Probabilistic Prediction Research Articles

New Features in the pyCSEP Toolkit for Earthquake Forecast Development and Evaluation

Abstract The Collaboratory for the Study of Earthquake Predictability (CSEP) is a global community dedicated to advancing earthquake predictability research by rigorously testing probabilistic earthquake forecast models and prediction algorithms. At the heart of this mission is the recent introduction of pyCSEP, an open-source software tool designed to evaluate earthquake forecasts. pyCSEP integrates modules to access earthquake catalogs, visualize forecast models, and perform statistical tests. Contributions from the CSEP community have reinforced the role of pyCSEP in offering a comprehensive suite of tools to test earthquake forecast models. This article builds on Savran, Bayona, et al. (2022), in which pyCSEP was originally introduced, by describing new tests and recent updates that have significantly enhanced the functionality and user experience of pyCSEP. It showcases the integration of new features, including access to authoritative earthquake catalogs from Italy (Bolletino Seismico Italiano), New Zealand (GeoNet), and the world (Global Centroid Moment Tensor), the creation of multiresolution spatial forecast grids, the adoption of non-Poissonian testing methods, applying a global seismicity model to specific regions for benchmarking regional models and evaluating alarm-based models. We highlight the application of these recent advances in regional studies, specifically through the New Zealand case study, which showcases the ability of pyCSEP to evaluate detailed, region-specific seismic forecasts using statistical functions. The enhancements in pyCSEP also facilitate the standardization of how the CSEP forecast experiments are conducted, improving the reliability, and comparability of the earthquake forecasting models. As such, pyCSEP exemplifies collaborative research and innovation in earthquake predictability, supporting transparent scientific practices, and community-driven development approaches.

Construction and verification of pancreatic fistula risk prediction model after pancreaticoduodenectomy based on ensemble machine learning

Objective: To construct an ensemble machine learning model for predicting the occurrence of clinically relevant postoperative pancreatic fistula (CR-POPF) after pancreaticoduodenectomy and evaluate its application value. Methods: This is a research on predictive model. Clinical data of 421 patients undergoing pancreaticoduodenectomy in the Department of Pancreatic Surgery,Union Hospital, Tongji Medical College,Huazhong University of Science and Technology from June 2020 to May 2023 were retrospectively collected. There were 241 males (57.2%) and 180 females (42.8%) with an age of (59.7±11.0)years (range: 12 to 85 years).The research objects were divided into training set (315 cases) and test set (106 cases) by stratified random sampling in the ratio of 3∶1. Recursive feature elimination is used to screen features,nine machine learning algorithms are used to model,three groups of models with better fitting ability are selected,and the ensemble model was constructed by Stacking algorithm for model fusion. The model performance was evaluated by various indexes,and the interpretability of the optimal model was analyzed by Shapley Additive Explanations(SHAP) method. The patients in the test set were divided into different risk groups according to the prediction probability (P) of the alternative pancreatic fistula risk score system (a-FRS). The a-FRS score was validated and the predictive efficacy of the model was compared. Results: Among 421 patients,CR-POPF occurred in 84 cases (20.0%). In the test set,the Stacking ensemble model performs best,with the area under the curve (AUC) of the subject's work characteristic curve being 0.823,the accuracy being 0.83,the F1 score being 0.63,and the Brier score being 0.097. SHAP summary map showed that the top 9 factors affecting CR-POPF after pancreaticoduodenectomy were pancreatic duct diameter,CT value ratio,postoperative serum amylase,IL-6,body mass index,operative time,albumin difference before and after surgery,procalcitonin and IL-10. The effects of each feature on the occurrence of CR-POPF after pancreaticoduodenectomy showed a complex nonlinear relationship. The risk of CR-POPF increased when pancreatic duct diameter<3.5 mm,CT value ratio<0.95,postoperative serum amylase concentration>150 U/L,IL-6 level>280 ng/L,operative time>350 minutes,and albumin decreased by more than 10 g/L. The AUC of a-FRS in the test set was 0.668,and the prediction performance of a-FRS was lower than that of the Stacking ensemble machine learning model. Conclusion: The ensemble machine learning model constructed in this study can predict the occurrence of CR-POPF after pancreaticoduodenectomy,and has the potential to be a tool for personalized diagnosis and treatment after pancreaticoduodenectomy.

Early Detection of Residual/Recurrent Lung Malignancies on Post-Radiation FDG PET/CT

Positron Emission Tomography/Computed Tomography (PET/CT) using Fluorodeoxyglucose (FDG) is an important imaging modality for assessing treatment outcomes in patients with pulmonary malignant neoplasms undergoing radiation therapy. However, distinguishing between benign post-radiation changes and residual or recurrent malignancies on PET/CT images is challenging. Leveraging the potential of artificial intelligence (AI), we aimed to develop a hybrid fusion model integrating radiomics and Convolutional Neural Network (CNN) architectures to improve differentiation between benign post-radiation changes and residual or recurrent malignancies on PET/CT images. We retrospectively collected post-radiation PET/CTs with identified labels for benign changes or residual/recurrent malignant lesions from 95 lung cancer patients who received radiation therapy. Firstly, we developed separate radiomics and CNN models using handcrafted and self-learning features, respectively. Then, to build a more reliable model, we fused the probabilities from the two models through an evidential reasoning approach to derive the final prediction probability. Five-folder cross-validation was performed to evaluate the proposed radiomics, CNN, and fusion models. Overall, the hybrid fusion model outperformed the other two models in terms of sensitivity, specificity, accuracy, and the area under the curve (AUC) with values of 0.67, 0.72, 0.69, and 0.72, respectively. Evaluation results on the three AI models we developed suggest that handcrafted features and learned features may provide complementary information for residual or recurrent malignancy identification in PET/CT.

Research of multi-label text classification based on label attention and correlation networks.

Multi-Label Text Classification (MLTC) is a crucial task in natural language processing. Compared to single-label text classification, MLTC is more challenging due to its vast collection of labels which include extracting local semantic information, learning label correlations, and solving label data imbalance problems. This paper proposes a model of Label Attention and Correlation Networks (LACN) to address the challenges of classifying multi-label text and enhance classification performance. The proposed model employs the label attention mechanism for a more discriminative text representation and uses the correlation network based on label distribution to enhance the classification results. Also, a weight factor based on the number of samples and a modulation function based on prediction probability are combined to alleviate the label data imbalance effectively. Extensive experiments are conducted on the widely-used conventional datasets AAPD and RCV1-v2, and extreme datasets EUR-LEX and AmazonCat-13K. The results indicate that the proposed model can be used to deal with extreme multi-label data and achieve optimal or suboptimal results versus state-of-the-art methods. For the AAPD dataset, compared with the suboptimal method, it outperforms the second-best method by 2.05% ∼ 5.07% in precision@k and by 2.10% ∼ 3.24% in NDCG@k for k = 1, 3, 5. The superior outcomes demonstrate the effectiveness of LACN and its competitiveness in dealing with MLTC tasks.

IoT Sensor Based Cross-Basin Natural Ecological Environment QualityMonitoring and Modeling Simulation with Artificial Intelligence RemoteSensing and GIS

This study presents an integrated approach for monitoring and modelling the quality of crossbasin natural ecological environments using advanced techniques such as Markov Random Field Clustering Classification (MRF-CC), Geographic Information Systems (GIS), Hidden Markov Models (HMM), and remote sensing. With key ecological parameters including water quality, biodiversity, habitat suitability, and land cover types across various scenarios and locations within a study area. The simulation results from MRF-CC revealed the significant impacts of different environmental scenarios and management actions, with restoration efforts showing improvements in ecological quality, while pollution mitigation and urbanization pressures led to declines. The simulation results from MRF-CC revealed the significant impacts of different environmental scenarios and management actions. For example, restoration efforts (Scenario 1) improved water quality (pH 7.5), biodiversity index (0.88), and habitat suitability (0.78) compared to the baseline values (pH 7.2, biodiversity index 0.85, habitat suitability 0.75). Conversely, pollution mitigation (Scenario 2) and urbanization pressure (Scenario 8) resulted in declines, with water quality dropping to pH 7.0 and 6.9, biodiversity indices to 0.82 and 0.81, and habitat suitability to 0.72 and 0.71, respectively. GIS estimation provided spatial insights into ecological parameters, revealing variability across different locations. For instance, Point A (forest) exhibited a water quality index of 0.78, while Point D (urban) showed a lower index of 0.54. The integration of HMM offered probabilistic predictions of land cover dynamics, with probabilities ranging from 0.85 for forest at Point A to 0.45 for urban land cover at Point D.

Machine learning model outperforms the ACS Risk Calculator in predicting non-home discharge following primary total knee arthroplasty.

Despite the increase in outpatient total knee arthroplasty (TKA) procedures, many patients are still discharged to non-home locations following index surgery. The ability to accurately predict non-home discharge (NHD) following TKAs has the potential to promote a reduction in associated adverse events and excess healthcare costs. This study aimed to evaluate whether a machine learning (ML) model could outperform the American College of Surgeons (ACS) Risk Calculator in predicting NHD following TKA, using the same set of clinical variables. We hypothesised that the ML model would outperform the ACS Risk Calculator. Data from 365,240 patients who underwent a primary TKA between 2013 and 2020 were extracted from the ACS-National Surgical Quality Improvement Program database and used to develop an artificial neural network (ANN) to predict discharge disposition following primary TKA. The ANN and ACS calculator were assessed and compared using discrimination, calibration and decision curve analysis. Age (>68 years), BMI (>35.5 kg/m2) and ASA Class (≥2) were found to be the most important variables in predicting NHD following TKA. When compared to the ACS calculator, the ANN model demonstrated a significantly superior ability to distinguish the area under the receiver operating characteristic curve (AUC) among NHD patients and provided probability predictions well aligned with the true outcomes (AUCANN = 0.69, AUCACS = 0.50, p = 0.002, slopeANN = 0.85, slopeACS = 4.46, interceptANN = 0.04, and interceptACS = 0.06). Our findings support the hypothesis that machine learning models outperform the ACS Risk Calculator in predicting non-home discharge after TKA, even when constrained to the same clinical variables. Our findings underscore the potential benefits of integrating machine learning models into clinical practice for improving preoperative patient risk identification, optimisation, counselling and clinical decision-making. III.

A High‐Precision Dynamic Movement Recognition Algorithm Using Multimodal Biological Signals for Human–Machine Interaction

Accurate recognition of human dynamic movement is essential for seamless human–machine interaction (HMI) across various domains. However, most of the existing methods are single‐modal movement recognition, which has inherent limitations, such as limited feature representation and instability to noise, which will affect its practical performance. To address these limitations, this article proposes a novel fusion approach that can integrate two biological signals, including electromyography (EMG) and bioelectrical impedance (BI). The fusion method combines EMG for capturing dynamic movement features and BI for discerning key postures representing discrete points within dynamic movements. In this method, the identification of key postures and their temporal sequences provide a guiding framework for the selection and weighted correction of probability prediction matrices in EMG‐based dynamic recognition. To verify the effectiveness of the method, six dynamic upper limb movements and nine key postures are defined, and a Universal Robot that can follow movements is employed for experimental validation. Experimental results demonstrate that the recognition accuracy of the dynamic movement reaches 96.2%, representing an improvement of nearly 10% compared with single‐modal signal. This study illustrates the potential of multimodal fusion of EMG and BI in movement recognition, with broad prospects for application in HMI fields.

The instant probability of object with deterministic motion. Another way to understand quantum mechanics

Usually, the idea of probability is applied only to the random motion in physics. Here, the instant probability concept is obtained by applying the probability approach to objects with deterministic motion trajectories via a coin falling to the ground as an example. Applied this concept to understand quantum mechanics, the following conclusions are inferred: The wave function/quantum state represents an instant probability distribution of physical quantities after taking into account Planck's constant; the collapse of the wave function is a special case of any function on time-dependent probability prediction has to collapse when performing measurement. The arrangement of measurements will affect the observed results; Quantum mechanics reveals the existence of non-local effects, which have nothing to do with the interaction in classical physics. Quantum mechanics cannot answer questions such as whether the cat is dead or alive before opening the box. The correct answer to this kind of question has to be from classic physics.

Navigating the gray zone: Machine learning can differentiate malignancy in PI-RADS 3 lesions

IntroductionThe objective of this study is to predict the probability of prostate cancer in PI-RADS 3 lesions using machine learning methods that incorporate clinical and mpMRI parameters. MethodsThe study included patients who had PI-RADS 3 lesions detected on mpMRI and underwent fusion biopsy between January 2020 and January 2024. Radiological parameters (Apparent diffusion coefficient (ADC), tumour ADC/contralateral ADC ratio, Ktrans value, periprostatic adipose tissue thickness, lesion size, prostate volume) and clinical parameters (age, body mass index, total prostate specific antigen, free PSA, PSA density, systemic inflammatory index, neutrophil-lymphocyte ratio [NLR], platelet lymphocyte ratio, lymphocyte monocyte ratio) were documented. The probability of prostate cancer prediction in PI-RADS 3 lesions was calculated using 6 different machine-learning models, with the input parameters being the aforementioned variables. ResultsOf the 235 participants in the trial, 61 had malignant fusion biopsy pathology and 174 had benign pathology. Among 6 different machine learning algorithms, the random forest model had the highest accuracy (0.86±0.04; 95% CI 0.85–0.87), F1 score (0.91±0.03; 95% CI 0.91–0.92) and AUC value (0.92±0.06; 95% CI 0.88–0.90). In SHAP analysis based on random forest model, tumour ADC, tumour ADC/contralateral ADC ratio and PSA density were the 3 most successful parameters in predicting malignancy. On the other hand, systemic inflammatory index and neutrophil lymphocyte ratio showed higher accuracy in predicting malignancy than total PSA, age, free PSA/total PSA and lesion size in SHAP analysis. ConclusionAmong the machine learning models we developed, especially the random forest model can predict malignancy in PI-RADS 3 lesions and prevent unnecessary biopsy. This model can be used in clinical practice with multicentre studies including more patients.

Big Brother Is Watching You: Leveraging Artificial Intelligence for Automated Fuel Cell Monitoring (Technical Report)

In this technical report, we present our second-generation digital twin model for monitoring of fuel cells. This digital twin combines two data-driven machine learning models: a stationary model for stationary cell voltage prediction and a degradation model for degradation correction. This combination of both models results in a precise probabilistic prediction of the cell voltage over time.Furthermore, we present a web-based framework for automated fuel cell monitoring in real time. This framework allows training a digital twin on existing data and subsequently applying the twin to automatically check new data for anomalies.

Big Brother Is Watching You: Leveraging Artificial Intelligence for Automated Fuel Cell Monitoring (Technical Report)

In this technical report, we present our second-generation digital twin model for monitoring of fuel cells. This digital twin combines two data-driven machine learning models: a stationary model for stationary cell voltage prediction and a degradation model for degradation correction. This combination of both models results in a precise probabilistic prediction of the cell voltage over time.Furthermore, we present a web-based framework for automated fuel cell monitoring in real time. This framework allows training a digital twin on existing data and subsequently applying the twin to automatically check new data for anomalies.

Automating urban soundscape enhancements with AI: In-situ assessment of quality and restorativeness in traffic-exposed residential areas

Formalized in ISO 12913, the “soundscape” approach is a paradigmatic shift towards perception-based urban sound management, aiming to alleviate the substantial socioeconomic costs of noise pollution to advance the United Nations Sustainable Development Goals. Focusing on traffic-exposed outdoor residential sites, we implemented an automatic masker selection system (AMSS) utilizing natural sounds to mask (or augment) traffic soundscapes. We employed a pre-trained AI model to automatically select the optimal masker and adjust its playback level, adapting to changes over time in the ambient environment to maximize “Pleasantness”, a perceptual dimension of soundscape quality in ISO 12913. Our validation study involving (N=68) residents revealed a significant 14.6% enhancement in “Pleasantness” after intervention, correlating with increased restorativeness and positive affect. Perceptual enhancements at the traffic-exposed site matched those at a quieter control site with 6dB(A) lower LA,eq and road traffic noise dominance, affirming the efficacy of AMSS as a soundscape intervention, while streamlining the labour-intensive assessment of “Pleasantness” with probabilistic AI prediction.

EXPRESS: Stratified Distributional Analysis - a Novel Perspective on RT Distributions.

Response times and their distributions serve as a powerful lens into cognitive processes. We present a novel statistical methodology called Stratified Distributional Analysis (SDA) to quantitatively assess how key determinants of response times (word frequency and length) shape their distributions. Taking advantage of the availability of millions of lexical decision response times in the English Lexicon Project and the British Lexicon Project, we made important advances into the theoretical issue of linking response times and word frequency by analysing RT distributions as a function of word frequency and word length. We tested these distributions against the lognormal, Wald, and Gamma distributions and three measures of word occurrence (word form frequencies obtained from subtitles and contextual diversity as operationalized as discourse contextual diversity and user contextual diversity). We found that the RT distributions were best described by a lnorm distribution across both megastudies when word occurrence was quantified by a contextual diversity measure. The link between the lnorm distribution and its generative process highlights the power of SDA in elucidating mechanisms that govern the generation of RTs through the fitting of probability distributions. Using a hierarchical Bayesian framework, SDA yielded posterior distributions for the distributional parameters at the single-participant level, enabling probabilistic predictions of response times as a function of word frequency and word length, which has the potential to serve as a diagnostic tool to uncover idiosyncratic features of word processing. Crucially, while we applied our parsimonious methodology to lexical decision response times, it is applicable to a variety of tasks such as word-naming and eye-tracking data.

Development and validation of asthma diagnostic scale for children.

To construct an asthma diagnostic scale for children under 6 years old. An electronic medical record database was used to develop the scale. Item pool was established through literature survey and expert opinion. Items were screened and optimized by using the Delphi method, t-test, reactivity analysis, Pearson correlation coefficient, factor analysis, reliability and validity test. The predictive probability of asthma was calculated using logistic regression, and receiver operating characteristic curve. Another childhood asthma database was used to validated the scale. The asthma diagnostic scale for children under 6 years old included five dimensions: dimension 1 (shortness of breath, three concave sign, cyanosis, moist rale, heart rhythm, heart sound and dyspnea), dimension 2 (respiratory sound, cough, sputum), dimension 3 (frequency of wheezing, allergic rhinitis, history of allergy in one or both parents), dimension 4 (sex, wheezing, atopic dermatitis), and dimension 5 (reversible airflow, positive in vitro or in vivo allergy test). The Cronbach's α coefficients for the five dimensions were 0.846, 0.459, 0.019, 0.202, and 0.024. The area under the ROC curve (AUC), sensitivity, and specificity were 0.737, 59.1%, and 81.2%. AUC, sensitivity, and specificity in the validation database were 0.614, 76.2%, and 46.7%. The scale has significant diagnostic value for asthma in children under 6 years old. 1. The aim of the study was to establish an asthma diagnosis scale for children younger than 6 years old. 2. Our study not only addresses the lack of diagnostic criteria of young children asthma, but also indicates the accuracy of the diagnostic scale. 3. The data may help to reduce the missed diagnosis and misdiagnosis rate of asthma, and improve the diagnostic accuracy of the disease, and thus reduces the harm of asthma to children's physical and mental health.

Time series forecasting using transformer neural network

Time series forecasting is a vital component of data science, giving essential insights that help decision-makers to predict future trends across a number of sectors. This paper focuses on projecting complicated stock market price dynamics, weather data variations, and hourly traffic occupancy rates. The Informer-Based Time Series Forecasting (IBTSF) model, which employs a transformer-based architecture to capture long-term relationships and give probabilistic predictions, is proposed.This model is assessed against Long Short-Term Memory (LSTM), Temporal Convolutional Network (TCN) and Gated Recurrent Unit (GRU) model to benchmark its performance. Simulation findings demonstrate that the IBTSF model considerably outperforms other models across all datasets. For traffic data, the IBTSF model achieves a Mean Squared Error (MSE) of 10.1 and a Mean Absolute Scaled Error (MASE) of 1.69, compared to other model's larger mistakes. In stock data forecasting, the IBTSF model reports an MSE of 8.04 and a MASE of 1.53, again outperforming other models. Similarly, using meteorological data, the IBTSF model obtains an MSE of 86.12 and a MASE of 0.7, whereas other models fall short. These results underline the IBTSF model's resilience and accuracy in capturing subtle patterns and relationships within time series data, giving precise insights that benefit decision-making processes.

The significance of P.V. Simonov’s “need-information theory of emotions” in the development of modern neurobiology of behavior

The current paper shows the importance of the need-information theory of P.V. Simonov in the development of modern neurobiology of behavior. The essence of the theory and the underlying fundamental principles of the organization of behavior-environmental uncertainty, probabilistic predictions of reinforcement are briefly described. The first section reviews the current data on the important role of uncertain environments and probabilistic predictions in organization of behavior. Attention is drawn to the reinforcement prediction error and its significance in the organization of both social and individual behavior, as well as its role in the consolidation and reconsolidation of memory. The second section shows the influence of need-information theory on the development of the theoretical and experimental basis of individual differences, with a scheme presented for explaining such differences based on the fundamental principles of theory. The next section examines the role of need-information theory in understanding the mechanisms of decision-making under risk conditions, and the importance of the theory as a conceptual basis for the new developing field of science – neuroeconomics. And finally, the 4th section considers in detail the model of emotional resonance (contagion) proposed by P. Simonov, and modern views on social behavior, in general, and the altruistic and selfish behavior of rodents, in particular.

Prediction of human error probability in helicopter to ship transfer operation under an evidential reasoning extended CREAM approach.

Although the helicopter to ship transfer operation is rarely applied in ports around the world, some of the hub ports have encouraged it. The operational process poses challenges to safety due to the nature of the work to be completed by the ship crew. This article performs a systematic human error prediction to ensure safe helicopter to ship transfer operations under different circumstances since the human factor is becoming a significant contributor to maritime accidents. To accomplish this, the article addresses a cognitive reliability and error analysis method (CREAM) under the evidential reasoning (ER) approach. In the proposed approach, the CREAM provides a comprehensive human error prediction tool and ER is capable of supporting experts' judgment in decision-making. The findings of the research show that 'Secure the loose objects within or adjacent to the operating area' poses a high human error probability (4.50E-02) during helicopter to ship transfer operations.

Secular and modulator-specific drifts in the predictive performance of a rapid lung function decline algorithm: a cystic fibrosis patient registry study

PurposePredicting the onset of pulmonary exacerbation for people with cystic fibrosis (CF) is critical to identify those at high risk in advance and allows timely clinical intervention. The FEV1-indicated exacerbation signal (FIES) has been proposed by the CF Learning Network to identify pulmonary exacerbation events, but temporal predictive performance of this marker is unknown.MethodsTo assess the epidemiologic impact of secular trends in CF care and treatments on algorithms aimed at prediction of pulmonary exacerbation, we evaluated prediction model “drift” using a longitudinal retrospective cohort of 29,476 individuals from the U.S. CF Foundation Patient Registry (2011–2018). The FIES marker was defined for each clinical encounter using relative drops in lung function, measured as forced expiratory volume in 1 s of % predicted (FEV1pp). We formed predictive probabilities for each FIES event via a target function for a predictive algorithm of FEV1pp. Models were trained using 2 year data windows with prediction performance tested on subsequent years.ResultsYear-over-year FIES event discrimination remained consistent (area under the receiver-operator characteristic curve, AUC > 80%). Annual drift was < 1% with actual drop in AUC ~ 3% in 6 years. The model performed best in individuals with moderate-to-low baseline FEV1pp (AUC range: 84–85%). For relatively higher FEV1pp levels, discrimination was slightly lower with higher AUC variability.ConclusionWhile model fit improved by accounting for modulator therapy, prediction performance was similar. Most impactful model drifts occurred after 2012 coinciding with increased ivacaftor modulator use. Additional evaluation will be necessary given recent, broad uptake of highly effective modulators.

ScPanel: a tool for automatic identification of sparse gene panels for generalizable patient classification using scRNA-seq datasets.

Single-cell RNA sequencing (scRNA-seq) technologies can generate transcriptomic profiles at a single-cell resolution in large patient cohorts, facilitating discovery of gene and cellular biomarkers for disease. Yet, when the number of biomarker genes is large, the translation to clinical applications is challenging due to prohibitive sequencing costs. Here, we introduce scPanel, a computational framework designed to bridge the gap between biomarker discovery and clinical application by identifying a sparse gene panel for patient classification from the cell population(s) most responsive to perturbations (e.g. diseases/drugs). scPanel incorporates a data-driven way to automatically determine a minimal number of informative biomarker genes. Patient-level classification is achieved by aggregating the prediction probabilities of cells associated with a patient using the area under the curve score. Application of scPanel to scleroderma, colorectal cancer, and COVID-19 datasets resulted in high patient classification accuracy using only a small number of genes (<20), automatically selected from the entire transcriptome. In the COVID-19 case study, we demonstrated cross-dataset generalizability in predicting disease state in an external patient cohort. scPanel outperforms other state-of-the-art gene selection methods for patient classification and can be used to identify parsimonious sets of reliable biomarker candidates for clinical translation.

Abstract B007: Racial/ethnic and socioeconomic disparities in over- and under-screening of cervical cancer among a national sample of commercially insured individuals

Abstract In the United States (US), disparities in cervical cancer screening uptake persist by race/ethnicity and socioeconomic status (SES). However, few have examined how race/ethnicity and SES contribute to specific guideline adherence patterns, which can exacerbate existing disparities. We investigated the impact of race/ethnicity and SES (education and household net worth [HHN]) on the probability of over- and under-screening. Optum Clinformatics Data Mart from 2012-2021 was used to identify average-risk females aged 30-64 and ≥1 cervical cancer screening with continuous health plan enrollment and no history of hysterectomy/cervical cancer. Over- and under-screening were defined as intervals between the index and subsequent screening which were shorter and longer than the interval recommended by the 2012 US Preventive Services Task Force guideline, respectively. Multivariate logistic regression was used to evaluate the association of race/ethnicity (Non-Hispanic [NH]-White, NH-Asian, Hispanic, NH-Black) and SES including HHN (&amp;lt;$250,000, $250,000-$500,000, &amp;gt;$500,000) and education level (&amp;lt;Bachelor’s Degree, &amp;gt;Bachelor’s Degree) with over- and under-screening. We tested for interactions between race/ethnicity and SES. Of the 686,884 eligible patients, 72.8% were NH-White, 71.4% had &amp;lt;Bachelor's Degree and 43.3% had HHN &amp;lt;$250,000. Only 7.2% were guideline adherent, 30.7% were under-screened, and 62.1% were over-screened. After regression adjustment, the odds of under-screening were higher among NH-Asian (OR, 1.2; 95%CI, 1.1- 1.2), NH-Black (OR, 1.1; 95%CI, 1.0-1.1), and Hispanic (OR, 1.2; 95%CI, 1.1-1.2) vs NH-White, &amp;lt;Bachelor’s degree (OR, 1.1; 95%CI, 1.0-1.1) vs &amp;gt;Bachelor’s degree, and HHN &amp;lt;$250,000 (OR, 1.3 95%CI, 1.2-1.3) and $250,000-$499,000 (OR, 1.1; 95%CI, 1.0-1.1) vs &amp;gt;$500,000. Odds of over-screening were higher among NH-Black (OR, 1.3; 95%CI, 1.2-1.3), and Hispanic (OR, 1.2; 95%CI, 1.1-1.2) vs NH-White, and lower among &amp;lt;Bachelor’s degree (OR, 0.9; 95%CI, 0.8-0.9) vs &amp;gt;Bachelor’s degree, and HHN &amp;lt;$250,000 (OR, 0.9; 95%CI, 0.9-1.0) and $250,000-$499,000 (OR, 0.9; 95%CI, 0.9-1.0) vs &amp;gt;$500,000. For under-screening, the interaction between race/ethnicity and HHN was significant, and the predicted probability (PP) increased with higher HHN among Hispanic but not in other groups (p&amp;lt;0.01). For over-screening, the interaction between race/ethnicity and HHN was significant and the PP increased with higher HHN among Hispanic and NH-White but decreased for NH-Black and NH-Asian. The interaction between race/ethnicity and education was significant and the PP increased with higher education among Hispanic and NH-White but decreased among NH-Black and NH-Asian. Even within a commercially insured population, we observed that the majority of patients received non-guideline adherent screening and SES contributed to disparities differently among racial/ethnic groups. Findings indicate the need to tailor interventions and policies for subpopulations experiencing over- and under-screening to address the overall screening quality and disparities in outcomes. Citation Format: Michelle B. Shin, Sarah Axeen, Jennifer Tsui. Racial/ethnic and socioeconomic disparities in over- and under-screening of cervical cancer among a national sample of commercially insured individuals [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr B007.