies the understanding of the causes of the tragedy, focusing mainly on heavy rains and neglecting factors such as global warming and El Niño. In addition, socioeconomic influences that amplify flood risks, such as poor infrastructure and housing conditions, were neglected in the discussions. The media also emphasized the importance of understanding the interaction between climate and geography to formulate strategies that minimize the effects of flooding and promote the sustainable development of the affected regions. Research Implications: By revealing how climate change narratives are being integrated into coverage of natural disasters, this study helps to understand the role of the media in educating the public and influencing public perception of the need for climate mitigation and adaptation policies. Originality/Value: This study examines how the Brazilian media approaches flooding in Rio Grande do Sul, highlighting the social consequences and the importance of understanding the interaction between climate and geography. Using text mining techniques to analyze a corpus of news stories, the research offers a succinct and objective analysis of media coverage, contributing to an understanding of communication about extreme weather events.

MEDICAL DATA EXTRACTION

The proposed system is a user-friendly and efficient Data Extraction software for Medical Documents and Images, integrating advanced technologies such as PHP, Android, database, ML-Kit, retrofit, recycler views. The application facilitates efficient user registration, user login, insurance claims, and management of claimed users. The system aims to enhance the accessibility and accuracy of medical data extraction, thereby optimizing insurance procedures for agents, Also This paper outlines the technical architecture, functionalities, and implementation details of the application, highlighting its potential to revolutionize insurance operations within the healthcare industry. Keywords: - Machine Learning Kit(ML-Kit)

Estrogen restores disordered lipid metabolism in visceral fat of prediabetic mice.

Visceral obesity is increasingly prevalent among adolescents and young adults and is commonly recognized as a risk factor for type 2 diabetes. Estrogen [17β-estradiol (E2)] is known to offer protection against obesity via diverse me-chanisms, while its specific effects on visceral adipose tissue (VAT) remain to be fully elucidated. To investigate the impact of E2 on the gene expression profile within VAT of a mouse model of prediabetes. Metabolic parameters were collected, encompassing body weight, weights of visceral and subcutaneous adipose tissues (VAT and SAT), random blood glucose levels, glucose tolerance, insulin tolerance, and overall body composition. The gene expression profiles of VAT were quantified utilizing the Whole Mouse Genome Oligo Microarray and subsequently analyzed through Agilent Feature Extraction software. Functional and pathway analyses were conducted employing Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analyses, respectively. Feeding a high-fat diet (HFD) moderately increased the weights of both VAT and SAT, but this increase was mitigated by the protective effect of endogenous E2. Conversely, ovariectomy (OVX) led to a significant increase in VAT weight and the VAT/SAT weight ratio, and this increase was also reversed with E2 treatment. Notably, OVX diminished the expression of genes involved in lipid metabolism compared to HFD feeding alone, signaling a widespread reduction in lipid metabolic activity, which was completely counteracted by E2 administration. This study provides a comprehensive insight into E2's local and direct protective effects against visceral adiposity in VAT at the gene level. In conclusion, the present study demonstrated that the HFD-induced over-nutritional challenge disrupted the gene expression profile of visceral fat, leading to a universally decreased lipid metabolic status in E2 deficient mice. E2 treatment effectively reversed this condition, shedding light on the mechanistic role and therapeutic potential of E2 in combating visceral obesity.

Phylogenetic Analysis and Mutation of Sars-Cov-2 in Bats in Karst Malang City, Indonesia

Introduction: A group of people in China were hospitalized with an initial diagnosis of pneumonia of unknown cause. The patients were linked to a wholesale wet seafood and animal market in Wuhan, Hubei Province, China. The disease has spread to other provinces in China, Thailand, Japan, and South Korea in less than a month. SARS-CoV-2 was found to originate from bats. Therefore, this research aims to analyze SARS-CoV-2 mutation in bats in Malang Karst, Indonesia. Methods: Other bat body parts used as research samples include the brain, liver, kidneys, intestines, pancreas, fetus, blood, lungs, and ectoparasites. The samples were taken separately and placed in a container containing 10% PBF. For further analysis, we used RNA Extraction, Real-Time PCR, Sequencing, and CoV Gisaid mutation analysis software to analyze the sequencing data. Then, EMBL software will be used to analyze the phylogenetically. Results and Discussion: There was 1 sample that showed a positive result for Covid-19, namely the intestine of the Cynoptera brachyotis species. There were differences between SARS-CoV-2 in bats in Malang Karst in Indonesia compared to SARS-CoV from 2000 to 2019. The spike protein's receptor binding domain (RBD) is the most variable part of the coronavirus genome. Conclusion: From the research results, one positive sample was obtained using Real-Time PCR, and based on mutation analysis, mutations were found in SARS-CoV-2 against the SARS-CoV virus from 2000-2019. Further research is needed, especially regarding SARS-CoV-2 as a vaccine.

MRI-based Texture Analysis in Differentiation of Benign and Malignant Vertebral Compression Fractures.

The diagnosis and characterization of vertebral compression fractures are very important for clinical management. In this evaluation, which is usually performed with diagnostic (conventional) imaging, the findings are not always typical or diagnostic. Therefore, it is important to have new information to support imaging findings. Texture analysis is a method that can evaluate information contained in diagnostic images and is not visually noticeable. This study aimed to evaluate the magnetic resonance images of cases diagnosed with vertebral compression fractures by the texture analysis method, compare them with histopathological data, and investigate the effectiveness of this method in the differentiation of benign and malignant vertebral compression fractures. Fifty-five patients with a total of 56 vertebral compression fractures were included in the study. Magnetic resonance images were examined and segmented using Local Image Feature Extraction (LIFEx) software, which is an open-source program for texture analysis. The results were compared with the histopathological diagnosis. The application of the Decision Tree algorithm to the dataset yielded impressively accurate predictions (≈95% in accuracy, precision, and recall). Interpreting tissue analysis parameters together with conventional magnetic resonance imaging findings can improve the abilities of radiologists, lead to accurate diagnoses, and prevent unnecessary invasive procedures. Further prospective trials in larger populations are needed to verify the role and performance of texture analysis in patients with vertebral compression fractures.

Comparison of ureteral stone measurements for predicting the efficacy of a single session of extracorporeal shockwave lithotripsy: one-, two-, and three-dimensional computed tomography measurements.

The objective of this study was to compare the value of one-, two- and three-dimensional computed tomography (CT) measurements for predicting the efficacy of a single session of extracorporeal shock wave lithotripsy (ESWL) in patients with a single ureteral stone. A total of 165 patients were included based on the inclusion and exclusion criteria. Different models were constructed using a combination of patients' clinical data and measurements obtained by manual sketching and automated extraction software. Multivariate logistic regression was used to develop the models. Receiver operating characteristic curves were used to assess the performance of the models. There was good interobserver agreement for all measurements in different dimensions (P < 0.001). We also found that hydronephrosis, the largest diameter, the largest area, volume, and mean CT value were significantly greater in the failure group than in the success group (P < 0.01). Furthermore, all sizes and CT measurement values were found to be independent predictors for predicting efficacy after one session of ESWL (P < 0.05). In addition, the multivariate logistic analysis showed that the area under the curve (AUC) for two-dimensional and three-dimensional measurements was superior to that of one-dimensional measurement (P < 0.01). However, when size alone was included as a measurable predictor, there was no significant difference in the AUC among the one-, two-, and three-dimensional measurements (P > 0.05). In summary, after adjusting for clinical data, two- and three-dimensional measurements combining ureteral stone size and CT values were found to be the best predictors of ESWL efficacy, and software-based three-dimensional measurements should be considered to avoid interobserver variability in clinical practice.

Study on Predicting Clinical Stage of Patients with Bronchial Asthma Based on CT Radiomics.

To explore the value of a new model based on CT radiomics in predicting the staging of patients with bronchial asthma (BA). Patients with BA from 2018 to 2021 were retrospectively analyzed and underwent plain chest CT before treatment. According to the guidelines for the prevention and treatment of BA (2016 edition), they were divided into two groups: acute attack and non-acute attack. The images were processed as follows: using Lung Kit software for image standardization and segmentation, using AK software for image feature extraction, and using R language for data analysis and model construction (training set: test set = 7: 3). The efficacy and clinical effects of the constructed model were evaluated with ROC curve, sensitivity, specificity, calibration curve and decision curve. A total of 112 patients with BA were enrolled, including 80 patients with acute attack (range: 2-86 years old, mean: 53.89±17.306 years old, males of 33) and 32 patients with non-acute attack (range: 4-79 years old, mean: 57.38±19.223 years old, males of 18). A total of 10 imaging features are finally retained and used to construct model using multi-factor logical regression method. In the training group, the AUC, sensitivity and specificity of the model was 0.881 (95% CI:0.808-0.955), 0.804 and 0.818, separately; while in the test group, it was 0.792 (95% CI:0.608-0.976), 0.792 and 0.80, respectively. The model constructed based on radiomics has a good effect on predicting the staging of patients with BA, which provides a new method for clinical diagnosis of staging in BA patients.

Clinical application of machine learning-based pathomics signature of gastric atrophy.

The diagnosis of gastric atrophy is highly subjective, and we aimed to establish a model of gastric atrophy based on pathological features to improve diagnostic consistency. We retrospectively collected the HE-stained pathological slides of gastric biopsies and used CellProfiler software for image segmentation and feature extraction of ten representative images for each sample. Subsequently, we employed the Least absolute shrinkage and selection operator (LASSO) to select features and different machine learning (ML) algorithms to construct the diagnostic models for gastric atrophy. We selected 289 gastric biopsy specimens for training, testing, and external validation. We extracted 464 pathological features and screened ten features by LASSO to establish the diagnostic model for moderate-to-severe atrophy. The range of area under the curve (AUC) for various machine learning algorithms was 0.835-1.000 in the training set, 0.786-0.949 in the testing set, and 0.689-0.818 in the external validation set. LR model had the highest AUC value, with 0.900 (95% CI: 0.852-0.947) in the training set, 0.901 (95% CI: 0.807-0.996) in the testing set, and 0.818 (95% CI: 0.714-0.923) in the external validation set. The atrophy pathological score based on the LR model was associated with endoscopic atrophy grading (Z=-2.478, P=0.013) and gastric cancer (GC) (OR=5.70, 95% CI: 2.63-12.33, P<0.001). The ML model based on pathological features could improve the diagnostic consistency of gastric atrophy, which is also associated with endoscopic atrophy grading and GC.