Classification Rate Research Articles

Methods and Strategies of Microsurgery Combined with Ilizarov Technique in the Treatment of Amputation of Limbs in Renji Hospital: A Report of 51 Cases.

Severe limb amputation trauma often results in bone and soft tissue defects after debridement. Traditional replantation aims to save the limb by shortening the ischemic period and using autologous transplantation for repair, but it can lead to surgical trauma, donor site damage, and prolonged operation time. Due to contusion, pollution, and complex injury, there is no unified standard for replantation and fixation. Improper operation can easily lead to complications such as bone infection, nonunion, bone defect, and joint stiffness. This study introduces the Ilizarov technique into microsurgery to improve limb lengthening after reconstruction and standardizes the steps of replantation fixation for complex limb avulsion injuries, with a focus on clinical efficacy. A retrospective analysis was performed on 51 patients with complex limb amputation who were treated in Zhengzhou Renji Hospital from June 2009 to March 2021. On the basis of microsurgical limb replantation, Ilizarov technology was introduced to innovate the internal and external combined stepwise fixation method for replantation. Patients' gender, age, height, weight, BMI, and other general information were collected. X-ray films were reviewed regularly to observe the surgical healing of fracture, that is, the degree of limb shortening. The lengthening time, carrying time after lengthening, follow-up time, Dahl classification, Paley fracture healing classification, and Chen Zhongwei's replantation function score were used to evaluate the recovery of the affected limb. A total of 51 patients were included in this group, including 36 male patients and 15 female patients. All the amputated wounds were single limb amputation. In this group of patients, the hind limbs were shortened by 2-12.5 cm (5.32 ± 2.24) after replantation. A total of 44 patients whose hind limbs were shortened by more than 2.5 cm were treated with two-stage Ilizarov lengthening for 1.5-5.5 months (3.19 ± 1.03). The carrying time was 3-7.5 months (4.25 ± 0.94), and the follow-up time was 1-7.8 years (3.76 ± 1.69). Among the 49 survived patients, the Dahl grade of external fixation was less than Grade 2 in 89.8%. The excellent and good rate of Paley fracture healing classification was 89.8%. The excellent and good rate of Chen Zhongwei's limb replantation function classification was 79.6%. Microsurgery combined with Ilizarov technique in the treatment of limb amputation injury, limb salvage reconstruction with internal and external combined step-by-step combined fixation, and one-stage shortening and two-stage limb lengthening can reduce the occurrence of osteomyelitis, bone defect and nonunion, expand the indications of limb replantation, improve the success rate of limb replantation, with satisfactory results, and facilitate the promotion of clinical techniques.

Method for enhanced gesture recognition under low light conditions based on wearable mechanoluminescence sensors

Gestures, as one of the key modes of human–computer interaction, are currently identified and processed mainly through two methods: machine vision imaging and wearable stress/strain sensing.However, the former has high requirements for lighting and background.This paper innovatively proposes a feature fusion algorithm that combines mechanoluminescent sensors for gesture recognition under low-light conditions. Unlike existing methods, this algorithm creates a system that integrates mechanoluminescent sensors for visual gesture recognition. By triggering the sensor’s luminescence through mechanical stress, it provides additional light source information to enhance image quality in low-light environments.For this purpose, this study employed the fluorescent emission properties of ZnS:Cu to fabricate a mechanoluminescence sensor. Its mechanoluminescence response characteristics were measured. Based on these findings, an algorithm for the fusion of gesture information features was developed for low illumination backgrounds. This algorithm decomposes the gesture visual images through YCbCr, extracting the HOG features of the Cr channel (Cr-HOG). Simultaneously, feature fusion is performed using the pooling layer (VGG16-Conv5-pool) following the convolutional layer (Conv5-3) in the VGG16 model. A one-to-one composite SVM classifier is constructed to complete the training and testing of the gesture recognition model. Ultimately, the feasibility of this enhanced gesture recognition method was validated through unmanned vehicle control experiment. Experimental results demonstrated that in environments with an illuminance level below 10 LUX measured by a light meter, the average recognition rate of the wearable mechanoluminescence sensor feature fusion algorithm reached 95.82 %. This is 32.24 % higher than the recognition rate of the single Cr-HOG feature classification and 28.27 % higher than the single VGG16-Conv5-pool feature classification. Compared with feature extraction using ResNet-50 and DenseNet-121 networks, the recognition rate for classifying gestures under low-light conditions is approximately 30 % higher.Compared with other image enhancement recognition methods, the recognition can be increased by up to 30.88 %.

Development Cyber Risk Assessment for Intrusion Detection Using Enhanced Random Forest

In cybersecurity, the lack of statistical data on cyber-attacks presents a significant challenge from an insurance perspective, hindering the accurate calculation of insurance premiums, furthermore assessing cybersecurity risk exposure and identifying high-risk threat categories. Effective intrusion detection systems (IDS) are paramount in addressing these issues. This research introduces a sophisticated cyber risk assessment model utilizing the Random Forest classification algorithm, tailored explicitly for IDS, and leverages the comprehensive CIC-IDS 2017 dataset. The central objective was to engineer robust models capable of classifying a broad array of cyber threats, focusing on classification accuracy. The model achieved an accurate average classification rate of 96.94% through systematic experimentation and hyperparameter tuning.This study found that 'n_estimators' values of 10 to 300 did not affect cyberattack performance. It was also shown that Bagging and bootstrapping improve model stability by mitigating variance and improving accuracy without many trees. Model performance was high, with an average F1-Score of 97.86%. Cyber-attack statistics are scarce, and from an insurance perspective, the lack of statistical data on cyber-attacks hinders the calculation of insurance premiums. Risk assessment allows for informed self-insurance or risk transfer processes ensuring that policies align with risk management strategies and premium calculations.

Enhancing Oral Squamous Cell Carcinoma Detection Using Histopathological Images: A Deep Feature Fusion and Improved Haris Hawks Optimization-Based Framework.

Oral cancer, also known as oral squamous cell carcinoma (OSCC), is one of the most prevalent types of cancer and caused 177,757 deaths worldwide in 2020, as reported by the World Health Organization. Early detection and identification of OSCC are highly correlated with survival rates. Therefore, this study presents an automatic image-processing-based machine learning approach for OSCC detection. Histopathological images were used to compute deep features using various pretrained models. Based on the classification performance, the best features (ResNet-101 and EfficientNet-b0) were merged using the canonical correlation feature fusion approach, resulting in an enhanced classification performance. Additionally, the binary-improved Haris Hawks optimization (b-IHHO) algorithm was used to eliminate redundant features and further enhance the classification performance, leading to a high classification rate of 97.78% for OSCC. The b-IHHO trained the k-nearest neighbors model with an average feature vector size of only 899. A comparison with other wrapper-based feature selection approaches showed that the b-IHHO results were statistically more stable, reliable, and significant (p < 0.01). Moreover, comparisons with those other state-of-the-art (SOTA) approaches indicated that the b-IHHO model offered better results, suggesting that the proposed framework may be applicable in clinical settings to aid doctors in OSCC detection.

Automatic Paddy Planthopper Detection and Counting Using Faster R-CNN

Counting planthoppers manually is laborious and yields inconsistent results, particularly when dealing with species with similar features, such as the brown planthopper (Nilaparvata lugens; BPH), whitebacked planthopper (Sogatella furcifera; WBPH), zigzag leafhopper (Maiestas dorsalis; ZIGZAG), and green leafhopper (Nephotettix malayanus and Nephotettix virescens; GLH). Most of the available automated counting methods are limited to populations of a small density and often do not consider those with a high density, which require more complex solutions due to overlapping objects. Therefore, this research presents a comprehensive assessment of an object detection algorithm specifically developed to precisely detect and quantify planthoppers. It utilises annotated datasets obtained from sticky light traps, comprising 1654 images across four distinct classes of planthoppers and one class of benign insects. The datasets were subjected to data augmentation and utilised to train four convolutional object detection models based on transfer learning. The results indicated that Faster R-CNN VGG 16 outperformed other models, achieving a mean average precision (mAP) score of 97.69% and exhibiting exceptional accuracy in classifying all planthopper categories. The correctness of the model was verified by entomologists, who confirmed a classification and counting accuracy rate of 98.84%. Nevertheless, the model fails to recognise certain samples because of the high density of the population and the significant overlap among them. This research effectively resolved the issue of low- to medium-density samples by achieving very precise and rapid detection and counting.

Meeting the challenge of varietal and geographical authentication of hazelnuts through lipid metabolite fingerprinting

Hazelnuts are high-quality products with significant economic importance in many European countries. Their market price depends on their qualitative characteristics, which are driven by cultivar and geographical origin, making hazelnuts susceptible to fraud. This study systematically compared two lipidomic fingerprinting strategies for the simultaneous authentication of hazelnut cultivar and provenance, based on the analysis of the unsaponifiable fraction (UF) and triacylglycerol (TAG) profiles by gas chromatography–mass spectrometry coupled with chemometrics. PLS-DA classification models were developed using a large sample set with high natural variability (n = 309) to discriminate hazelnuts by cultivar and origin. External validation results demonstrated the suitability of the UF fingerprint as a hazelnut authentication tool, both tested models showing a high efficiency (>94 %). The correct classification rate of the TAG fingerprinting method was lower (>80 %), but due to its faster analysis time, it is recommended as a complementary screening tool to UF fingerprinting.

Using autonomous recording units to identify and monitor western yellow‐billed cuckoo habitat

AbstractAutonomous recording units (ARUs) paired with signal classification software can be used to detect species‐specific calls, making them useful for evaluating patterns in avian occurrence and activity. However, classification of target signals is not always reliable and may be especially challenging for species that vocalize infrequently. We assessed the use of ARUs to identify and monitor habitat for a cryptic and federally threatened distinct population segment, the western yellow‐billed cuckoo (Coccyzus americanus) in mountainous xeroriparian drainages. Using Kaleidoscope Pro, we developed a call‐classifier and processed acoustic data collected in sites also surveyed using traditional human‐observer methods, applying the same spatial and temporal detection criteria to estimate breeding territories for each method. The classifier detected a total of 4,061 true positive calls at 4 sites, had an overall precision score of 0.07, recall score of 0.09, and F‐score (beta = 1) of 0.08, indicating high false positive and false negative classification rates. Our results were, however, consistent with other ARU studies of rare and cryptic species and ARUs estimated occupancy as effectively as human surveys with as little as 2 hours of daily recording. Total detections varied among sites, likely due to differences in cuckoo population densities and the interaction between topography and ARU detection space. Our results suggest that despite performance shortcomings of call‐classifiers, ARUs can be effective for monitoring cuckoos, with potential for providing higher resolution temporal and spatial information on activity and habitat use and may be particularly effective in remote locations where cuckoos often occur.

Gas Sensing with Nanoporous In2O3 under Cyclic Optical Activation: Machine Learning-Aided Classification of H2 and H2O

Clean hydrogen is a key aspect of carbon neutrality, necessitating robust methods for monitoring hydrogen concentration in critical infrastructures like pipelines or power plants. While semiconducting metal oxides such as In2O3 can monitor gas concentrations down to the ppm range, they often exhibit cross-sensitivity to other gases like H2O. In this study, we investigated whether cyclic optical illumination of a gas-sensitive In2O3 layer creates identifiable changes in a gas sensor’s electronic resistance that can be linked to H2 and H2O concentrations via machine learning. We exposed nanostructured In2O3 with a large surface area of 95 m2 g−1 to H2 concentrations (0–800 ppm) and relative humidity (0–70%) under cyclic activation utilizing blue light. The sensors were tested for 20 classes of gas combinations. A support vector machine achieved classification rates up to 92.0%, with reliable reproducibility (88.2 ± 2.7%) across five individual sensors using 10-fold cross-validation. Our findings suggest that cyclic optical activation can be used as a tool to classify H2 and H2O concentrations.

Specifications of the ACMG/AMP variant curation guidelines for the analysis of germline ATM sequence variants.

The ClinGen Hereditary Breast, Ovarian, and Pancreatic Cancer (HBOP) Variant Curation Expert Panel (VCEP) is composed of internationally recognized experts in clinical genetics, molecular biology, and variant interpretation. This VCEP made specifications for the American College of Medical Genetics and Association for Molecular Pathology (ACMG/AMP) guidelines for the ataxia telangiectasia mutated (ATM) gene according to the ClinGen protocol. These gene-specific rules for ATM were modified from the ACMG/AMP guidelines and were tested against 33 ATM variants of various types and classifications in a pilot curation phase. The pilot revealed a majority agreement between the HBOP VCEP classifications and the ClinVar-deposited classifications. Six pilot variants had conflicting interpretations in ClinVar, and re-evaluation with the VCEP's ATM-specific rules resulted in four that were classified as benign, one as likely pathogenic, and one as a variant of uncertain significance (VUS) by the VCEP, improving the certainty of interpretations in the public domain. Overall, 28 of the 33 pilot variants were not VUS, leading to an 85% classification rate. The ClinGen-approved, modified rules demonstrated value for improved interpretation of variants in ATM.

An unsupervised learning model based on CT radiomics features accurately predicts axillary lymph node metastasis in breast cancer patients: diagnostic study.

The accuracy of traditional clinical methods for assessing the metastatic status of axillary lymph nodes (ALNs) is unsatisfactory. In this study, the authors propose the use of radiomic technology and three-dimensional (3D) visualization technology to develop an unsupervised learning model for predicting axillary lymph node metastasis in patients with breast cancer (BC), aiming to provide a new method for clinical axillary lymph node assessment in patients with this disease. In this study, we retrospectively analyzed the data of 350 patients with invasive BC who underwent lung-enhanced computed tomography (CT) and axillary lymph node dissection surgery at the Department of Breast Surgery of the Second Xiangya Hospital of Central South University. The authors used 3D visualization technology to create a 3D atlas of ALNs and identified the region of interest for the lymph nodes. Radiomic features were subsequently extracted and selected, and a prediction model for ALNs was constructed using the K-means unsupervised algorithm. To validate the model, the authors prospectively collected data from 128 BC patients who were clinically evaluated as negative at our center. Using 3D visualization technology, we extracted and selected a total of 36 CT radiomics features. The unsupervised learning model categorized 1737 unlabeled lymph nodes into two groups, and the analysis of the radiomic features between these groups indicated potential differences in lymph node status. Further validation with 1397 labeled lymph nodes demonstrated that the model had good predictive ability for axillary lymph node status, with an area under the curve of 0.847 (0.825-0.869). Additionally, the model's excellent predictive performance was confirmed in the 128 axillary clinical assessment negative cohort (cN0) and the 350 clinical assessment positive (cN+) cohort, for which the correct classification rates were 86.72 and 87.43%, respectively, which were significantly greater than those of clinical assessment methods. The authors created an unsupervised learning model that accurately predicts the status of ALNs. This approach offers a novel solution for the precise assessment of ALNs in patients with BC.

Sodium Bicarbonate Tolerance During Seedling Stages of Maize (Zea mays L.) Lines

ABSTRACT(1) Soil alkalinization and salinization represent a growing global challenge. Maize (Zea mays L.), with its relatively low tolerance to salt and alkali, is increasingly vulnerable to saline‐alkali stress. Identifying maize genotypes that can withstand salinity and alkalinity is crucial to broaden the base of salt‐alkali‐tolerant maize germplasm. (2) In this study, we screened 65 maize germplasm resources for alkali stress using a 60 mM NaHCO3 solution. We measured fifteen morphological and physiological indices, including seedling height, stem thickness, and leaf area. Various analytical methods—correlation analysis, principal component analysis, subordinate function analysis, cluster analysis, stepwise discriminant analysis, and ridge regression analysis—were used to assess the seedling alkali tolerance of these maize germplasm resources. The physiological indices of six tested maize varieties were analyzed in greater detail. (3) The findings revealed complex correlations among traits, particularly strong negative associations between conductivity and root traits such as length, volume, surface area, diameter, and number of branches. The 15 evaluation indices were reduced to 7 principal components, explaining 77.89% of the variance. By applying affiliation functions and weights, we derived a comprehensive evaluation of maize seedling alkali tolerance. Notably, three germplasms—Liang Yu 99, Bi Xiang 638, and Gan Xin 2818—demonstrated significant comprehensive seedling alkali tolerance. Cluster analysis grouped the 65 maize germplasm resources into four distinct categories (I, II, III, and IV). The results of the cluster analysis were confirmed by multiclass stepwise discriminant analysis, which achieved a correct classification rate of 92.3% for 60 maize genotypes regarding alkalinity tolerance. Using principal component and ridge regression analyses, we formulated a regression equation for alkali tolerance: D‐value = −1.369 + 0.002 * relative root volume + 0.003 * relative number of root forks + 0.006 * relative chlorophyll SPAD + 0.005 * relative stem thickness + 0.005 * relative plant height + 0.001 * relative conductivity + 0.002 * relative dry weight of underground parts. Under sodium bicarbonate stress, morphological indices and germination rates were significantly reduced, germination was inhibited, photosynthetic pigment levels in maize leaves decreased to varying degrees, and the activities of peroxidase (POD), superoxide dismutase (SOD), and catalase (CAT) significantly increased. Alkali stress markedly enhanced the antioxidant enzyme activities in maize varieties, with alkali‐resistant varieties exhibiting a greater increase in antioxidant enzyme activities than alkali‐sensitive varieties under such stress. (4) By screening for alkali tolerance in maize seedlings, the identified alkali‐tolerant genotypes can be further utilized as suitable donor parents, thereby enhancing the use of alkali‐tolerant germplasm resources and providing theoretical guidance for maize cultivation in saline‐alkaline environments.

Monitor and classify dough based on color image with deep learning

To address subjective, time-consuming, and labor-intensive manual methods for monitoring and classifying fermenting dough, an automated and non-destructive method is proposed. Deep learning model YOLOv8s and extracted features including dough surface area, contrast, and homogeneity from RGB color images were employed to monitor fermenting dough. The features were input to a stacked ensemble model (SEM) with base models SVM, AdaBoost, KNN, and RF, with AdaBoost as meta-learner to classify fermenting dough into under-fermented, fermented, and over-fermented. SEM demonstrated a high dough classification rate of 83%, with specific rates of 75% for under-fermented, 71% for fermented, and 90% for over-fermented dough. Results reviewed that combining dough surface area and texture features is effective for monitoring dough, and can be used in adjusting chamber conditions. Furthermore, SEM showed great ability in classifying fermenting dough. The proposed method offers a promising solution for improved bread quality and consistency in bread-making.

Optimal Pathological Prediction of Vocal Fold Leukoplakia Based on IPCLs and Morphology.

To establish a combined classification based on intrapapillary capillary loops (IPCLs) and morphological characteristics to improve the accuracy of pathological prediction of vocal fold leukoplakia (VFL). A prospective research was conducted of VFL patients diagnosed by IPCLs and morphology. The VFL cases were classified as Type I-III based on IPCLs and morphological characteristics. Type I referred to VFL with dendritic vessels but not IPCLs. Type II defined VFL without any IPCLs or vessels and classified by morphology into two subtypes as non-rough Type IIa and rough Type IIb. Type III referred to VFL with IPCLs and classified into two subtypes as Type IIIa with small IPCLs and Type IIIb with large IPCLs or vascular distortion in or around lesions. Predicting pathology accuracy was analyzed. 182 eligible patients were recruited. The prediction accuracy rates of VFL pathology were 81.5% according to the 2019 Ni classification. The combined classification includes 4 cases of Type I, 28 Type IIa, 35 Type IIb, 56 Type IIIa, and 59 Type IIIb VFLs. The prediction accuracy rate of combined classification ranged from 95.1% to 97.3% in three observers. The average sensitivity, specificity, positive predictive value, negative predictive value was 97.8%, 86.2%, 97.4%, 88.2%, respectively. The inter-observer agreement varied from 84.1% to 94.0%, and mean area under curve of receiver-operating curve analysis was 0.954. The new combined classification based on IPCLs and morphological characteristics could predict pathology of VFL accurately. 4 Laryngoscope, 2024.

PTBP1 is a potential indicator of disease progression in acute myeloid leukemia.

This study aimed to verify a novel potential indicator of disease progression in acute myeloid leukemia (AML) patients. Bone marrow samples were collected from 27 AML patients and 27 controls without hematological malignancies. Polypyrimidine tract-binding protein 1 (PTBP1) expression in bone marrow samples was measured, and the association of PTBP1 with the French-American-British (FAB) classification, cytogenetics, risk stratification, and complete remission (CR) rate was analyzed. The correlation between PTBP1 and Ki-67/p53 expression in AML patients was ultimately evaluated. The results showed that PTBP1 mRNA and protein levels were greater in AML patients than in controls. PTBP1 expression was able to distinguish between AML patients and controls (area under the curve, 0.8601; 95% confidence interval, 0.7632-0.9570). Furthermore, PTBP1 expression was associated with an increased frequency of internal tandem duplication mutations within FMS-like tyrosine kinase-3 (FLT3) and a complex karyotype, while PTBP1 expression was not correlated with FAB classification, monosomal karyotype, isolated biallelic CCAAT/enhancer-binding protein α (CEBPA) mutation, or nucleophosmin 1 (NPM1) mutation in patients with AML. Moreover, PTBP1 expression was associated with a poorer prognosis according to risk stratification and a lower CR rate in AML patients. In addition, PTBP1 expression was positively correlated with the expression of the proliferation marker Ki-67 and negatively correlated with the expression of the apoptosis marker p53 in AML patients. Overall, PTBP1 is a viable biomarker that contributes to the risk prediction and the determination of potential drug targets for AML.

A stacking ensemble machine learning based approach for classification of plant diseases through leaf images

Diseases and pests in plants/crops are major causes of significant agricultural losses with economic, social and ecological impacts. Therefore, there is a need for early identification of plant diseases and pests through automated systems. Recently, machine learning-based methods have become popular in solving agricultural problems such as plant diseases faced by technically-noob farmers. This work proposes a novel method based on stacking ensemble machine learning to detect plant diseases in Uradbean precisely. Two classifiers: support vector machine (SVM), random forest (RF) are trained on a dataset consists of Uradbean infected and healthy leaf images. These classifiers are stacked with logistic regression (LR) classifier. In the diverse ensemble, LR classifier is used as a meta-learner which enhanced the precision of the disease classification. The fuzzy C-Means clustering with particle swarm optimization is used for image segmentation. Haralick, Hu Moments and color histogram methods are used in feature extraction. During the tests, the proposed model is also compared with pre-trained networks: DenseNet-201, ResNet-50, and VGG19. It achieved an impressive classification accuracy of 96.82 % which is higher than the individual classifiers and pre-trained networks. To validate model performance, it is evaluated on a benchmark public dataset consists of Apple leaf images and achieved 98.30% accuracy. It is observed that ensemble method reflects an advantage over individual models in increasing the classification rates and reducing the computational overhead in comparison to pre-trained networks which struggle due to the issues such as irrelevant features, generation of pertinent characteristics, and noise

Drivers of snag fall rates in Fennoscandian boreal forests

Abstract Persistence of standing dead trees (snags) is an important determinant for their role for biodiversity and dead wood associated carbon fluxes. How fast snags fall varies widely among species and regions and is further influenced by a variety of stand‐ and tree‐level factors. However, our understanding of this variation is fragmentary at best, partly due to lack of empirical data. Here, we took advantage of the accruing time series of snag observations in the Finnish, Norwegian and Swedish National Forest Inventories that have been followed in these programs since the mid‐1990s. We first harmonized observations from slightly different inventory protocols and then, using this harmonized dataset of ca. 43,000 observations that had a consistent 5‐year census interval, we modelled the probability of snags of the main boreal tree species Pinus sylvestris, Picea abies and Betula spp. falling, as a function of tree‐ and stand‐level variables, using Bayesian logistic regression modelling. The models were moderately good at predicting snags remaining standing or falling, with a correct classification rate ranging from 68% to 75% among species. In general, snag persistence increased with tree size and climatic wetness, and decreased with temperature sum, advancing stage of decay, site productivity and disturbance intensity (mainly harvesting). Synthesis and applications: The effect of harvesting demonstrates that an efficient avenue to increase the amount of snags in managed forests is protecting them during silvicultural operations. In the warmer future, negative relationship between snag persistence and temperature suggests decreasing the time snags remain standing and hence decreasing habitat availability for associated species. As decomposition rates generally increase after fall, decreasing snag persistence also implies substantially faster release of carbon from dead wood.

Predicting Arsenic Contamination in Groundwater: A Comparative Analysis of Machine Learning Models in Coastal Floodplains and Inland Basins

Arsenic (As) contamination in groundwater represents a major global health threat, potentially impacting billions of individuals. Elevated As concentrations are found in river floodplains across south and southeast Asia, as well as in the inland basins of China, despite varying sedimentological and hydrogeochemical conditions. The specific mechanisms responsible for these high As levels remain poorly understood, complicating efforts to predict and manage the contamination. Applying hydro-chemical, geological, and soil parameters as explanatory variables, this study employs multiple linear regression (MLIR) and random forest regression (RFR) models to estimate groundwater As concentrations in these regions. Additionally, random forest classification (RFC) and multivariate logistic regression (MLOR) models are applied to predict the probability of As levels exceeding 10 μg/L in the Hetao Basin (China) and Bangladesh. Model validation reveals that RFR explains 80% and 70% of spatial variability of As concentration in the Hetao Basin and Bangladesh, respectively, outperforming MLIR, which accounts for only 35% and 32%. Similarly, RFC outperforms MLOR in predicting high As probability, achieving correct classification rates of 98.70% (Hetao Basin) and 98.25% (Bangladesh) on training datasets, and 82.76% (Hetao Basin) and 91.20% (Bangladesh) on validation datasets. The performance of the MLOR model on the validation set yields accuracy rates of 81.60% and 72.18%, respectively. In the Hetao Basin, Ca2+, redox potential (Eh), Fe, pH, SO42−, and Cl− are key predictors of As contamination, while in Bangladesh, soil organic carbon (SOC), pH, and SO42− are significant predictors. This study underscores the potential of random forest (RF) models as robust tools for predicting groundwater As contamination.

Neural networks for the analysis of 2D radio-xenon beta gamma spectra

Atmospheric traces of radioactive xenon can be a strong indicator for underground nuclear fission reactions. 131mXe, 133Xe, 133mXe and 135Xe are the primary gaseous isotopes/isomers currently used to identify and classify nuclear events. During decay, each of these radioactive species produces a unique beta-gamma energy spectra, which can be measured using beta-gamma coincidence counting. Current operational Xe beta-gamma spectrum analysis software relies on Region of Interest (ROI) counting (Bowyer et al. in J Environ Radioact 59(2):139–151, 2002). This algorithm occasionally produces mismeasurements, especially when quantifying meta-stable isomers, due to overlapping ROIs and shifts in detector calibration in fielded systems over time (Ringbom and Axelsson in Appl Radiat Isot 156:108950, 2020). In an attempt to better de-convolve overlapping isotope spectra we have developed a technique that applies a supervised neural-network implemented in TensorFlow with Keras to classify and quantify the isotopes and mixtures of isomers based on their beta-gamma spectra—similar to image recognition. From this, we have improved upon the false-positive rate for classification and regression models, however challenges remain with dealing with differing detector energy calibrations and with estimating measurement uncertainty.

Racial Composition of Social Environments Over the Life Course Using the Pictorial Racial Composition Measure: Development and Validation Study.

Studies investigating the impact of racial segregation on health have reported mixed findings and tended to focus on the racial composition of neighborhoods. These studies use varying racial composition measures, such as census data or investigator-adapted questions, which are currently limited to assessing one dimension of neighborhood racial composition. This study aims to develop and validate a novel racial segregation measure, the Pictorial Racial Composition Measure (PRCM). The PRCM is a 10-item questionnaire of pictures representing social environments across adolescence and adulthood: neighborhoods and blocks (adolescent and current), schools and classrooms (junior high and high school), workplace, and place of worship. Cognitive interviews (n=13) and surveys (N=549) were administered to medically underserved patients at a primary care clinic at the Barnes-Jewish Hospital. Development of the PRCM occurred across pilot and main phases. For each social environment and survey phase (pilot and main), we computed positive versus negative pairwise comparisons: mostly Black versus all other categories, half Black versus all other categories, and mostly White versus all other categories. We calculated the following validity metrics for each pairwise comparison: sensitivity, specificity, correct classification rate, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, false positive rate, and false negative rate. For each social environment, the mostly Black and mostly White dichotomizations generated better validity metrics relative to the half Black dichotomization. Across all 10 social environments in the pilot and main phases, mostly Black and mostly White dichotomizations exhibited a moderate-to-high sensitivity, specificity, correct classification rate, positive predictive value, and negative predictive value. The positive likelihood ratio values were >1, and the negative likelihood ratio values were close to 0. The false positive and negative rates were low to moderate. These findings support that using either the mostly Black versus other categories or the mostly White versus other categories dichotomizations may provide accurate and reliable measures of racial composition across the 10 social environments. The PRCM can serve as a uniform measure across disciplines, capture multiple social environments over the life course, and be administered during one study visit. The PRCM also provides an added window into understanding how structural racism has impacted minoritized communities and may inform equitable intervention and prevention efforts to improve lives.

Estimation and inference on the partial volume under the receiver operating characteristic surface.

measures of biomarker accuracy that employ the receiver operating characteristic surface have been proposed for biomarkers that classify patients into one of three groups: healthy, benign, or aggressive disease. The volume under the receiver operating characteristic surface summarizes the overall discriminatory ability of a biomarker in such configurations, but includes cutoffs associated with clinically irrelevant true classification rates. Due to the lethal nature of pancreatic cancer, cutoffs associated with a low true classification rate for identifying patients with pancreatic cancer may be undesirable and not appropriate for use in a clinical setting. In this project, we study the properties of a more focused criterion, the partial volume under the receiver operating characteristic surface, that summarizes the diagnostic accuracy of a marker in the three-class setting for regions restricted to only those of clinical interest. We propose methods for estimation and inference on the partial volume under the receiver operating characteristic surface under parametric and non-parametric frameworks and apply these methods to the evaluation of potential biomarkers for the diagnosis of pancreatic cancer.