Probabilistic Prediction Research Articles

Transcriptome profiling of pediatric extracranial solid tumors and lymphomas enables rapid low-cost diagnostic classification

Approximately 80% of pediatric tumors occur in low- and middle-income countries (LMIC), where diagnostic tools essential for treatment decisions are often unavailable or incomplete. Development of cost-effective molecular diagnostics will help bridge the cancer diagnostic gap and ultimately improve pediatric cancer outcomes in LMIC settings. We investigated the feasibility of using nanopore whole transcriptome sequencing on formalin-fixed paraffin embedded (FFPE)-derived RNA and a composite machine learning model for pediatric solid tumor diagnosis. Transcriptome cDNA sequencing was performed on a heterogenous set of 221 FFPE and 32 fresh frozen pediatric solid tumor and lymphoma specimens on Oxford Nanopore Technologies’ sequencing platforms. A composite machine learning model was then used to classify transcriptional profiles into clinically actionable tumor types and subtypes. In total, 95.6% and 89.7% of pediatric solid tumors and lymphoma specimens were correctly classified, respectively. 71.5% of pediatric solid tumors had prediction probabilities > 0.8 and were classified with 100% accuracy. Similarly, for lymphomas, 72.4% of samples that had prediction probabilities > 0.6 were classified with 97.6% accuracy. Additionally, FOXO1 fusion status was predicted accurately for 97.4% of rhabdomyosarcomas and MYCN amplification was predicted with 88% accuracy in neuroblastoma. Whole transcriptome sequencing from FFPE-derived pediatric solid tumor and lymphoma samples has the potential to provide clinical classification of both tissue lineage and core genomic classification. Further expansion, refinement, and validation of this approach is necessary to explore whether this technology could be part of the solution of addressing the diagnostic limitations in LMIC.

Machine learning-based reproducible prediction of type 2 diabetes subtypes

Aims/hypothesisClustering-based subclassification of type 2 diabetes, which reflects pathophysiology and genetic predisposition, is a promising approach for providing personalised and effective therapeutic strategies. Ahlqvist’s classification is currently the most vigorously validated method because of its superior ability to predict diabetes complications but it does not have strong consistency over time and requires HOMA2 indices, which are not routinely available in clinical practice and standard cohort studies. We developed a machine learning (ML) model to classify individuals with type 2 diabetes into Ahlqvist’s subtypes consistently over time.MethodsCohort 1 dataset comprised 619 Japanese individuals with type 2 diabetes who were divided into training and test sets for ML models in a 7:3 ratio. Cohort 2 dataset, comprising 597 individuals with type 2 diabetes, was used for external validation. Participants were pre-labelled (T2Dkmeans) by unsupervised k-means clustering based on Ahlqvist’s variables (age at diagnosis, BMI, HbA1c, HOMA2-B and HOMA2-IR) to four subtypes: severe insulin-deficient diabetes (SIDD), severe insulin-resistant diabetes (SIRD), mild obesity-related diabetes (MOD) and mild age-related diabetes (MARD). We adopted 15 variables for a multiclass classification random forest (RF) algorithm to predict type 2 diabetes subtypes (T2DRF15). The proximity matrix computed by RF was visualised using a uniform manifold approximation and projection. Finally, we used a putative subset with missing insulin-related variables to test the predictive performance of the validation cohort, consistency of subtypes over time and prediction ability of diabetes complications.ResultsT2DRF15 demonstrated a 94% accuracy for predicting T2Dkmeans type 2 diabetes subtypes (AUCs ≥0.99 and F1 score [an indicator calculated by harmonic mean from precision and recall] ≥0.9) and retained the predictive performance in the external validation cohort (86.3%). T2DRF15 showed an accuracy of 82.9% for detecting T2Dkmeans, also in a putative subset with missing insulin-related variables, when used with an imputation algorithm. In Kaplan–Meier analysis, the diabetes clusters of T2DRF15 demonstrated distinct accumulation risks of diabetic retinopathy in SIDD and that of chronic kidney disease in SIRD during a median observation period of 11.6 (4.5–18.3) years, similarly to the subtypes using T2Dkmeans. The predictive accuracy was improved after excluding individuals with low predictive probability, who were categorised as an ‘undecidable’ cluster. T2DRF15, after excluding undecidable individuals, showed higher consistency (100% for SIDD, 68.6% for SIRD, 94.4% for MOD and 97.9% for MARD) than T2Dkmeans.Conclusions/interpretationThe new ML model for predicting Ahlqvist’s subtypes of type 2 diabetes has great potential for application in clinical practice and cohort studies because it can classify individuals with missing HOMA2 indices and predict glycaemic control, diabetic complications and treatment outcomes with long-term consistency by using readily available variables. Future studies are needed to assess whether our approach is applicable to research and/or clinical practice in multiethnic populations.Graphical

Intrinsic Rewards for Exploration Without Harm From Observational Noise: A Simulation Study Based on the Free Energy Principle.

In reinforcement learning (RL), artificial agents are trained to maximize numerical rewards by performing tasks. Exploration is essential in RL because agents must discover information before exploiting it. Two rewards encouraging efficient exploration are the entropy of action policy and curiosity for information gain. Entropy is well established in the literature, promoting randomized action selection. Curiosity is defined in a broad variety of ways in literature, promoting discovery of novel experiences. One example, prediction error curiosity, rewards agents for discovering observations they cannot accurately predict. However, such agents may be distracted by unpredictable observational noises known as curiosity traps. Based on the free energy principle (FEP), this letter proposes hidden state curiosity, which rewards agents by the KL divergence between the predictive prior and posterior probabilities of latent variables. We trained six types of agents to navigate mazes: baseline agents without rewards for entropy or curiosity and agents rewarded for entropy and/or either prediction error curiosity or hidden state curiosity. We find that entropy and curiosity result in efficient exploration, especially both employed together. Notably, agents with hidden state curiosity demonstrate resilience against curiosity traps, which hinder agents with prediction error curiosity. This suggests implementing the FEP that may enhance the robustness and generalization of RL models, potentially aligning the learning processes of artificial and biological agents.

A conceptual risk modelling for cargo tank fire/explosion in chemical tanker by using Evidential Reasoning -SLIM and Bayesian belief network approach

The transportation of hazardous cargoes, particularly in chemical tankers, represents significant risks such as fire and explosion, which can have catastrophic consequences for both human life, commodity and maritime environment. This article presents a conceptual risk modelling for cargo tank fire/explosion incidents in chemical tanker ships under the Evidential Reasoning (ER) - Success likelihood index method (SLIM) approach with the Bayesian belief network (BBN) to provide a practical tool for assessing and managing the associated risks. Whilst the ER-SLIM approach offers a systematic human error probability (HEP) prediction and handling uncertainty in risk, allowing for the incorporation of expert judgments and data-driven information, the BBN provides a probabilistic graphical model to represent the causal relationships among various risk factors. The result shows that the risk of cargo tank fire/explosion during the cargo tank cleaning operation is calculated as 2.83E-02. The findings of the research including consequences analysis provide valuable insights for decision-makers, safety managers, superintendents, ship masters and officers in the maritime industry to prioritize risk management strategies and enhance the safety of chemical tanker operations.

High-percentage new energy distribution network line loss frequency division prediction based on wavelet transform and BIGRU-LSTM.

The access of new energy improves the flexibility of distribution network operation, but also leads to more complex mechanism of line loss. Therefore, starting from the nonlinear, fluctuating and multi-scale characteristics of line loss data, and based on the idea of decomposition prediction, this paper proposes a new method of line loss frequency division prediction based on wavelet transform and BIGRU-LSTM (Bidirectional Gated Recurrent Unit-Long Short Term Memory Network).Firstly, the grey relation analysis and the improved NARMA (Nonlinear Autoregressive Moving Average) correlation analysis method are used to extract the non-temporal and temporal influencing factors of line loss, and the corresponding feature data set is constructed. Then, the historical line loss data is decomposed into physical signals of different frequency bands by using wavelet transform, and the multi-dimensional input data of the prediction network is formed with the above characteristic data set. Finally, the BIGRU-LSTM prediction network is built to realize the probabilistic prediction of high-frequency and low-frequency components of line loss. The effectiveness and applicability of the method proposed in this paper were verified through numerical simulation. By dividing the line loss data into different frequency bands for frequency prediction, the mapping relationship between different line loss components and influencing factors was accurately matched, thereby improving the prediction accuracy.

Combining reproductive endocrinology and ROC analysis to identify changes with sex, age, and pregnancy status in botos Inia geoffrensis

Reproductive hormone profiles were described against physical characteristics during growth and development in male and female botos (Amazon River dolphins Inia geoffrensis) and during pregnancy. We determined hormone concentrations in 226 wild botos located in Mamirauá Reserve, Amazonas State, Brazil, as a part of the long-term population monitoring program known as Projeto Boto. Additionally, we applied receiver operator characteristic (ROC) analysis to compare diagnostic probabilities of using ultrasound, hormones, or combinations of these to detect pregnancy. Based on single-point analysis of serum testosterone (T), males with <2.5 ng ml-1 T and a mean 163 cm total body length were classified as immature, 2.5 to <4.9 ng ml-1 T and 183 cm as pubescent, and >5 ng ml-1 T and 227 cm as adult botos. For females, only progesterone (P4), T, relaxin (Rlx), and the combination of P4 × T2 were significantly different between non-pregnant and pregnant females, but androstenedione (A4) and the P4:T ratio were not. ROC analysis indicated that ultrasound and P4 × T2 were considered excellent as pregnancy diagnostic tests, and P4, T, and Rlx were classified as good predictors. Results indicated that negative and positive predictive probabilities from each diagnostic test could be used to accurately predict a pregnancy and calf loss rate of 13% for this population. Application of these methods for evaluating wild population reproductive success from a single serum sample can now be used for health evaluations of wild populations of boto and provide timely information for the development or evaluation of any conservation initiatives.

Generalized collaborative relevance vector regression for soft sensors

Soft sensors often struggle to adapt to changing conditions due to their overreliance on single data sources, severely undermining the generalizability and applicability of models. This paper proposes a novel generalized collaborative relevance vector regression (GCRVR) framework to substantially improve the accuracy and adaptability of soft sensors under varying conditions. The core innovation of GCRVR lies in an adaptive collaborative strategy designed to combine complementary predictive distributions based on analyzed variance differences, enabling the acquisition of more reasonable and reliable results. The GCRVR approach leverages both smoothed and unsmoothed process data to train independent relevance vector regression models, generating probabilistic predictions. A theoretical analysis provides insight into ensuring the stability of the model. In an evaluation conducted across four sophisticated industrial processes, the proposed GCRVR method outperformed a suite of conventional methods—namely, linear regression, support vector regression, Gaussian process regression, a stacked autoencoder, long short-term memory, attention-based long short-term memory, and extreme gradient boosting. The GCRVR method exhibited substantial predictive accuracy enhancements, as evidenced by average reductions of 39% in the root mean square error (RMSE), 43% in the mean absolute error (MAE), and 42% in the mean absolute percentage error (MAPE). This work addresses the generalization and adaptability limitations of soft sensors through an integrated framework that capitalizes on the collaborative use of complementary models. GCRVR constitutes a major advancement in real-time monitoring and optimization for industrial process control.

Bayesian Predictive Probability Based on a Bivariate Index Vector for Single-Arm Phase II Study With Binary Efficacy and Safety Endpoints.

In oncology, Phase II studies are crucial for clinical development plans as such studies identify potent agents with sufficient activity to continue development in the subsequent Phase III trials. Traditionally, Phase II studies are single-arm studies, with the primary endpoint being short-term treatment efficacy. However, drug safety is also an important consideration. In the context of such multiple-outcome designs, predictive probability-based Bayesian monitoring strategies have been developed to assess whether a clinical trial will provide enough evidence to continue with a Phase III study at the scheduled end of the trial. Therefore, we propose a new simple index vector to summarize the results that cannot be captured by existing strategies. Specifically, we define the worst and most promising situations for the potential effect of a treatment, then use the proposed index vector to measure the deviation between the two situations. Finally, simulation studies are performed to evaluate the operating characteristics of the design. The obtained results demonstrate that the proposed method makes appropriate interim go/no-go decisions.

Time-domain fatigue reliability analysis for floating offshore wind turbine substructures using coupled nonlinear aero-hydro-servo-elastic simulations

The present study aims to develop a novel methodology for the fatigue reliability analysis and design for a floating offshore wind turbine substructure using its fully coupled nonlinear dynamic responses in the time domain. The developed methodology offers a computationally efficient yet robust assessment for designing fatigue-critical welded joints on floating substructures. The methodology starts with the sea state selection based on the fatigue damage contribution of all the sea states in the scatter diagram. To this end, the dynamic response is analysed by fully coupled aero-hydro-servo-elastic simulations using OpenFAST. The resulting short-term fatigue loading is then used to estimate the hotspot stress history using the analytical solution for global structural analysis and the empirical solution for the stress concentration factor. The long-term fatigue damage is calculated as the joint probability-weighted sum of the short-term fatigue damage using Palmgren-Miner's linear damage rule. Afterwards, the selected sea states are used to perform dynamic response simulations with multiple random seeds, ensuring no significant accuracy loss. To obtain the probabilistic fatigue life prediction, a novel time-domain fatigue reliability analysis algorithm is developed based on the bootstrapping method, which creates a pool of short-term fatigue responses with different random seeds and combines them within a Monte Carlo Simulation. Finally, the fatigue reliability analysis considering the S-N and damage-tolerant approaches for design is carried out.

Machine learning-based noninvasive diagnostic classifiers for the prediction of cancer tissue of origin using serum microRNAs.

101 Background: Noninvasive multi-cancer early detection (MCED) with or without tissue of origin (TOO) has the potential to reduce cancer-related mortality by analyzing circulating cell-free nucleic acids and/or proteins in blood. Accurate prediction of TOO following a positive MCED test would guide selection of confirmatory tests, thereby expediting the definitive diagnosis and prompt initiation of the most appropriate treatment, tailored to the specific cancer type. Here, we report the development of machine learning-based diagnostic classifiers that predict TOO for 13 cancer types with high accuracy using serum microRNAs. Methods: Eight serum miRNA microarray datasets from GEO totaling 6,283 patients across 13 cancer types were used in this study. The patients were split, with an approximate 3:2 ratio, into a training (n=3,844) and a validation set (n=2,439). An ensemble of classifiers was constructed in the training set via the “one vs. rest” approach, thus one classifier for each cancer type. Random forest models with recursive feature elimination (RFE) selected the optimal set of miRNAs that was fed into support vector machine models to generate a prediction probability for each cancer type. The type with the highest probability was considered the predicted cancer type. The performance of these classifiers was evaluated in the validation set in two steps with the 1st using all cancer types and the 2nd using the top 2 or 3 cancer types from the 1st step to achieve a refined prediction. Results: RFE selected 426 miRNAs for building the 13 classification models. In the validation set comprising 2,439 patients across 12 cancer types, the classifiers correctly predicted cancer types for 1,922 (79%) samples based on the highest prediction probability. The accuracy increased to 92% and 95% based on top 2 and 3 predictions. In particular, based on top 3 predictions, the accuracy was >95% for bladder, breast, prostate, gastric, glioma and lung cancers, >85% for ovarian, liver and esophageal cancers, 78% for pancreatic cancer and sarcoma, and 67% for colorectal cancer. Conclusions: With 95% accuracy in narrowing TOO down to 3 organ sites, the miRNA-based TOO classifiers could be used clinically as a reflex test for the simple and highly accurate MCED screening models previously developed ( Cancers 2022,14:1450; ESMO 2024). Together, they support the development of an inexpensive, accurate and noninvasive blood test for MCED with TOO.

Stereotypes as Bayesian prediction of social groups

ABSTRACT A stereotype is a generalization about a class of people which is often used to make probabilistic predictions about individuals within that class. Can stereotypes can be understood as conditional probabilities that distinguish among groups in ways that follow Bayesian posterior prediction? For instance, the stereotype of Germans as industrious can be understood as the conditional probability of someone being industrious given that they are German. Whether such representations follow Bayes’ rule was tested in a replication and extension of past work. Across three studies (N = 2,652), we found that people’s judgments of different social categories were appropriately Bayesian, in that their direct posterior predictions were aligned with what Bayes’ rule suggests they should be. Moreover, across social categories, traits with a high calculated diagnostic ratio generally distinguished stereotypic from non-stereotypic traits. The effects of cognitive ability, political orientation, and motivated stereotyping were also explored.

Feasibility of Coseismic Landslide Prediction Based on GNSS Observations: A Case Study of the 2022 Ms 6.8 Luding, China, Earthquake

Abstract On 5 September 2022, an M 6.8 earthquake struck Luding County in Sichuan Province, China, triggering extensive landslides and causing severe damages. In this study, taking this event as an exemplary case study, we test the feasibility of fast earthquake-induced landslide prediction utilizing Global Navigation Satellite System (GNSS) observations. Particularly, we construct finite-slip models based on static offsets and 1 Hz displacement waveforms. Employing these slip models, physics-based simulation (PBS) is applied separately to obtain peak ground velocity (PGV). The PGVs are then integrated into landslide spatial distribution probability prediction based on the Deep Forest algorithm. Our results show that the predicted landslides probability distribution of fast inversion models using static and high-rate GNSS data align well with the landslide catalog. Furthermore, high-rate GNSS data can improve the model performance by providing the evolution information of rupture. In addition, we also derive PGV from the empirically regressed ground-motion prediction equations (GMPEs) and incorporate it into landslide prediction. The GMPEs exhibits an advantage in terms of prediction recall for landslides and a relatively reduced accuracy compared with the PBS. Despite the inherent uncertainties in this study, based on the case study of the 2022 M 6.8 Luding earthquake, we utilize GNSS data and present a set of methods for real-time landslide prediction. The achieved model performance is relatively satisfactory, considering the challenges and uncertainties involved.

Performance evaluation of multivariate deep-time convolution neural architectures for short-term electricity forecasting: Findings and failures

Deep learning (DL) models hold great promise in enhancing the decision-making abilities of electricity market participants and system operators in the short term, as they excel at capturing the intricate interdependencies and uncertainties in electricity consumption. Nevertheless, currently, the adoption of DL models for electricity prediction is not keeping up with the advancements in DL research. Within this framework, the aims of this study are to examine the probabilistic, multivariate, and multihorizon time series prediction of deep time convolution neural (DTCN) models. The effectiveness of DTCN in forecasting electricity consumption over daily, weekly and monthly time horizons is emperically assessed. We analyze both probabilistic and deterministic predictions. The evaluation data comprises real residential electricity consumption, average temperature and humidity with hourly granularity, collected over one year from a residential building. For evaluation, we considered accuracy and uncertainties, sensitivity to hyperparameters, the impact of exogeneous variables, as well as computational efficiency. Results show that the probabilistic models offer greater advantages when it comes to forecasting on multihorizons, especially for diverse datasets with non-uniform time series models. Unlike traditional models, the DTCN models demonstrated superior performance in capturing complex patterns and reducing forecast errors in short-term scenarios. However, their performance can be influenced by the sensitivity of hyperparameters. These findings can act as a benchmark for quantitatively assessing the performance of novel multivariate DL models in predicting electricity demand.

What Is the Relationship between Chronotype and Disordered Eating in Adolescents? The EHDLA Study.

Evidence assessing the relationship between chronotype and disordered eating in adolescents is scarce. The current study tried to evaluate the association between chronotype and disordered eating in a sample of Spanish adolescents. This secondary cross-sectional study analyzed data from the Eating Healthy and Daily Life Activities (EHDLA) study. The sample consisted of 703 adolescents (56.3% girls) aged between 12 and 17 years from the Valle de Ricote (Region of Murcia, Spain). Chronotype was assessed using the Morningness/Eveningness Scale in Children (MESC). Disordered eating was evaluated by two psychologists using the Sick, Control, One, Fat, and Food (SCOFF) questionnaire. Adolescents with an eveningness chronotype showed a higher SCOFF score (estimated marginal mean [M] = 1.1; 95% confidence interval [CI] 0.7 to 1.5) in comparison with adolescents with a morningness chronotype (M = 0.7; 95% CI 0.5 to 0.8) (p = 0.010), as well as with those with an intermediate chronotype (M = 0.6; 95% CI 0.5 to 0.8) (p = 0.032). A higher predictive probability of having disordered eating was identified in adolescents with an eveningness chronotype (39.5%; 95% CI 22.8% to 59.1%), compared to adolescents with an intermediate chronotype (14.9%; 95% CI 10.8% to 20.1%) (p = 0.008) and with their counterparts with a morningness chronotype (16.9%; 95% CI 11.6% to 24.0%) (p = 0.021). This study reveals that adolescents with an eveningness chronotype are more likely to exhibit disordered eating behaviors compared to those with morningness or intermediate chronotypes. These findings highlight the importance of considering chronotype in adolescent health, particularly in developing targeted interventions to prevent eating disorders.

Cross-domain fault diagnosis network based on attributes and features transfer with dual classifier under limited and unbalanced datasets

Abstract Deep learning (DL)-based methods have demonstrated significant success in fault diagnosis owing to their robust feature extraction and non-linear fitting capabilities. Meanwhile, their remarkable performance is accompanied by constant operating conditions and sufficient monitoring data. However, in real engineering environments, variable working conditions or limited and unbalanced data are common, which can widen the gap between fault diagnosis methods and real industrial applications. In this paper, we proposed a cross-domain fault diagnosis network based on a dual classifier (CFDNet) with input being limited and unbalanced data to learn attributes and features for unsupervised domain adaptation. We found that the diagnostic performance is commonly bounded by the underlying knowledge, especially feature extraction from original data. Therefore, we designed a new feature encoder with features and relationships, i.e. using a convolutional neural network and graph convolutional network, which improves extraction efficiency while retaining valuable information. Then, we discovered that enforced feature transfer can lead to negative transfer. To mitigate this, we present a feature and attribute transfer framework, which not only achieves features transfer but also enables attributes transfer. Furthermore, it was noted that limited and unbalanced datasets can introduce label bias and lead to biased model training. Hence, we designed dual classifiers to improve the probability of high-confidence final prediction by synthesizing diagnostic results. Comprehensive experiments conducted on three case studies demonstrate the effectiveness and superiority of our method for cross-domain fault diagnosis under limited and unbalanced datasets, which outperforms state-of-the-art methods in this study.

Modeling of the cascading impacts of drought and forest fire based on a Bayesian network

The cascading impact of disastrous events has become an important focus of disaster research. In recent years, forest fires have occurred frequently in China, causing huge economic losses. Studying the cascading impacts of drought and forest fire is of great significance for reducing disaster risks. Taking Yunnan Province of China as an example, meteorological data and forest fire point data from 2005 to 2018 were collected and statistically analyzed. A Bayesian network model of the cascading impacts of drought and forest fire was established, enabling the prior probability and conditional probability of nodes to be determined. Based on this information, a probability prediction was established using causal reasoning. Finally, in the case test, the Brier score was used to test the accuracy of the model. The Brier test value was 0.305, which was less than the qualified threshold of 0.6. The results indicated that the Bayesian network model established in this study had a good prediction performance, which was basically consistent with the facts. The results provide an insight into the mechanism by which drought induced forest fires occur and will be of use in forest fire prevention work.

Small data reliability analysis in concrete three-point bending tests: A Weibull mixture model approach based on Weibull fracture theory

Due to the high costs and time-consuming nature of experiments, data from concrete three-point bending tests are usually limited. Additionally, the performance of concrete is influenced by various factors such as mix composition and sample preparation methods, leading to significant variability in test results. This poses a major challenge for the reliability analysis of small data from repeated tests. In reliability analysis, statistical models are particularly important as they allow for the prediction of performance and failure probabilities based on existing data. However, traditional statistical methods require a large number of similar samples for data processing and may need specific adjustments due to the diversity of practical applications. To address these issues, this study proposes a mixture model that combines domain knowledge of concrete with Weibull fracture theory and uses the Clustering-Circular Block Bootstrap method for sample augmentation. By analyzing real data obtained from concrete three-point bending tests, we accurately estimated the failure probability of concrete. Additionally, the study uses Gaussian process regression with a spectral mixture kernel function to fit the threshold fracture strength of concrete and estimate failure probabilities. The results show that the proposed method can accurately fit small datasets, with a RMSE of 0.320 and a MAPE of 0.100. The nominal bending strength range at a 5% failure probability is 2.0 MPa to 6.5 MPa, depending on the sample height and pre-crack length. Statistical learning demonstrates that the proposed method excels in fitting small datasets, providing a solid framework for predicting concrete strength and reliability.

Cellular proliferation biases clonal lineage tracing and trajectory inference.

Lineage tracing and trajectory inference from single-cell RNA-sequencing data hold tremendous potential for uncovering the genetic programs driving development and disease. Single cell datasets are thought to provide an unbiased view on the diverse cellular architecture of tissues. Sampling bias, however, can skew single cell datasets away from the cellular composition they are meant to represent. We demonstrate a novel form of sampling bias, caused by a statistical phenomenon related to repeated sampling from a growing, heterogeneous population. Relative growth rates of cells influence the probability that they will be sampled in clones observed across multiple time points. We support our probabilistic derivations with a simulation study and an analysis of a real time-course of T-cell development. We find that this bias can impact fate probability predictions, and we explore how to develop trajectory inference methods which are robust to this bias. Source code for the simulated datasets and to create the figures in this manuscript is freely available in python at https://github.com/rbonhamcarter/simulate-clones. A python implementation of the extension of the LineageOT method is freely available at https://github.com/rbonhamcarter/LineageOT/tree/multi-time-clones.

Differentiating low-risk thymomas from high-risk thymomas: preoperative radiomics nomogram based on contrast enhanced CT to minimize unnecessary invasive thoracotomy

BackgroundThis study was designed to develop a combined radiomics nomogram to preoperatively predict the risk categorization of thymomas based on contrast-enhanced computed tomography (CE-CT) images.MaterialsThe clinical and CT data of 178 patients with thymoma (100 patients with low-risk thymomas and 78 patients with high-risk thymomas) collected in our hospital from March 2018 to July 2023 were retrospectively analyzed. The patients were randomly divided into a training set (n = 125) and a validation set (n = 53) in a 7:3 ratio. Qualitative radiological features were recorded, including (a) tumor diameter, (b) location, (c) shape, (d) capsule integrity, (e) calcification, (f) necrosis, (g) fatty infiltration, (h) lymphadenopathy, and (i) enhanced CT value. Radiomics features were extracted from each CE-CT volume of interest (VOI), and the least absolute shrinkage and selection operator (LASSO) algorithm was performed to select the optimal discriminative ones. A combined radiomics nomogram was further established based on the clinical factors and radiomics scores. The differentiating efficacy was determined using receiver operating characteristic (ROC) analysis.ResultsOnly one clinical factor (incomplete capsule) and seven radiomics features were found to be independent predictors and were used to establish the radiomics nomogram. In differentiating low-risk thymomas (types A, AB, and B1) from high-risk ones (types B2 and B3), the nomogram demonstrated better diagnostic efficacy than any single model, with the respective area under the curve (AUC), accuracy, sensitivity, and specificity of 0.974, 0.921, 0.962 and 0.900 in the training cohort, 0.960, 0.892, 0923 and 0.897 in the validation cohort, respectively. The calibration curve showed good agreement between the prediction probability and actual clinical findings.ConclusionsThe nomogram incorporating clinical factors and radiomics features provides additional value in differentiating the risk categorization of thymomas, which could potentially be useful in clinical practice for planning personalized treatment strategies.

Dynamic Relapse Prediction by Peripheral Blood WT1mRNA after Allogeneic Hematopoietic Cell Transplantation for Myeloid Neoplasms

Although various relapse prediction models based on pretransplant information have been reported, they cannot update the predictive probability considering post-transplant patient status. Therefore, these models are not appropriate for deciding on treatment adjustment and preemptive intervention during post-transplant follow-up. A dynamic prediction model can update the predictive probability by considering the information obtained during follow-up. This study aimed to develop and assess a dynamic relapse prediction model after allogeneic hematopoietic cell transplantation (allo-HCT) for acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) using peripheral blood Wilms' tumor 1 messenger RNA (WT1mRNA). We retrospectively analyzed patients with AML or MDS who underwent allo-HCT at our institution. To develop dynamic models, we employed the landmarking supermodel approach, using age, refined disease risk index, conditioning intensity, and number of transplantations as pretransplant covariates and both pre- and post-transplant peripheral blood WT1mRNA levels as time-dependent covariates. Finally, we compared the predictive performances of the conventional and dynamic models by area under the time-dependent receiver operating characteristic curves. A total of 238 allo-HCT cases were included in this study. The dynamic model that considered all pretransplant WT1mRNA levels and their kinetics showed superior predictive performance compared to models that considered only pretransplant covariates or factored in both pretransplant covariates and post-transplant WT1mRNA levels without their kinetics; their time-dependent areas under the curve were 0.89, 0.73, and 0.87, respectively. The predictive probability of relapse increased gradually from approximately 90 days before relapse. Furthermore, we developed a web application to make our model user-friendly. This model facilitates real-time, highly accurate, and personalized relapse prediction at any time point after allo-HCT. This will aid decision-making during post-transplant follow-up by offering objective relapse forecasts for physicians.