Early Prediction Research Articles

To what extent has machine learning achieved in predicting online at-risk students? Evidence from quantitative meta-analysis

Machine learning and data mining techniques hold promise in predicting at-risk students in online learning. This meta-analysis aimed to provide quantitative evidence to validate whether and to what extent machine learning techniques have been achieved in identifying online at-risk students. Meta-regressions examined the impacts of predictor data types, classical or deep learning approaches, and prediction stages on performance. A random-effects meta-analysis of 47 studies with 309 models showed good classification accuracy, higher in summative than early predictions. Deep learning models and diverse predictors can significantly enhance model performance. No publication bias was detected. Implications and recommendations for practice are discussed.

Predictors of Post-Hepatectomy Liver Failure in Klatskin Tumors: The Role of Preoperative Glucose, Future Liver Remnant to Spleen Ratio, and Early Bilirubin Monitoring

Background: Post-hepatectomy liver failure (PHLF) is a serious complication following hepatic resection for Klatskin tumors, significantly affecting patient prognosis. Identifying reliable preoperative and early postoperative predictors of PHLF can help optimize patient outcomes and guide surgical planning. Method: We conducted a retrospective review of 34 patients who underwent hemi-hepatectomy for extrahepatic cholangiocarcinoma at Kosin University Gospel Hospital between April 2019 and April 2024, and at Chonnam National University Hwasun Hospital between September 2017 and April 2024. Demographics, laboratory data, and volumetric measurements including spleen volume, were analyzed to assess their roles in predicting PHLF. Logistic regression and receiver operating characteristic (ROC) curve analyses were performed to evaluate the predictive value of these factors. Results: Elevated preoperative glucose levels and reduced future liver remnant to spleen ratio (FLR/SV) were significantly associated with an increased risk of PHLF. Additionally, elevated postoperative day 1 total bilirubin (POD 1 TB) was identified as a key postoperative predictor of PHLF. Multivariable analysis confirmed the significance of these factors, with FLR/SV, preoperative glucose, and POD 1 TB demonstrating good discriminative ability in ROC analysis, with AUC values of 0.779, 0.782, and 0.786 respectively. Conclusion: Preoperative glucose control, evaluation of FLR/SV, and early postoperative monitoring of TB are useful for improving outcomes in patients undergoing major hepatectomy for Klatskin tumors.

Factors associated with predicting knee pain using knee X-ray and personal factors: A multivariate logistic regression and XGBoost model analysis from the Nationwide Korean Database (KNHANES).

With increasing life expectancy, knee pain has become more prevalent, highlighting the need for early prediction. Although X-rays are commonly used for diagnosis, knee pain and X-ray findings do not always match. This study aims to identify factors contributing to knee pain in individuals with both normal and abnormal knee X-ray results to bridge the gap between X-ray findings and knee pain. Data from the fifth Korea National Health and Nutrition Examination Survey (KNHANES), collected from 2010 to 2012, including data from 5,191 participants, were analyzed. The focus was on epidemiological characteristics, medical histories, knee pain, and X-ray grades. Multivariate logistic regression and extreme gradient boosting (XGBoost) models were used to predict knee pain in individuals with normal and abnormal knee X-rays, categorized by Kellgren-Lawrence grades. For normal X-rays, the logistic regression model identified aging, being female, higher BMI, lower fat percentage, osteoporosis, depression, and rural living as factors associated with knee pain. The XGBoost model highlighted BMI, age, and sex as key predictors, with a feature importance >0.1. For abnormal X-rays, logistic regression indicated that aging, being female, higher BMI, osteoporosis, depression, and rural living were associated with knee pain. The XGBoost model highlighted age, BMI, sex, and osteoporosis as key predictors, with a feature importance >0.1. Aging and being female were associated with knee pain due to hormonal changes in women, as well as cartilage and bone deterioration. Lower fat percentage was significantly associated with increased pain, which might be attributable to higher activity levels. Higher BMI and osteoporosis were significantly associated with knee pain, possibly due to increased stress and reduced resistance on knee structures, respectively. Depression was identified as a key predictor of knee pain in patients with normal X-rays, potentially attributable to psychosomatic factors. The study's limitations include its cross-sectional nature, which does not allow for the establishment of causal relationships, the lack of detailed medical history such as trauma history, and recall bias due to self-reported questionnaires. Future research should address these limitations to support our hypothesis.

An evolving machine-learning-based algorithm to early predict response to anti-CGRP monoclonal antibodies in patients with migraine.

The present study aimed to determine whether machine-learning (ML)-based models can predict 3-, 6, and 12-month responses to the monoclonal antibodies (mAbs) against the calcitonin gene-related peptide (CGRP) or its receptor (anti-CGRPmAbs) in patients with migraine using early predictors (up to one month) and to create an evolving prediction tool. In this prospective cohort study, data from patients with migraine who had received anti-CGRP mAbs for 12 months were collected. Demographic and monthly clinical variables were collected, including monthly headache days (MHDs), days with acute medication use, number of analgesics and Headache Impact Test-6. Response rates were categorized as <25%, 26-50%, 51-75% and >75% reduction in MHDs. ML models were trained using random forest algorithm and optimized to maximize the F1 score. ML model performance was also evaluated using standard evaluation metrics, including accuracy, precision and area under the receiver operating characteristic curve (AUC-ROC). Sequential backward feature selection was employed to identify the most relevant predictors for each model. Each model was given 11 baseline data inputs and month-based predictors for months 1, 3 and 6. Each model was then validated against an external test cohort of patients who had received anti-CGRP mAbs for 12 months. Three hundred thirty-six patients treated with anti-CGRP mAbs were included. The external cohort included 93 patients treated with anti-CGRP mAbs. We developed six models to predict 3- 6- and 12-month responses using early predictors. ML-based models yielded predictions with an accuracy score in the range 0.40-0.73 and an AUC-ROC score in the range 0.56-0.76 during internal testing and yielding predictions with an accuracy in the range 0.39-0.64 and an AUC-ROC score in the range 0.52-0.78 when tested against an external test cohort. Shapley Additive explanations summary plots were generated to interpret the contribution of each feature for each model. Based on these findings, a response prediction tool was developed. Each model was run through a backward feature selection to find the most relevant features for the models. The MHDs reduction of the previous data point tends to be the most relevant, while the migraine with aura indicator tends to be the least effective predictor. The response prediction tool utilizing evolving ML-based models holds promise in the early prediction of treatment outcomes for patients with migraine undergoing anti-CGRP mAbs treatment.

A Spatial-Temporal Method for Early Prediction of Fatigue Crack Region and Orientation in Metallic Cellular Materials Using In-Situ Infrared Thermography (IRT)

This study seeks an early prediction method of crack failure location and orientation due to low cycle fatigue in additively manufactured metallic cellular materials by leveraging experimentally observed accumulation of plastic deformation. To study this, a novel spatial-temporal approach for analyzing Infrared Thermographic (IRT) video was developed to detect heat generated by local plastic deformation. The method was validated experimentally by conducting fully reversed low cycle fatigue tests of Inconel 718 (IN718) honeycomb specimens manufactured using Laser Powder Bed Fusion (LPBF). Using the approach developed, results showed that localized heating due to plastic work could be detected and used for early prediction of the most probable path, and orientation of crack propagation. Furthermore, the method developed was found to be able to predict these results within the first 1.5% of the total life of the specimen apriori to crack initiation.

The role of phenology in crop yield prediction: Comparison of ground-based phenology and remotely sensed phenology

Precise and timely crop yield predictions at large scales are crucial for safeguarding global food security. A key factor in accurate yield forecasting is the integration of multi-source environmental data, where the choice of time windows for aggregating these variables plays a pivotal role. Segmenting time windows by phenological stages allows for more precise extraction of environmental variables, capturing the complex interactions between crop development and external factors. However, the effectiveness of this approach in improving yield prediction accuracy, especially when comparing ground-based and remote sensing-derived land surface phenology data, remains largely unexplored. In this study, we investigate how phenology-based time windows affect corn yield predictions, using machine learning algorithms and multi-source environmental data from the U.S. Corn Belt. We systematically analyzed and compared models incorporating either ground-based or land surface phenology. By segmenting the growing season into six crucial stages (BBCH 00-99) and formulating specific yield forecasting models for each stage, we determined the optimal lead times for predictions utilizing both sources of phenological data. Our findings suggested that the incorporation of phenology-derived crop growth windows significantly enhances the accuracy of yield prediction by approximately 10 % compared to the fixed-season method. Ground-based phenology data from the USDA crop progress report slightly outperformed MODIS-based land surface phenology data, achieved the highest accuracy with the XGBoost model (R2 = 0.668, RMSE = 1.09 t/ha, MAE = 0.84 t/ha). Furthermore, this study demonstrated that reliable corn yield predictions could be made as early as the second phenological stage (Emerged-Silking for ground phenology, MidGreenup-Maturity for MODIS phenology). In regions where ground observations are limited or unavailable, land surface phenology emerges as a promising alternative. This study presents a robust framework for precise and extensive crop yield modeling and early prediction, which is crucial for making informed agricultural decisions and ensuring food security.

Biopsychosocial predictors of rapid weight gain from birth to 6 months.

Childhood obesity remains a public health crisis and identification of unique prenatal and early infancy predictors of obesity risk are critically needed. We test a comprehensive biopsychosocial model of the predictors of rapid weight gain (RWG) in the first 6 months of life. Two hundred and ninety nine pregnant women and their infants participated. Maternal prenatal psychobiological risk (PPBR) was assessed during the third trimester via maternal anthropometrics, serum biomarkers (insulin, leptin, adiponectin), and maternal report of pregnancy complications, substance use, mental health and stress. Infant stress reactivity was measured at 2 months (cortisol output, resting RSA, observed irritability, negative emotionality). At 2 and 6 months, maternal self-report of obesogenic feeding practices and observed maternal sensitivity during three tasks were collected. RWG was classified based on change in weight-for-age z scores from birth to 6 months (>0.67 SD). Obesogenic feeding practices predicted greater likelihood of RWG, β = 0.30, p = .0.01, independent of other predictors and covariates. Obesogenic feeding practices was the only proposed intervening mechanism that produced a significant indirect effect of PPBR on RWG, b = 0.05, S.E. = 0.04, 95% CI [0.002, 0.15], β = 0.06. Identifying proclivity towards obesogenic feeding practices and providing support to reduce these behaviours may enhance childhood obesity prevention efforts.

An Empirical-Informed Model for the Early Degradation Trajectory Prediction of Lithium-ion Battery

An Empirical-Informed Model for the Early Degradation Trajectory Prediction of Lithium-ion Battery

Predicting adolescent psychopathology from early life factors: A machine learning tutorial

ObjectiveThe successful implementation and interpretation of machine learning (ML) models in epidemiological studies can be challenging without an extensive programming background. We provide a didactic example of machine learning for risk prediction in this study by determining whether early life factors could be useful for predicting adolescent psychopathology. MethodsIn total, 9643 adolescents ages 9–10 from the Adolescent Brain and Cognitive Development (ABCD) Study were included in ML analysis to predict high Child Behavior Checklist (CBCL) scores (i.e., t-scores ≥ 60). ML models were constructed using a series of predictor combinations (prenatal, family history, sociodemographic) across 5 different algorithms. We assessed ML performance through sensitivity, specificity, F1-score, and area under the curve (AUC) metrics. ResultsA total of 1267 adolescents (13.1 %) were found to have high CBCL scores. The best performing algorithms were elastic net and gradient boosted trees. The best performing elastic net models included prenatal and family history factors (Sensitivity 0.654, Specificity 0.713; AUC 0.742, F1-score 0.401) and prenatal, family, history, and sociodemographic factors (Sensitivity 0.668, Specificity 0.704; AUC 0.745, F1-score 0.402). Across all 5 ML algorithms, family history factors (e.g., either parent had nervous breakdowns, trouble holding jobs/fights/police encounters, and counseling for mental issues) and sociodemographic covariates (e.g., maternal age, child's sex, caregiver income and caregiver education) tended to be better predictors of adolescent psychopathology. The most important prenatal predictors were unplanned pregnancy, birth complications, and pregnancy complications. ConclusionOur results suggest that inclusion of prenatal, family history, and sociodemographic factors in ML models can generate moderately accurate predictions of adolescent psychopathology. Issues associated with model overfitting, hyperparameter tuning, and system seed setting should be considered throughout model training, testing, and validation. Future early risk predictions models may improve with the inclusion of additional relevant covariates.

Predictors of attachment in early- and late-placed adoptees

Although early experiences with primary caregivers have been recognised to play a significant role in individuals’ attachment relationships, little is known about adoptees’ attachment trajectories once placed within their respective adoptive families. This study aims to address this by investigating how attachment representations might be explained by early predictors including childhood attachment representations and other pertinent pre-placement variables. In early childhood, the Story Stem Assessment Profile (SSAP) is used to capture these representations, whilst in adolescence, an attachment-orientated interview (Friends and Family Interview; FFI) is employed. The sample here consisted of 35 early adoptees, placed before 12 months of age, and 35 maltreated, late-adoptees, placed between four and nine years old. Regression models were statistically significant for three out of four FFI constructs ( Security, Disorganisation, Coherence), though most of these effects were driven by IQ. When exploring the late adoptees alone, all regression models were significant in explaining variance in adolescent attachment, with SSAP Defensive Avoidance and Abuse as the strongest predictors. When examining the early adoptees alone, only two models were statistically significant, with most of these effects from IQ. These findings add to current knowledge on the role of childhood variables in predicting attachment and suggest potential areas to improve adoptees’ outcomes.

MiRNAs and Neutrophil-Related Membrane Proteins from Plasma-Derived Extracellular Vesicles for Early Prediction of Organ Dysfunction and Prognosis in Septic Patients.

The pathogenesis of sepsis-induced organ dysfunction remains elusive, and the mortality remains alarmingly high. We sought to investigate the profile of extracellular vesicles (EVs)-mediated communication between plasma and polymorphonuclear neutrophils (PMNs) in sepsis, and to elucidate whether miRNAs and PMN-related membrane proteins from plasma-derived EVs (plasma-EVs) are associated with sepsis-induced organ dysfunction and prognosis. PMN-derived EVs (PMN-EVs) were isolated from the blood samples of healthy controls (N=3) and patients with septic shock (N=3) after ICU admission. We performed miRNA sequencing of the isolated EVs, followed by bioinformatic analysis. A miRNA model for comparing PMN-EVs and plasma-EVs was successfully established in the training cohort. Furthermore, miRNAs and PMN-related membrane proteins from the plasma-EV model were confirmed in the validation cohort. A logistic regression model, receiver operating characteristic (ROC) curves, and Kaplan-Meier analyses were performed to evaluate the efficiency of diagnostic and/or prognostic performance. Further, in vivo and in vitro experiments were conducted to explore the involvement of plasma-EVs in PMNs autophagy. Fifty-five miRNAs from PMN-EVs differed significantly between the healthy controls and patients with septic shock. Furthermore, the plasma-EV model (six miRNAs and eight PMN-related membrane proteins) was confirmed in the validation cohort, demonstrating that miR-34a-5p, miR-503-5p, miR-4772-3p, ITGAM, MPO, and MMP9 serve as sepsis biomarkers for distinguishing lung, liver, and kidney dysfunction. Kaplan-Meier survival analysis showed that miR-34a-5p, miR-4772-3p, ITGAM, and MMP9 were potential prognostic predictors. Finally, we found that plasma-EVs from sepsis patients exert an inhibitory effect on PMNs autophagy, which can be reversed by EV inhibitors such as GW4869 and enoxaparin. These findings suggest that miRNAs and PMN-related membrane proteins from plasma-EVs could be valuable diagnostic tools for identifying sepsis-induced organ dysfunction and predicting prognosis, enabling proactive management of sepsis by physicians and improving the prognosis of sepsis patients.

Infrared thermographic assessment of cutaneous temperature changes during labour epidural analgesia initiation: An observational pilot study

BackgroundInadequate labour epidural analgesia within 30––45 min of insertion is typically considered primary epidural failure and deemed unsuccessful. Early prediction of analgesic success would have clinical benefit, enabling corrective interventions. Sympathectomy accompanying a successful epidural block can cause foot temperature changes. We sought to evaluate the relationship between foot temperature changes and successful epidural, defined as adequate analgesia in 45 min of epidural administration. For the purpose of this study, we defined primary epidural failure as inadequate analgesia within 45 min of administration. MethodsFollowing ethical committee approval, this observational study was conducted in a tertiary level obstetric centre between January 2021 and March 2024. Patients attending for induction of labour were included; with labour epidural analgesia comprised the epidural study group and without acted as a control group. A FlirT540 infrared thermography camera was used to take bilateral foot images every 5 min for 30 min. Images were taken following epidural administration in the epidural group and at any convenient time once in established labour in the control group. We studied temperature changes over time in the epidural and control groups and compared the temperature changes in patients with successful and failed epidurals. ResultsThirty eight patients were included in the epidural group and eleven in the control group. 29 patients (76.3 %) had successful analgesia and 9 patients (23.7 %) had primary epidural failure. Patients with successful analgesia had a statistically significant rise in temperature after 10 min. The maximum rate of rise was between 5 and 15 min and was 0.5 (0.1)°C per minute on both sides. Primary epidural failure was associated with minimal temperature changes, while all patients who had a 2 °C rise in hallux temperature at 10 min had successful analgesia. ConclusionsSympathetic blockade associated with a successful analgesic epidural block produces cutaneous temperature elevation at the plantar surface of the hallux. These can be detected within ten minutes of instituting an epidural and have a potential role in guiding timely troubleshooting of an epidural.

A Data-Driven and Hybrid Approach for Real-Time Threshold-Based Landslide Prediction in Hilly Regions using Sensor Networks

ABSTRACT Landslides pose a constant threat to the Himalayan regions, causing severe harm to people and property, and disastrous consequences demand novel ways of early prediction and prevention. The study gives a comprehensive insight into real-time landslide prediction using data from several sensors. The study investigates the applicability of advanced machine learning algorithms such as Multiple Linear Regression (MLR), XGBoost (XGB), Random Forest (RF), and a hybrid MLR-LSTM model in predicting landslide disasters. Further, the performance of these models was evaluated by utilizing the evaluation metrics such as MSE, MAE, and RMSE. However, the hybrid MLR-LSTM model excels with MSE = 0.014, MAE = 0.140, and RMSE = 0.120. Additionally, the study demonstrates the hybrid model’s practical application by demonstrating the real-time alert generation method.

Self-regulation of slow cortical potential and seizure suppression by scalp electroencephalography: Early prediction of therapeutic efficacy

Self-regulation of slow cortical potential and seizure suppression by scalp electroencephalography: Early prediction of therapeutic efficacy

Multimodal transformer for early alarm prediction

Alarms are an essential part of distributed control systems designed to help plant operators keep the processes stable and safe. In reality, however, alarms are often noisy and thus can be easily overlooked. Early alarm prediction can give the operator more time to assess the situation and introduce corrective actions to avoid downtime and negative impact on human safety and environment. Existing studies on alarm prediction typically rely on signals directly coupled with these alarms. However, using more sources of information could benefit early prediction by letting the model learn characteristic patterns in the interactions of signals and events. Meanwhile, multimodal deep learning has recently seen impressive developments. Combination (or fusion) of modalities has been shown to be a key success factor, yet choosing the best fusion method for a given task introduces a new degree of complexity, in addition to existing architectural choices and hyperparameter tuning. This is one of the reasons why real-world problems are still typically tackled with unimodal approaches. To bridge this gap, we introduce a multimodal Transformer model for early alarm prediction based on a combination of recent events and signal data. The model learns the optimal representation of data from multiple fusion strategies automatically. The model is validated on real-world industrial data. We show that our model is capable of predicting alarms with the given horizon and that the proposed multimodal fusion method yields state-of-the-art predictive performance while eliminating the need to choose among conventional fusion techniques, thus reducing tuning costs and training time.

Predictive Capacity of Screening Methods for Preeclampsia in Early Pregnancies

Background: Preeclampsia (PE) is a multiorgan and multifactorial disorder that occurs worldwide and is responsible for maternal mortality between 10 to 15% as a direct obstetric cause. Its early prediction would allow us to act in a timely manner to avoid possible complications. Objective: To evaluate the predictive capacity of screening methods. Methods: It is a prospective longitudinal study. the mean gestational age by LCC was 12.6 weeks. in which clinical risk criteria for preeclampsia (major and moderate), mean arterial pressure (MAP), uterine artery pulsatility index (IP AU), biomarkers (PIGF and PAPPA) were taken as early predictors. Results: Of the 632 participants who met the inclusion criteria, 62 (9.8%) presented preeclampsia and 570 (90.1%) participants did not develop it. They were grouped by gestational age at delivery and the different markers were applied, obtaining them by combining them in the case of Pregnancies less than 34 weeks a prediction of 88.24% (Low PE p ≤ 0.001 compared to High PE), in pregnancies less than 37 weeks 70% (Low PE p ≤ 0.04 compared to High PE), in pregnancies greater than 37 weeks 68% (Low PE p ≤ 0.003 compared to High PE). Conclusions: The application of the different markers individually is not usually highly effective, compared to their combined application, which in our studies achieves a higher percentage of prediction, especially in early-onset preeclampsia.

A Review on: Rice Disease Prediction

Abstract: Rice disease prediction is an emerging technology that leverages machine learning and image processing techniques to forecast disease outbreaks in rice crops. Early and accurate prediction of diseases like rice blast, bacterial leaf blight, and sheath blight is critical for maintaining crop health, optimizing resource use, and enhancing yield quality. This paper explores the use of high-resolution imaging and data-driven models to identify early signs of disease in rice plants. By analyzing data from multiple sources, including aerial imagery and environmental sensors, these predictive systems enable targeted and timely interventions, reducing the spread of infections and minimizing the use of pesticides. The application of predictive technologies not only aids in sustainable farming practices by lowering environmental impact but also aligns with the goals of precision agriculture to support food security and increase productivity. As agricultural practices continue to evolve, predictive models for rice disease management offer a promising solution for addressing the challenges posed by crop diseases in a changing climate

Yield prediction through UAV-based multispectral imaging and deep learning in rice breeding trials

ContextPredicting crop yields with high precision and timeliness is essential for crop breeding, enabling the optimization of planting strategies and efficients resource allocation while ensuring food security. Current research in this field typically does not address the problem of yield prediction in the diverse context of breeding experiments involving numerous varieties. However, evaluating the performance of prediction models across multiple varieties is vital for further model refining and enhancing model robustness and adaptability. ObjectiveThis study aims to evaluate the performance of feature- and image-based yield prediction models for yields with multiple varieties to compare their capabilities and determine an appropriate timing for early yield prediction. MethodsThis study combines unmanned aerial vehicle (UAV)-based multispectral remote sensing imagery with machine learning and deep learning-based algorithms to develop rice yield prediction models across multiple varieties. The performances of both feature- and image-based models are evaluated. The feature-based models considered in this study include random forest (RF), deep neural network (DNN), and long short-term memory (LSTM) algorithms, and the image-based models are convolutional neural network (CNN) architectures, including both two-dimensional (2D) and three-dimensional (3D) CNN models. To assess the performance of the multi-variety crop yield prediction models thoroughly, this study considers two sampling scenarios: stratified sampling and group sampling. Results and conclusionsThe results show that the image-based deep learning models outperform the feature-based machine learning models, which indicates their superior robustness in multi-variety scenarios and highlights their significant potential of directly extracting spatiotemporal features from images for yield prediction. The results indicate that the multi-temporal 2D CNN model (i.e., the CNN-M2D model) can achieve the best yield prediction performance among all models, achieving RRMSE = 8.13 % and R2 = 0.73. The prediction results also demonstrate good consistency with the observed data, indicating an efficient capturing of spatial pattern variations in yield across different varieties. Based on the results, with the crops progressing along the growth stages, the accuracy of the yield prediction models improves gradually, achieving the best prediction performance during the flowering to grain-filling stage. Finally, according to the results, the optimal lead time for predicting rice yield is approximately one month before harvest. SignificanceOur study can provide a reference for the research community in yield prediction and high-yield variety selection in breeding trials.

Development of a Regression Model for Prediction of Chronic Kidney Disease Risk

In recent years, chronic kidney disease (CKD) has been widespread in public health. Therefore, the early prediction of these diseases can save many lives. Keeping this fact in mind, this study presents a new way to predict CKD using regression modeling, aiming to improve early detection and save lives. For this purpose, the first authors collected the data of 104 patients, then re-arranged them in ten different parameters and calculated their scores. Thereafter, a composite CATH score is calculated as an output variable. Then, a suitable regression model will be identified based on various parameters such as R-squared, Adjusted R-squared, and PRESS values. Thereafter, to identify the significance of the selected model, the authors performed an Analysis of Variances (ANOVA) at a confidence interval of &lt;0.05. Results revealed that the developed model has a higher degree of fitness and is suitable for prediction purposes. Finally, the authors performed parameter analysis to identify the effects of various parameters on CKD.

High one-hour plasma glucose is an intermediate risk state and an early predictor of type 2 diabetes in a longitudinal Korean cohort

AimsBecause one-hour post-load plasma glucose (1 h-PG) ≥ 155 mg/dL (8.6 mmol/L) has been proposed as an early marker for future diabetes but lacks sufficient longitudinal confirmation of its risk, we aimed to evaluate the risk of T2D based on 1 h-PG and track changes of insulin sensitivity and β-cell function over time by 1 h-PG in a longitudinal cohort. MethodsOGTTs were conducted every 2 years in the 10-year longitudinal Korean Genome Epidemiology study (n = 6144) with three groups characterized at baseline: Low 1 h-PG (< 155 mg/dL) with Normal Glucose Tolerance (NGT), High 1 h-PG (≥155 mg/dL) with NGT, and prediabetes (PreDM). ResultsT2D risk was higher in people with High 1 h-PG with NGT and PreDM than those with Low 1 h-PG with NGT. Baseline insulin sensitivity in Low 1 h-PG as measured by the insulin sensitivity and secretion (ISS) model and Matsuda insulin sensitivity index (ISI) was higher than in High 1 h-PG, which was comparable to PreDM. β-cell function as assessed by ISS and the insulinogenic index decreased from Low 1 h-PG to High 1 h-PG to PreDM. Over time, insulin sensitivity decreased in the three groups. Time from High 1 h-PG to T2D was 0.9 years shorter than from Low 1 h-PG. All participants passed the 1 h-PG threshold for T2D (209 mg/dL, 11.6 mmol/L) first, and 74 % passed the 1 h-PG threshold for impaired glucose tolerance (IGT; 155 mg/dL) first. ConclusionsHigh 1 h-PG NGT is an intermediate risk category between Low 1 h-PG NGT and PreDM and may provide an opportunity for early intervention to prese rve ß-cell function.