Bayesian network analysis of factors influencing type 2 diabetes, coronary heart disease, and their comorbidities.

OBJECTIVE: Bayesian network (BN) models were developed to explore the specific relationships between influencing factors and type 2 diabetes mellitus (T2DM), coronary heart disease (CAD), and their comorbidities. The aim was to predict disease occurrence and diagnose etiology using these models, thereby informing the development of effective prevention and control strategies for T2DM, CAD, and their comorbidities. METHOD: Employing a case-control design, the study compared individuals with T2DM, CAD, and their comorbidities (case group) with healthy counterparts (control group). Univariate and multivariate Logistic regression analyses were conducted to identify disease-influencing factors. The BN structure was learned using the Tabu search algorithm, with parameter estimation achieved through maximum likelihood estimation. The predictive performance of the BN model was assessed using the confusion matrix, and Netica software was utilized for visual prediction and diagnosis. RESULT: The study involved 3,824 participants, including 1,175 controls, 1,163 T2DM cases, 982 CAD cases, and 504 comorbidity cases. The BN model unveiled factors directly and indirectly impacting T2DM, such as age, region, education level, and family history (FH). Variables like exercise, LDL-C, TC, fruit, and sweet food intake exhibited direct effects, while smoking, alcohol consumption, occupation, heart rate, HDL-C, meat, and staple food intake had indirect effects. Similarly, for CAD, factors with direct and indirect effects included age, smoking, SBP, exercise, meat, and fruit intake, while sleeping time and heart rate showed direct effects. Regarding T2DM and CAD comorbidities, age, FBG, SBP, fruit, and sweet intake demonstrated both direct and indirect effects, whereas exercise and HDL-C exhibited direct effects, and region, education level, DBP, and TC showed indirect effects. CONCLUSION: The BN model constructed using the Tabu search algorithm showcased robust predictive performance, reliability, and applicability in forecasting disease probabilities for T2DM, CAD, and their comorbidities. These findings offer valuable insights for enhancing prevention and control strategies and exploring the application of BN in predicting and diagnosing chronic diseases.

Measuring protective efficacy and quantifying the impact of drug resistance: A novel malaria chemoprevention trial design and methodology.

BACKGROUND: Recently revised WHO guidelines on malaria chemoprevention have opened the door to more tailored implementation. Countries face choices on whether to replace old drugs, target additional age groups, and adapt delivery schedules according to local drug resistance levels and malaria transmission patterns. Regular routine assessment of protective efficacy of chemoprevention is key. Here, we apply a novel modelling approach to aid the design and analysis of chemoprevention trials and generate measures of protection that can be applied across a range of transmission settings. METHODS AND FINDINGS: We developed a model of genotype-specific drug protection, which accounts for underlying risk of infection and circulating genotypes. Using a Bayesian framework, we fitted the model to multiple simulated scenarios to explore variations in study design, setting, and participant characteristics. We find that a placebo or control group with no drug protection is valuable but not always feasible. An alternative approach is a single-arm trial with an extended follow-up (>42 days), which allows measurement of the underlying infection risk after drug protection wanes, as long as transmission is relatively constant. We show that the currently recommended 28-day follow-up in a single-arm trial results in low precision of estimated 30-day chemoprevention efficacy and low power in determining genotype differences of 12 days in the duration of protection (power = 1.4%). Extending follow-up to 42 days increased precision and power (71.5%) in settings with constant transmission over this time period. However, in settings of unstable transmission, protective efficacy in a single-arm trial was overestimated by 24.3% if recruitment occurred during increasing transmission and underestimated by 15.8% when recruitment occurred during declining transmission. Protective efficacy was estimated with greater precision in high transmission settings, and power to detect differences by resistance genotype was lower in scenarios where the resistant genotype was either rare or too common. CONCLUSIONS: These findings have important implications for the current guidelines on chemoprevention efficacy studies and will be valuable for informing where these studies should be optimally placed. The results underscore the need for a comparator group in seasonal settings and provide evidence that the extension of follow-up in single-arm trials improves the accuracy of measures of protective efficacy in settings with more stable transmission. Extension of follow-up may pose logistical challenges to trial feasibility and associated costs. However, these studies may not need to be repeated multiple times, as the estimates of drug protection against different genotypes can be applied to different settings by adjusting for transmission intensity and frequency of resistance.

A fully autonomous robotic ultrasound system for thyroid scanning.

The current thyroid ultrasound relies heavily on the experience and skills of the sonographer and the expertise of the radiologist, and the process is physically and cognitively exhausting. In this paper, we report a fully autonomous robotic ultrasound system, which is able to scan thyroid regions without human assistance and identify malignant nod- ules. In this system, human skeleton point recognition, reinforcement learning, and force feedback are used to deal with the difficulties in locating thyroid targets. The orientation of the ultrasound probe is adjusted dynamically via Bayesian optimization. Experimental results on human participants demonstrated that this system can perform high-quality ultrasound scans, close to manual scans obtained by clinicians. Additionally, it has the potential to detect thyroid nodules and provide data on nodule characteristics for American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) calculation.

Reverse engineering morphogenesis through Bayesian optimization of physics-based models.

Morphogenetic programs coordinate cell signaling and mechanical interactions to shape organs. In systems and synthetic biology, a key challenge is determining optimal cellular interactions for predicting organ shape, size, and function. Physics-based models defining the subcellular force distribution facilitate this, but it is challenging to calibrate parameters in these models from data. To solve this inverse problem, we created a Bayesian optimization framework to determine the optimal cellular force distribution such that the predicted organ shapes match the experimentally observed organ shapes. This integrative framework employs Gaussian Process Regression, a non-parametric kernel-based probabilistic machine learning modeling paradigm, to learn the mapping functions relating to the morphogenetic programs that maintain the final organ shape. We calibrated and tested the method on Drosophila wing imaginal discs to study mechanisms that regulate epithelial processes ranging from development to cancer. The parameter estimation framework successfully infers the underlying changes in core parameters needed to match simulation data with imaging data of wing discs perturbed with collagenase. The computational pipeline identifies distinct parameter sets mimicking wild-type shapes. It enables a global sensitivity analysis to support the regulation of actomyosin contractility and basal ECM stiffness to generate and maintain the curved shape of the wing imaginal disc. The optimization framework, combined with experimental imaging, identified that Piezo, a mechanosensitive ion channel, impacts fold formation by regulating the apical-basal balance of actomyosin contractility and elasticity of ECM. This workflow is extensible toward reverse-engineering morphogenesis across organ systems and for real-time control of complex multicellular systems.

Spatial analysis of cholangiocarcinoma in relation to diabetes mellitus and Opisthorchis viverrini infection in Northeast Thailand.

Cholangiocarcinoma (CCA) exhibits a heightened incidence in regions with a high prevalence of Opisthorchis viverrini infection, with previous studies suggesting an association with diabetes mellitus (DM). Our study aimed to investigate the spatial distribution of CCA in relation to O. viverrini infection and DM within high-risk populations in Northeast Thailand. Participants from 20 provinces underwent CCA screening through the Cholangiocarcinoma Screening and Care Program between 2013 and 2019. Health questionnaires collected data on O. viverrini infection and DM, while ultrasonography confirmed CCA diagnoses through histopathology. Multiple zero-inflated Poisson regression, accounting for covariates like age and gender, assessed associations of O. viverrini infection and DM with CCA. Bayesian spatial analysis methods explored spatial relationships. Among 263,588 participants, O. viverrini infection, DM, and CCA prevalence were 32.37%, 8.22%, and 0.36%, respectively. The raw standardized morbidity ratios for CCA was notably elevated in the Northeast's lower and upper regions. Coexistence of O. viverrini infection and DM correlated with CCA, particularly in males and those aged over 60 years, with a distribution along the Chi, Mun, and Songkhram Rivers. Our findings emphasize the association of the spatial distribution of O. viverrini infection and DM with high-risk CCA areas in Northeast Thailand. Thus, prioritizing CCA screening in regions with elevated O. viverrini infection and DM prevalence is recommended.

Adverse events associated with immune checkpoint inhibitors in non-small cell lung cancer: a safety analysis of clinical trials and FDA pharmacovigilance system.

Objectives: Immune checkpoint inhibitors (ICIs) have revolutionized the treatment of non-small cell lung cancer (NSCLC). However, the application of ICIs can also cause treatment-related adverse events (trAEs) and immune-related adverse events (irAEs). This study was to evaluate both the irAEs and trAEs of different ICI strategies for NSCLC based on randomized clinical trials (RCTs). The study also examined real-world pharmacovigilance data from the Food and Drug Administration Adverse Event Reporting System (FAERS) regarding claimed ICI-associated AEs in clinical practice. Methods: Based on Pubmed, Embase, Medline, and the Cochrane CENTRAL, we retrieved RCTs comparing ICIs with chemotherapy drugs or with different ICI regimens for the treatment of NSCLC up to October 20, 2023. Bayesian network meta-analysis (NMA) was performed using odds ratios (ORs) with 95% credible intervals (95%CrI). Separately, a retrospective pharmacovigilance study was performed based on FAERS database, extracting ICI-associated AEs in NSCLC patients between the first quarter (Q1) of 2004 and Q4 of 2023. The proportional reports reporting odds ratio was calculated to analyze the disproportionality. Results: The NMA included 51 RCTs that involved a total of 26,958 patients with NSCLC. Based on the lowest risk of any trAEs, cemiplimab, tislelizumab, and durvalumab were ranked as the best. Among the agents associated with the lowest risk of grades 3-5 trAEs, tislelizumab, avelumab, and nivolumab were most likely to rank highest. As far as any or grades 3-5 irAEs are concerned, atezolizumab plus bevacizumab plus chemotherapy is considered the most safety option. However, it is associated with a high risk of grades 3-5 trAEs. As a result of FAERS pharmacovigilance data analysis, 9,420 AEs cases have been identified in 7,339 NSCLC patients treated with ICIs, and ICIs were related to statistically significant positive signal with 311 preferred terms (PTs), and comprehensively investigated and identified those AEs highly associated with ICIs. In total, 152 significant signals were associated with Nivolumab, with malignant neoplasm progression, death, and hypothyroidism being the most frequent PTs. Conclusion: These findings revealed that ICIs differed in their safety profile. ICI treatment strategies can be improved and preventive methods can be developed for NSCLC patients based on our results.

Replication of null results: Absence of evidence or evidence of absence?

In several large-scale replication projects, statistically non-significant results in both the original and the replication study have been interpreted as a 'replication success.' Here, we discuss the logical problems with this approach: Non-significance in both studies does not ensure that the studies provide evidence for the absence of an effect and 'replication success' can virtually always be achieved if the sample sizes are small enough. In addition, the relevant error rates are not controlled. We show how methods, such as equivalence testing and Bayes factors, can be used to adequately quantify the evidence for the absence of an effect and how they can be applied in the replication setting. Using data from the Reproducibility Project: Cancer Biology, the Experimental Philosophy Replicability Project, and the Reproducibility Project: Psychology we illustrate that many original and replication studies with 'null results' are in fact inconclusive. We conclude that it is important to also replicate studies with statistically non-significant results, but that they should be designed, analyzed, and interpreted appropriately.

Comparison of quantitative whole body PET parameters on [68Ga]Ga-PSMA-11 PET/CT using ordered Subset Expectation Maximization (OSEM) vs. bayesian penalized likelihood (BPL) reconstruction algorithms in men with metastatic castration-resistant prostate cancer.

BACKGROUND: PSMA PET/CT is a predictive and prognostic biomarker for determining response to [177Lu]Lu-PSMA-617 in patients with metastatic castration resistant prostate cancer (mCRPC). Thresholds defined to date may not be generalizable to newer image reconstruction algorithms. Bayesian penalized likelihood (BPL) reconstruction algorithm is a novel reconstruction algorithm that may improve contrast whilst preventing introduction of image noise. The aim of this study is to compare the quantitative parameters obtained using BPL and the Ordered Subset Expectation Maximization (OSEM) reconstruction algorithms. METHODS: Fifty consecutive patients with mCRPC who underwent [68Ga]Ga-PSMA-11 PET/CT using OSEM reconstruction to assess suitability for [177Lu]Lu-PSMA-617 therapy were selected. BPL algorithm was then used retrospectively to reconstruct the same PET raw data. Quantitative and volumetric measurements such as tumour standardised uptake value (SUV)max, SUVmean and Molecular Tumour Volume (MTV-PSMA) were calculated on both reconstruction methods. Results were compared (Bland-Altman, Pearson correlation coefficient) including subgroups with low and high-volume disease burdens (MTV-PSMA cut-off 40 mL). RESULTS: The SUVmax and SUVmean were higher, and MTV-PSMA was lower in the BPL reconstructed images compared to the OSEM group, with a mean difference of 8.4 (17.5%), 0.7 (8.2%) and - 21.5 mL (-3.4%), respectively. There was a strong correlation between the calculated SUVmax, SUVmean, and MTV-PSMA values in the OSEM and BPL reconstructed images (Pearson r values of 0.98, 0.99, and 1.0, respectively). No patients were reclassified from low to high volume disease or vice versa when switching from OSEM to BPL reconstruction. CONCLUSIONS: [68Ga]Ga-PSMA-11 PET/CT quantitative and volumetric parameters produced by BPL and OSEM reconstruction methods are strongly correlated. Differences are proportional and small for SUVmean, which is used as a predictive biomarker. Our study suggests that both reconstruction methods are acceptable without clinical impact on quantitative or volumetric findings. For longitudinal comparison, committing to the same reconstruction method would be preferred to ensure consistency.

Exploring inflammatory bowel disease therapy targets through druggability genes: a Mendelian randomization study.

Background: Inflammatory bowel disease is an incurable group of recurrent inflammatory diseases of the intestine. Mendelian randomization has been utilized in the development of drugs for disease treatment, including the therapeutic targets for IBD that are identified through drug-targeted MR. Methods: Two-sample MR was employed to explore the cause-and-effect relationship between multiple genes and IBD and its subtypes ulcerative colitis and Crohn's disease, and replication MR was utilized to validate this causality. Summary data-based Mendelian randomization analysis was performed to enhance the robustness of the outcomes, while Bayesian co-localization provided strong evidential support. Finally, the value of potential therapeutic target applications was determined by using the estimation of druggability. Result: With our investigation, we identified target genes associated with the risk of IBD and its subtypes UC and CD. These include the genes GPBAR1, IL1RL1, PRKCB, and PNMT, which are associated with IBD risk, IL1RL1, with a protective effect against CD risk, and GPX1, GPBAR1, and PNMT, which are involved in UC risk. Conclusion: In a word, this study identified several potential therapeutic targets associated with the risk of IBD and its subtypes, offering new insights into the development of therapeutic agents for IBD.

Consumer electronics based smart technologies for enhanced terahertz healthcare having an integration of split learning with medical imaging.

The proposed work contains three major contribution, such as smart data collection, optimized training algorithm and integrating Bayesian approach with split learning to make privacy of the patent data. By integrating consumer electronics device such as wearable devices, and the Internet of Things (IoT) taking THz image, perform EM algorithm as training, used newly proposed slit learning method the technology promises enhanced imaging depth and improved tissue contrast, thereby enabling early and accurate disease detection the breast cancer disease. In our hybrid algorithm, the breast cancer model achieves an accuracy of 97.5 percent over 100 epochs, surpassing the less accurate old models which required a higher number of epochs, such as 165.

Comparison of interventions for Barrett's esophagus: A network meta-analysis.

BACKGROUND AND OBJECTIVE: Barrett's esophagus (BE) is a precancerous condition that has the potential to develop into esophageal cancer (EC). Currently, there is a wide range of management options available for individuals at different pathological stages in Barrett's esophagus (BE). However, there is currently a lack of knowledge regarding their comparative efficacy. To address this gap, we conducted a network meta-analysis of published randomized controlled trials to examine the comparative effectiveness of all regimens. METHODS: Data extracted from eligible randomized controlled trials were utilized in a Bayesian network meta-analysis to examine the relative effectiveness of BE's treatment regimens and determine their ranking in terms of efficacy. The ranking probability for each regimen was assessed using the surfaces under cumulative ranking values. The outcomes under investigation were complete ablation of BE, neoplastic progression of BE, and complete eradication of dysplasia. RESULTS: We identified twenty-three RCT studies with a total of 1675 participants, and ten different interventions. Regarding complete ablation of non-dysplastic BE, the comparative effectiveness ranking indicated that argon plasma coagulation (APC) was the most effective regimen, with the highest SUCRA value, while surveillance and PPI/H2RA were found to be the least efficacious regimens. For complete ablation of BE with low-grade dysplasia, high-grade dysplasia, or esophageal cancer, photodynamic therapy (PDT) had the highest SUCRA value of 94.1%, indicating it as the best regimen. Additionally, for complete eradication of dysplasia, SUCRA plots showed a trend in ranking PDT as the highest with a SUCRA value of 91.2%. Finally, for neoplastic progression, radiofrequency ablation (RFA) and surgery were found to perform significantly better than surveillance. The risk of bias assessment revealed that 6 studies had an overall high risk of bias. However, meta-regression with risk of bias as a covariate did not indicate any influence on the model. In terms of the Confidence in Network Meta-Analysis evaluation, a high level of confidence was found for all treatment comparisons. CONCLUSION: Endoscopic surveillance alone or PPI/H2RA alone may not be sufficient for managing BE, even in cases of non-dysplastic BE. However, APC has shown excellent efficacy in treating non-dysplastic BE. For cases of BE with low-grade dysplasia, high-grade dysplasia, or esophageal cancer, PDT may be the optimal intervention as it can induce regression of BE metaplasia and prevent future progression of BE to dysplasia and EC.

Global burden and prediction study of cutaneous squamous cell carcinoma from 1990 to 2030: A systematic analysis and comparison with China.

Background: China has the highest number of new cancer cases and deaths globally. Due to particularly low scores in health care quality for cutaneous squamous cell carcinoma (cSCC), the country's cSCC burden requires greater awareness. Consequently, we aimed to evaluate and predict the trend of the cSCC burden globally and in China from 1990 to 2030. Methods: We retrieved data from the Global Burden of Disease 2019 study, which provided estimates of the incidence, mortality, prevalence, and disability-adjusted life years (DALYs) of cSCC from 1990 to 2019. We set up joint-point analyses and Bayesian age-period-cohort (BAPC) models to predict the disease burden of cSCC up to 2030. Results: In 2019, China reported age-standardised rates of cSCC prevalence, incidence, mortality, and DALYs of 2.54, 2.12, 0.88, and 16.76 per 100 000 population, respectively. The country's prevalence and incidence rates from 1990 to 2019 were lower than the global levels, but its mortality and DALY rates were higher. The age-standardised rates were higher for males, and the disease burden increased with each age group globally and in China. Moreover, the average annual percentage change showed all indicators were growing faster than the global levels. According to the BAPC model, there will be an upward trend in the prevalence and incidence globally and in China between 2020 and 2030, with a decrease in mortality and DALYs. Conclusions: We observed an upward trend in the cSCC burden over the past 30 years in China. Prevalence and incidence are expected to continue at a higher rate than the global average in the next decade, while mortality and DALYs are predicted to decrease. As the Chinese population ages, efforts toward managing and preventing cSCC should be targeted towards the elderly population.

A novel non-negative Bayesian stacking modeling method for Cancer survival prediction using high-dimensional omics data.

BACKGROUND: Survival prediction using high-dimensional molecular data is a hot topic in the field of genomics and precision medicine, especially for cancer studies. Considering that carcinogenesis has a pathway-based pathogenesis, developing models using such group structures is a closer mimic of disease progression and prognosis. Many approaches can be used to integrate group information; however, most of them are single-model methods, which may account for unstable prediction. METHODS: We introduced a novel survival stacking method that modeled using group structure information to improve the robustness of cancer survival prediction in the context of high-dimensional omics data. With a super learner, survival stacking combines the prediction from multiple sub-models that are independently trained using the features in pre-grouped biological pathways. In addition to a non-negative linear combination of sub-models, we extended the super learner to non-negative Bayesian hierarchical generalized linear model and artificial neural network. We compared the proposed modeling strategy with the widely used survival penalized method Lasso Cox and several group penalized methods, e.g., group Lasso Cox, via simulation study and real-world data application. RESULTS: The proposed survival stacking method showed superior and robust performance in terms of discrimination compared with single-model methods in case of high-noise simulated data and real-world data. The non-negative Bayesian stacking method can identify important biological signal pathways and genes that are associated with the prognosis of cancer. CONCLUSIONS: This study proposed a novel survival stacking strategy incorporating biological group information into the cancer prognosis models. Additionally, this study extended the super learner to non-negative Bayesian model and ANN, enriching the combination of sub-models. The proposed Bayesian stacking strategy exhibited favorable properties in the prediction and interpretation of complex survival data, which may aid in discovering cancer targets.

Duhamel and transanal endorectal pull-throughs for Hirschsprung disease: a Bayesian network meta-analysis.

BACKGROUND: To comprehensively compare the effects of open Duhamel (OD), laparoscopic-assisted Duhamel (LD), transanal endorectal pull-through (TEPT), and laparoscopic-assisted endorectal pull-through (LEPT) in Hirschsprung disease. METHODS: PubMed, Embase, Cochrane Library, Web of Science, CNKI, WanFang, and VIP were comprehensively searched up to August 4, 2022. The outcomes were operation-related indicators and complication-related indicators. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach was used to evaluate the quality of evidence. Network plots, forest plots, league tables and rank probabilities were drawn for all outcomes. For measurement data, weighted mean differences (WMDs) and 95% credibility intervals (CrIs) were reported; for enumeration data, relative risks (RRs) and 95%CrIs were calculated. RESULTS: Sixty-two studies of 4781 patients were included, with 2039 TEPT patients, 1669 LEPT patients, 951 OD patients and 122 LD patients. Intraoperative blood loss in the OD group was more than that in the LEPT group (pooled WMD = 44.00, 95%CrI: 27.33, 60.94). Patients lost more blood during TEPT versus LEPT (pooled WMD = 13.08, 95%CrI: 1.80, 24.30). In terms of intraoperative blood loss, LEPT was most likely to be the optimal procedure (79.76%). Patients undergoing OD had significantly longer gastrointestinal function recovery time, as compared with those undergoing LEPT (pooled WMD = 30.39, 95%CrI: 16.08, 44.94). The TEPT group had significantly longer gastrointestinal function recovery time than the LEPT group (pooled WMD = 11.49, 95%CrI: 0.96, 22.05). LEPT was most likely to be the best operation regarding gastrointestinal function recovery time (98.28%). Longer hospital stay was observed in patients with OD versus LEPT (pooled WMD = 5.24, 95%CrI: 2.98, 7.47). Hospital stay in the TEPT group was significantly longer than that in the LEPT group (pooled WMD = 1.99, 95%CrI: 0.37, 3.58). LEPT had the highest possibility to be the most effective operation with respect to hospital stay. The significantly reduced incidence of complications was found in the LEPT group versus the LD group (pooled RR = 0.24, 95%CrI: 0.12, 0.48). Compared with LEPT, OD was associated with a significantly increased incidence of complications (pooled RR = 5.10, 95%CrI: 3.48, 7.45). Patients undergoing TEPT had a significantly greater incidence of complications than those undergoing LEPT (pooled RR = 1.98, 95%CrI: 1.63, 2.42). For complications, LEPT is most likely to have the best effect (99.99%). Compared with the LEPT group, the OD group had a significantly increased incidence of anastomotic leakage (pooled RR = 5.35, 95%CrI: 1.45, 27.68). LEPT had the highest likelihood to be the best operation regarding anastomotic leakage (63.57%). The incidence of infection in the OD group was significantly higher than that in the LEPT group (pooled RR = 4.52, 95%CrI: 2.45, 8.84). The TEPT group had a significantly increased incidence of infection than the LEPT group (pooled RR = 1.87, 95%CrI: 1.13, 3.18). LEPT is most likely to be the best operation concerning infection (66.32%). Compared with LEPT, OD was associated with a significantly higher incidence of soiling (pooled RR = 1.91, 95%CrI: 1.16, 3.17). Patients with LEPT had the greatest likelihood not to develop soiling (86.16%). In contrast to LD, LEPT was significantly more effective in reducing the incidence of constipation (pooled RR = 0.39, 95%CrI: 0.15, 0.97). LEPT was most likely not to result in constipation (97.81%). LEPT was associated with a significantly lower incidence of Hirschprung-associated enterocolitis (HAEC) than LD (pooled RR = 0.34, 95%CrI: 0.13, 0.85). The OD group had a significantly higher incidence of HAEC than the LEPT group (pooled RR = 2.29, 95%CrI: 1.31, 4.0). The incidence of HAEC was significantly greater in the TEPT group versus the LEPT group (pooled RR = 1.74, 95%CrI: 1.24, 2.45). LEPT was most likely to be the optimal operation in terms of HAEC (98.76%). CONCLUSION: LEPT may be a superior operation to OD, LD and TEPT in improving operation condition and complications, which might serve as a reference for Hirschsprung disease treatment.

Choice of baseline hazards in joint modeling of longitudinal and time-to-event cancer survival data.

Longitudinal time-to-event analysis is a statistical method to analyze data where covariates are measured repeatedly. In survival studies, the risk for an event is estimated using Cox-proportional hazard model or extended Cox-model for exogenous time-dependent covariates. However, these models are inappropriate for endogenous time-dependent covariates like longitudinally measured biomarkers, Carcinoembryonic Antigen (CEA). Joint models that can simultaneously model the longitudinal covariates and time-to-event data have been proposed as an alternative. The present study highlights the importance of choosing the baseline hazards to get more accurate risk estimation. The study used colon cancer patient data to illustrate and compare four different joint models which differs based on the choice of baseline hazards [piecewise-constant Gauss-Hermite (GH), piecewise-constant pseudo-adaptive GH, Weibull Accelerated Failure time model with GH & B-spline GH]. We conducted simulation study to assess the model consistency with varying sample size (N = 100, 250, 500) and censoring (20 %, 50 %, 70 %) proportions. In colon cancer patient data, based on Akaike information criteria (AIC) and Bayesian information criteria (BIC), piecewise-constant pseudo-adaptive GH was found to be the best fitted model. Despite differences in model fit, the hazards obtained from the four models were similar. The study identified composite stage as a prognostic factor for time-to-event and the longitudinal outcome, CEA as a dynamic predictor for overall survival in colon cancer patients. Based on the simulation study Piecewise-PH-aGH was found to be the best model with least AIC and BIC values, and highest coverage probability(CP). While the Bias, and RMSE for all the models showed a competitive performance. However, Piecewise-PH-aGH has shown least bias and RMSE in most of the combinations and has taken the shortest computation time, which shows its computational efficiency. This study is the first of its kind to discuss on the choice of baseline hazards.

BN-BacArena: Bayesian network extension of BacArena for the dynamic simulation of microbial communities.

MOTIVATION: Simulating gut microbial dynamics is extremely challenging. Several computational tools, notably the widely used BacArena, enable modeling of dynamic changes in the microbial environment. These methods, however, do not comprehensively account for microbe-microbe stimulant or inhibitory effects or for nutrient-microbe inhibitory effects, typically observed in different compounds present in the daily diet. RESULTS: Here, we present BN-BacArena, an extension of BacArena consisting on the incorporation within the native computational framework of a Bayesian network model that accounts for microbe-microbe and nutrient-microbe interactions. Using in vitro experiments, 16S rRNA gene sequencing data and nutritional composition of 55 foods, the output Bayesian network showed 23 significant nutrient-bacteria interactions, suggesting the importance of compounds such as polyols, ascorbic acid, polyphenols and other phytochemicals, and 40 bacteria-bacteria significant relationships. With test data, BN-BacArena demonstrates a statistically significant improvement over BacArena to predict the time-dependent relative abundance of bacterial species involved in the gut microbiota upon different nutritional interventions. As a result, BN-BacArena opens new avenues for the dynamic modeling and simulation of the human gut microbiota metabolism. AVAILABILITY AND IMPLEMENTATION: MATLAB and R code are available in https://github.com/PlanesLab/BN-BacArena.

Use of risk chart algorithms for the identification of psoriatic arthritis patients at high risk for cardiovascular disease: findings derived from the project CARMA cohort after a 7.5-year follow-up period.

OBJECTIVE: To assess the predictive value of four cardiovascular (CV) risk algorithms for identifying high-risk psoriatic arthritis (PsA) patients. METHODS: Evaluation of patients with PsA enrolled in the Spanish prospective project CARdiovascular in RheuMAtology. Baseline data of 669 PsA patients with no history of CV events at the baseline visit, who were followed in rheumatology outpatient clinics at tertiary centres for 7.5 years, were retrospectively analysed to test the performance of the Systematic Coronary Risk Assessment (SCORE), the modified version (mSCORE) European Alliance of Rheumatology Associations (EULAR) 2015/2016, the SCORE2 algorithm (the updated and improved version of SCORE) and the QRESEARCH risk estimator version 3 (QRISK3). RESULTS: Over 4790 years of follow-up, there were 34 CV events, resulting in a linearised rate of 7.10 per 1000 person-years (95% CI 4.92 to 9.92). The four CV risk scales showed strong correlations and all showed significant associations with CV events (p<0.001). SCORE, mSCORE EULAR 2015/2016 and QRISK3 effectively differentiated between low and high CV risk patients, although the cumulative rate of CV events observed over 7.5 years was lower than expected based on the frequency predicted by these risk scales. Additionally, model improvement was observed when combining QRISK3 with any other scale, particularly the combination of QRISK3 and SCORE2, which yielded the lowest Akaike information criterion (411.15) and Bayesian information criterion (420.10), making it the best predictive model. CONCLUSIONS: Risk chart algorithms are very useful for discriminating PsA at low and high CV risk. An integrated model featuring QRISK3 and SCORE2 yielded the optimal synergy of QRISK3's discrimination ability and SCORE2's calibration accuracy.

Trial of Early Minimally Invasive Removal of Intracerebral Hemorrhage.

BACKGROUND: Trials of surgical evacuation of supratentorial intracerebral hemorrhages have generally shown no functional benefit. Whether early minimally invasive surgical removal would result in better outcomes than medical management is not known. METHODS: In this multicenter, randomized trial involving patients with an acute intracerebral hemorrhage, we assessed surgical removal of the hematoma as compared with medical management. Patients who had a lobar or anterior basal ganglia hemorrhage with a hematoma volume of 30 to 80 ml were assigned, in a 1:1 ratio, within 24 hours after the time that they were last known to be well, to minimally invasive surgical removal of the hematoma plus guideline-based medical management (surgery group) or to guideline-based medical management alone (control group). The primary efficacy end point was the mean score on the utility-weighted modified Rankin scale (range, 0 to 1, with higher scores indicating better outcomes, according to patients' assessment) at 180 days, with a prespecified threshold for posterior probability of superiority of 0.975 or higher. The trial included rules for adaptation of enrollment criteria on the basis of hemorrhage location. A primary safety end point was death within 30 days after enrollment. RESULTS: A total of 300 patients were enrolled, of whom 30.7% had anterior basal ganglia hemorrhages and 69.3% had lobar hemorrhages. After 175 patients had been enrolled, an adaptation rule was triggered, and only persons with lobar hemorrhages were enrolled. The mean score on the utility-weighted modified Rankin scale at 180 days was 0.458 in the surgery group and 0.374 in the control group (difference, 0.084; 95% Bayesian credible interval, 0.005 to 0.163; posterior probability of superiority of surgery, 0.981). The mean between-group difference was 0.127 (95% Bayesian credible interval, 0.035 to 0.219) among patients with lobar hemorrhages and -0.013 (95% Bayesian credible interval, -0.147 to 0.116) among those with anterior basal ganglia hemorrhages. The percentage of patients who had died by 30 days was 9.3% in the surgery group and 18.0% in the control group. Five patients (3.3%) in the surgery group had postoperative rebleeding and neurologic deterioration. CONCLUSIONS: Among patients in whom surgery could be performed within 24 hours after an acute intracerebral hemorrhage, minimally invasive hematoma evacuation resulted in better functional outcomes at 180 days than those with guideline-based medical management. The effect of surgery appeared to be attributable to intervention for lobar hemorrhages. (Funded by Nico; ENRICH ClinicalTrials.gov number, NCT02880878.).

Prediction of the Ki-67 expression level in head and neck squamous cell carcinoma with machine learning-based multiparametric MRI radiomics: a multicenter study.

BACKGROUND: This study aimed to develop and validate a machine learning (ML)-based fusion model to preoperatively predict Ki-67 expression levels in patients with head and neck squamous cell carcinoma (HNSCC) using multiparametric magnetic resonance imaging (MRI). METHODS: A total of 351 patients with pathologically proven HNSCC from two medical centers were retrospectively enrolled in the study and divided into training (n = 196), internal validation (n = 84), and external validation (n = 71) cohorts. Radiomics features were extracted from T2-weighted images and contrast-enhanced T1-weighted images and screened. Seven ML classifiers, including k-nearest neighbors (KNN), support vector machine (SVM), logistic regression (LR), random forest (RF), linear discriminant analysis (LDA), naive Bayes (NB), and eXtreme Gradient Boosting (XGBoost) were trained. The best classifier was used to calculate radiomics (Rad)-scores and combine clinical factors to construct a fusion model. Performance was evaluated based on calibration, discrimination, reclassification, and clinical utility. RESULTS: Thirteen features combining multiparametric MRI were finally selected. The SVM classifier showed the best performance, with the highest average area under the curve (AUC) of 0.851 in the validation cohorts. The fusion model incorporating SVM-based Rad-scores with clinical T stage and MR-reported lymph node status achieved encouraging predictive performance in the training (AUC = 0.916), internal validation (AUC = 0.903), and external validation (AUC = 0.885) cohorts. Furthermore, the fusion model showed better clinical benefit and higher classification accuracy than the clinical model. CONCLUSIONS: The ML-based fusion model based on multiparametric MRI exhibited promise for predicting Ki-67 expression levels in HNSCC patients, which might be helpful for prognosis evaluation and clinical decision-making.

Predictive modeling for breast cancer classification in the context of Bangladeshi patients by use of machine learning approach with explainable AI.

Breast cancer has rapidly increased in prevalence in recent years, making it one of the leading causes of mortality worldwide. Among all cancers, it is by far the most common. Diagnosing this illness manually requires significant time and expertise. Since detecting breast cancer is a time-consuming process, preventing its further spread can be aided by creating machine-based forecasts. Machine learning and Explainable AI are crucial in classification as they not only provide accurate predictions but also offer insights into how the model arrives at its decisions, aiding in the understanding and trustworthiness of the classification results. In this study, we evaluate and compare the classification accuracy, precision, recall, and F1 scores of five different machine learning methods using a primary dataset (500 patients from Dhaka Medical College Hospital). Five different supervised machine learning techniques, including decision tree, random forest, logistic regression, naive bayes, and XGBoost, have been used to achieve optimal results on our dataset. Additionally, this study applied SHAP analysis to the XGBoost model to interpret the model's predictions and understand the impact of each feature on the model's output. We compared the accuracy with which several algorithms classified the data, as well as contrasted with other literature in this field. After final evaluation, this study found that XGBoost achieved the best model accuracy, which is 97%.