PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Bayesian modelling”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed1987

Bayesian statistical theory in the preoperative diagnosis of pulmonary lesions.

We used a computerized Bayesian algorithm to assist in the preoperative diagnosis of pulmonary lesions. One hundred consecutive patients who were undergoing exploratory thoracotomy for newly discovered pulmonary lesions were prospectively evaluated. The Bayesian model used a total of 44 preoperative clinical and roentgenographic factors to categorize the lesions as benign or malignant. The Bayesian algorithm correctly categorized 96 of the 100 lesions, thereby providing an accuracy of 96 percent. The sensitivity of the model was 98 percent and the specificity was 87 percent. All but two of the 85 malignant lesions were correctly categorized and 13 of the 15 benign lesions were correctly analyzed by the model. These results indicate that computer-assisted diagnosis using the Theorem of Bayes may provide valuable preoperative information for the management of selected patients.

Adolescent↗
PubMed1987

The "constant intake rate" assumption in interim recruitment goal methodology for multicenter clinical trials.

A primary concern of any multihospital clinical trial is the recruitment of a predetermined number of patients during a prespecified interval of time. In several recent papers a Poisson based model was used to estimate the time needed to recruit a predetermined number of patients and the probabilities of recruiting specified fractions of the sample during subintervals. The Poisson model requires the assumption that patients be recruited at a constant rate over the entire length of the interval. In this paper we test the adequacy of this model and assumption using patient intake data from nine multihospital VA clinical trials and propose an alternative Bayesian model.

Bayes Theorem↗
PubMed1992

Ranitidine pharmacokinetics and adverse central nervous system reactions.

BACKGROUND: Treatment with histamine2-receptor antagonists has been associated with adverse central nervous system reactions (CNS-ADRs). Previous studies of cimetidine have shown an association between CNS-ADRs and high cimetidine drug levels. While case reports of ranitidine CNS-ADRs have appeared, we wanted to study a series of patients, some of whom were critically ill, for the presence of CNS-ADRs and to correlate these with ranitidine pharmacokinetics. METHODS: A prospective, observational, open study included 163 consecutive patients, of whom 41 met entry criteria. A nonlinear least-squares regression analysis was used to establish a ranitidine pharmacokinetic dosing model. Ranitidine levels were determined by a high-performance liquid chromatographic assay. Individual ranitidine pharmacokinetics were determined by means of a bayesian model. Observations on 13 possible CNS-ADRs were recorded. The CNS-ADRs were evaluated by the Naranjo rating system. RESULTS: Ranitidine-associated CNS-ADRs, particularly lethargy, confusion, somnolence, and disorientation, occurred more frequently in patients with renal function impairment, and these were associated with higher peak concentrations, average plasma concentrations, and area under the curve. CONCLUSIONS: Ranitidine, when given in conventional doses, can cause CNS-ADRs, particularly in older patients who have substantial renal function impairment. These CNS-ADRs occur as a consequence of altered ranitidine disposition. Ranitidine doses should be reduced when renal function impairment is present, and patients should be carefully observed for CNS-ADRs.

Aged↗
PubMed2025

Inferring the sensitivity of wastewater metagenomic sequencing for early detection of viruses: a statistical modelling study.

BACKGROUND: Metagenomic sequencing of wastewater (W-MGS) can in principle detect any known or novel pathogen in a population. We aimed to quantify the sensitivity and cost of W-MGS for viral pathogen detection by jointly analysing W-MGS and epidemiological data for a range of human-infecting viruses. METHODS: In this statistical modelling study, we analysed sequencing data from four studies of untargeted W-MGS to estimate the relative abundance of 11 human-infecting viruses. Corresponding prevalence and incidence estimates were obtained or calculated from academic and public health reports. We combined these estimates using a hierarchical Bayesian model to predict relative abundance at set prevalence or incidence values, allowing comparison across studies and viruses. These predictions were then used to estimate the sequencing depth and concomitant cost required for pathogen detection using W-MGS with or without use of a hybridisation capture enrichment panel. FINDINGS: After controlling for variation in local infection rates, relative abundance varied by orders of magnitude across studies for a given virus. For instance, a local SARS-CoV-2 weekly incidence of 1% corresponded to a predicted SARS-CoV-2 relative abundance ranging from 3·8 × 10-10 to 2·4 × 10-7 across studies, translating to orders-of-magnitude variation in the cost of operating a system able to detect a SARS-CoV-2-like pathogen at a given sensitivity. Use of a respiratory virus enrichment panel in two studies greatly increased predicted relative abundance of SARS-CoV-2, lowering yearly costs by 27-fold (from US$7·87 million to $287 000) and 29-fold (from $1·98 million to $69 100) for a system able to detect a SARS-CoV-2-like pathogen before reaching 0·01% cumulative incidence. INTERPRETATION: The large variation in viral relative abundance after controlling for epidemiological factors indicates that other sources of inter-study variation, such as differences in sewershed hydrology and laboratory protocols, have a substantial impact on the sensitivity and cost of W-MGS. Well chosen hybridisation capture panels can greatly increase sensitivity and reduce cost for viruses in the panel, but might reduce sensitivity to unknown or unexpected pathogens. FUNDING: The Wellcome Trust, Open Philanthropy, and Musk Foundation.

Humans↗
PubMed2025

BTS: a scalable Bayesian Tissue Score for prioritizing GWAS variants and their functional contexts across >1000s of omics datasets.

MOTIVATION: statistics from genome-wide association studies (GWAS) are widely used in fine-mapping and colocalization analyses to identify causal variants and their enrichment in functional contexts, such as affected cell types and genomic features. With the expansion of functional genomic (FG) datasets, which now include hundreds of thousands of tracks across various cell and tissue types, it is critical to establish scalable algorithms integrating thousands of diverse FG annotations with GWAS results. RESULTS: We propose BTS (Bayesian Tissue Score), a novel, highly efficient algorithm uniquely designed for (i) identifying affected cell types and functional elements (context-mapping) and (ii) fine-mapping potentially causal variants in a context-specific manner using large collections of cell type-specific FG annotation tracks. BTS leverages GWAS summary statistics and annotation-specific Bayesian models to analyze genome-wide annotation tracks, including enhancers, open chromatin, and histone marks. We evaluated BTS on GWAS summary statistics for immune and cardiovascular traits, such as Inflammatory Bowel Disease (IBD), Rheumatoid Arthritis (RA), Systemic Lupus Erythematosus (SLE), and Coronary Artery Disease (CAD). Our results demonstrate that BTS is over 100× more efficient in estimating functional annotation effects and context-specific variant fine-mapping compared to existing methods. Importantly, this large-scale Bayesian approach prioritizes both known and novel annotations, cell types, genomic regions, and variants and provides valuable biological insights into the functional contexts of these diseases. AVAILABILITY AND IMPLEMENTATION: Docker image is available at https://hub.docker.com/r/wanglab/bts with preinstalled BTS R package (https://bitbucket.org/wanglab-upenn/BTS-R) and BTS GWAS summary statistics analysis pipeline (https://bitbucket.org/wanglab-upenn/bts-pipeline).

Genome-Wide Association Study↗
PubMed2026

Polygenic Risk Scores for Incident Dementia in the Multi-Ethnic Study of Atherosclerosis.

Over 75 Alzheimer's disease (AD) and dementia-associated variants have been identified through genome-wide association studies, but the utility of polygenic risk scores (PRS) for predicting AD and dementia in diverse and admixed populations remains unclear. We compared how PRS approaches differing in p-value thresholds, variant weights, and source ancestry perform in predicting dementia in 6338 African American, Chinese, Hispanic, and White individuals from the Multi-Ethnic Study of Atherosclerosis. We tested clumping and thresholding (C+T) methods with varying parameters against Bayesian approaches (PRS-CS, PRS-CSx). We compared the ability of each method to predict incident dementia in all participants and in groups stratified by self-reported race/ethnicity. We additionally analyzed performance across groups stratified by estimated proportion of non-Finnish European (NFE)-like ancestry. Including more variants does not improve performance. We found comparable associations between dementia and PRS when comparing a C+T method with only 15 SNPs and PRS derived from Bayesian models that include >&#x2009;800,000 SNPs (HR5e-08 = 1.18, 95% CI: 1.08-1.28; HRCSx = 1.17, 95% CI: 1.07-1.27). The p&#x2009;<&#x2009;5e-08 C+T method was more strongly associated with incident dementia in populations genetically dissimilar from the source data (HRlowNFE_5e-08 = 1.27, 95% CI: 1.08-1.50; HRlowNFE_CSx = 1.12, 95% CI: 0.94-1.33). More selective PRS models using genome-wide significant SNPs may be preferable for dementia prediction in diverse populations.

Aged↗
PubMed1991

Ruling out acute myocardial infarction. A prospective multicenter validation of a 12-hour strategy for patients at low risk.

BACKGROUND: Although previous investigations have suggested that 24 hours is required to exclude acute myocardial infarction in patients who are admitted to a coronary care unit for the evaluation of acute chest pain, we hypothesized that a 12-hour period might be adequate for patients with a low probability of infarction at the time of admission. METHODS: Using a Bayesian model, we developed a strategy to identify candidates for a shorter period of observation from an analysis of a derivation set of 976 patients with acute chest pain who were admitted to three teaching and four community hospitals. In the derivation set, patients whose clinical characteristics in the emergency room predicted a low (less than or equal to 7 percent) probability of myocardial infarction had only a 0.4 percent risk of infarction if they had neither abnormal levels of cardiac enzymes nor recurrent ischemic pain during the first 12 hours of hospitalization. In an independent testing set of 2684 patients from the seven hospitals, 957 admitted patients (36 percent) were classified as candidates for this 12-hour period of observation according to a previously published multivariate algorithm. Few of these patients were actually transferred from a monitored setting at 12 hours. RESULTS: Of the 771 candidates for a 12-hour period of observation who did not have enzyme abnormalities or recurrent pain during the first 12 hours, 4 (0.5 percent) were subsequently found to have acute myocardial infarction, and only 3 (0.4 percent) died after primary cardiac arrests, all of which occurred three to five days after admission. Rates of other major cardiovascular complications were low in the patients who might have been transferred from the coronary care unit after 12 hours with this strategy. In patients with a higher initial risk of infarction, the standard strategy of 24-hour observation identified all but 11 of 739 acute myocardial infarctions (1 percent). CONCLUSIONS: Emergency room clinical data can be used to identify a large subgroup of patients for whom a 12-hour period of observation is normally sufficient to exclude acute myocardial infarction. Patient-specific evaluation and treatment can then proceed without the restrictions imposed by "rule-out" protocols for myocardial infarction.

Adult↗
PubMed2026

Mutational signatures in blood-brain barrier: mechanisms, computational insights, and clinical applications in precision oncology.

The blood - brain barrier (BBB) plays a central role in maintaining central nervous system (CNS) homeostasis, and its disruption is a defining feature of malignant brain tumors such as glioblastoma. Emerging evidence indicates that BBB dysfunction not only alters the tumor microenvironment but also shapes the mutational processes that drive genomic instability in CNS malignancies. This review synthesizes current understanding of the biological mechanisms linking BBB breakdown with distinct mutational signatures, including those arising from oxidative stress, hypoxia-induced replication stress, lipid peroxidation, inflammation, and metabolic reprogramming. Advances in next-generation sequencing, coupled with computational tools such as non-negative matrix factorization, Bayesian modeling, and deep learning, have enabled precise extraction of these signatures and their integration with multi-omics data. Clinically, BBB-associated mutational signatures offer significant promise for therapeutic stratification, prediction of treatment response, and noninvasive monitoring through cerebrospinal fluid - derived circulating tumor DNA. Despite these advances, challenges persist due to limited tissue accessibility, low-yield CSF samples, incomplete mechanistic models, and the lack of CNS-specific analytical frameworks. A deeper understanding of BBB-driven mutational processes, supported by improved computational approaches and integrative datasets, holds potential to advance precision oncology in neuro-oncology.

Humans↗
PubMed2026

Identifying multigenic modules under selection in the tumor genome.

MOTIVATION: Genomic alterations in cancer arise from selective pressures acting on hallmark molecular modules, layered over a background of random mutagenic events. Methods to detect selection at the level of modules, as opposed to genes or nucleotides, are relatively underdeveloped. RESULTS: Here we present CanSRMaPP (Cancer Selection Recovery by Maximum Posterior Probability), a Bayesian model of the cancer genome that infers mutational selection on single genes and multi-genic modules while simultaneously modeling background events. Applying CanSRMaPP to lung adenocarcinoma genomes, we identify positive selection on 63 modules, yielding a model that parsimoniously explains the observed pattern of genetic alterations observed in new cancer cohorts. We further show that CanSRMaPP is adaptable to more tumor types and to alternative module definitions. We show that these modules serve as an effective scaffold for translating the cancer genome to molecular states, with prediction of cancer biomarker status as demonstration. AVAILABILITY: CanSRMaPP is freely available on GitHub. SUPPLEMENTARY INFORMATION: Supplementary Figs. S1-5, Supplementary Tables S1-5, and Supplementary Notes 1 and 2 are available at Bioinformatics online.

Journal Article↗
PubMed1989

Evaluation of a two-compartment Bayesian forecasting program for predicting vancomycin concentrations.

The application of a two-compartment Bayesian forecasting program for vancomycin was tested retrospectively in 45 adult patients with stable renal function. Serial blood samples from 25 of these patients were used to determine population-based parameter estimates. The predictive performance of the Bayesian program was assessed by using both non-steady-state and steady-state vancomycin concentrations as feedback information. Overall, the program tended to underpredict peak and trough steady-state vancomycin serum concentrations. A larger mean prediction error (ME) was seen when non-steady-state feedback serum concentrations were used compared with using population-based parameter estimates (no feedback). In contrast, a marked improvement in ME (peaks: -1.03 versus -2.61; troughs: -1.60 versus -2.07) was seen when steady-state feedback serum concentrations were used compared with no feedback data. Precision improved when either feedback serum concentrations were used to predict steady-state peak and trough vancomycin concentrations. The results from this clinical evaluation demonstrate that the initial pharmacokinetic parameter estimates for a two-compartment Bayesian model provided accurate prediction of steady-state vancomycin concentrations. Prediction bias and precision were improved when steady-state vancomycin concentrations were used to determine individualized pharmacokinetic parameters.

Adult↗
PubMed1992

Predictive performance of Bayesian and nonlinear least-squares regression programs for lidocaine.

The predictive performance of two computer programs for lidocaine dosing were evaluated. Two-compartment Bayesian and nonlinear least-squares regression programs were used in two groups of patients (15 acute arrhythmia patients and 14 chronic arrhythmia patients). Lidocaine was given as a 1.5 mg/kg bolus and a 2.8 mg/min infusion for 48 h. A second bolus (0.5 mg/kg) was given 10 min after the first bolus over 2 min. Serum samples of the patients receiving lidocaine were drawn at 2, 15, 30 min and 1, 2, and 4 h and were used in forecasting the serum concentrations at 6, 8, 12, and 48 h. Predictive performance was assessed by mean error and mean-squared error. The results (mean +/- 95% confidence intervals) demonstrated the Bayesian program predicted a significant (p less than 0.05) difference at 12 h between the two arrhythmia groups (acute 0.52 [-0.95; -0.09] and chronic 0.28 [0.12; 0.44]). The results also demonstrated the Bayesian method was significantly more precise compared to the nonlinear least-squares regression program at 8, 12, and 48 h for the acute group. While caution is warranted, this study demonstrated that the predictive performance by a two-compartment Bayesian model is more accurate in predicting future lidocaine serum concentrations than that by nonlinear least-squares regression.

Acute Disease↗
PubMed2026

Phylogeographic epidemiology of Dabie bandavirus in East Asia: divergent transmission networks and genotype&#x2011;linked clinical severity.

BACKGROUND: Severe fever with thrombocytopenia syndrome (SFTS), caused by Dabie bandavirus (SFTSV), exhibits geographically decoupled incidence and fatality patterns across East Asia. We aimed to elucidate the distinct ecological drivers and phylogeographic dynamics underlying this inland-coastal epidemiological divergence. METHODS: Integrating 1820 high-quality global genomes of SFTSV with well-characterized clinical cohorts (936 patients) and nationwide surveillance data (27,457 cases) from China, we constructed a comprehensive analytical framework. Ecological modeling, Bayesian phylogeography, and genotype-phenotype association analyses were employed to trace the evolutionary trajectories and clinical implications of the virus. RESULTS: A pronounced "inland-high-incidence vs. coastal-high-fatality" pattern of SFTS was identified. The incidence of SFTS exhibited divergent sensitivities to meteorological factors; inland transmission was sensitive to thermal fluctuations, whereas coastal dynamics were constrained by a sunshine threshold (>&#x2009;200&#xa0;h/month). In contrast, spatial divergence in clinical severity correlated with the distribution of regional viral genetic structures. Inland regions mainly co-circulated genotypes A, C, and D, while coastal regions were dominated by genotype B. Zhejiang province was identified as a genetic hub with significantly higher recombination frequencies than inland regions (11.0% vs. 3.5%, P < 0.001). Bayesian phylogeographic inference indicated frequent lineage exchange of Zhejiang province in China with the Republic of Korea and Japan. Clinically, genotypes B and D were associated with elevated mortality in coastal and inland regions, respectively, suggesting that the severe coastal phenotype is shaped by its genotype B-dominated structure. Additionally, the RdRp-N828S mutation emerged as a robust molecular correlate of fatal outcomes, warranting further functional validation. CONCLUSIONS: Divergent meteorological factors and plausible maritime transmission networks may underlie the geographically decoupled epidemiology of SFTS. These findings highlight that risk assessment must extend beyond incidence alone and provide a phylogeographically informed framework for targeted surveillance and genotype-specific interventions in high-risk hotspots.

Humans↗
PubMed1990

Bayesian forecasting of serum gentamicin concentrations in intensive care patients.

This study retrospectively evaluated the predictive performance of a 1-compartment Bayesian forecasting program in adult intensive care unit (ICU) patients with stable renal function. A comparison was made of the reliability of 3 sets of population-based parameter estimates and 2 serum concentration monitoring strategies. A larger mean error for prediction of peak gentamicin concentrations was seen with literature-derived parameters than when ICU population-based parameter estimates were used. Bias and precision improved when non-steady-state peak and trough concentrations were used to predict those at steady-state; the addition of steady-state values did not provide additional information for predictions once non-steady-state feedback concentrations were incorporated. The addition of 4 serial gentamicin concentrations obtained at both non-steady-state and steady-state did not noticeably improve the predictive performance. The results demonstrate that initial ICU pharmacokinetic parameter estimates for a 1-compartment Bayesian model provide accurate prediction of steady-state gentamicin concentrations. Prediction bias and precision showed the greatest improvement when non-steady-state gentamicin concentrations were used to determine individualised pharmacokinetic parameters.

Adult↗
PubMed2026

Shielding the First 24 Postnatal Months of Life: A Proposal for a Prospective Cohort Study of Early-Life Electromagnetic Exposure and Autism Risk.

BACKGROUND: Autism Spectrum Disorder (ASD) involves Mirror Neuron System (MNS) dysfunction, driving core social and imitative impairments. Systemic physiological alterations such as autonomic dysregulation, mitochondrial dysfunction and neuroinflammation are known to impair synchronization and plasticity of neuronal clusters. A less-evident environmental cofactor, coinciding with rising ASD prevalence, is the considerable world-wide increase in electromagnetic radiation (EMR) overall exposure among children. Experimental evidence shows how low-intensity EMR influences cellular processes, via voltage-gated calcium channels (VGCCs), oxidative stress, and mitochondrial metabolism. The Resonant Convergence framework, allow to predict how chronic EMR exposure during the first 24 postnatal months of life can act as a factor in ASD pathogenesis. The best candidate mechanism is chronic Ion Cyclotron Resonance (ICR) detuning the Ca2+-calmodulin pathway, thus disrupting MNS synchronization. METHODS AND ANALYSIS: A prospective observational pilot cohort study (24-month follow-up) proposes to enroll 1000 full-term newborns into two arms: an EMR-reduced cohort (n = 500, rest and sleep-phase Faraday shielding) and a standard exposure cohort (n = 500). Exposure is quantified via radiofrequency (RF)/extremely low frequency(ELF) measurements, proximity analysis, device inventories and wearable dosimetry. The primary endpoint is a continuous neurodevelopmental trajectory score (joint attention, language, electroencephalogram (EEG) mu-rhythm); binary ASD diagnosis (Autism Diagnostic Observation Schedule, Second Edition (ADOS-2), Autism Diagnostic Interview-Revised (ADI-R)) is a secondary, exploratory endpoint. Moreover, an optional genomic screening will evaluate gene-environment interactions within extremely low-frequency electromagnetic field (ELF-EMF) vulnerable pathways, including ASD-associated genes upregulated by RF via bromodomain and extraterminal protein (BET)-mediated epigenetic mechanisms. Analyses will employ risk ratios, Fisher's exact tests and logistic regression adjusted for confounders; mixed-effects and Bayesian modeling will evaluate longitudinal outcomes and exposure reduction effects. Given a 2-3% baseline prevalence, approximately 20-30 ASD cases are expected. The study is therefore powered for exploratory signal detection rather than definitive causal inference, providing the critical baseline data required to justify and design future confirmatory trials. Sex-stratified modeling will address the 4:1 male-to-female prevalence ratio. ETHICS AND DISSEMINATION: Ethics committee approval is not yet sought; full protocol review and approval will be obtained prior to the study initiation, in strict accordance with the Declaration of Helsinki. Written parental informed consent will be mandatory for all participants prior to enrollment. Study findings and methodological milestones will be disseminated through peer-reviewed international scientific publications. This protocol provides a structured methodological framework for the first prospective investigation of sleep-phase EMR reduction as a potential modulator of ASD incidence during early neurodevelopment. Results will inform adequately powered confirmatory trials in electromagnetic neurodevelopmental epidemiology.

autism spectrum disorder↗
PubMed1991

A computational model of approximate Bayesian inference for associating clinical algorithms with decision analyses.

The lack of rationale or explanation is a major deficiency of clinical algorithms. To address this issue, the authors present a computational model for associating decision analyses with clinical algorithms. Automata theory is used to model categorical reasoning with approximate Bayesian inference based on probability intervals. This approximation reduces the number of computations to linear-order instead of the exponential-order combinations of clinical findings in exact Bayes. The linkage of decision analyses and clinical algorithms by means of this model exploits a new concept of "regular" clinical algorithms and their equivalency in theory and provides valuable perspectives in practice for developers of clinical algorithms.

Algorithms↗
PubMed1979

The Chatham Blood Pressure Study. An application of Bayesian growth curve models to a longitudinal study of blood pressure in children.

Recent developments in statistics have produced powerful methods that facilitate the analysis of longitudinal studies. These methods are illustrated by an analysis of a longitudinal study of blood pressure in children. The results of the study show a clear tendency for blood pressure to increase with age, and Asian children tend to have lower blood pressures than their Caucasian counterparts of the same age. There is evidence to support the hypothesis that blood pressures track.

Age Factors↗
PubMed1992

Suramin: rapid loading and weekly maintenance regimens for cancer patients.

PURPOSE: Suramin is an anticancer agent with a narrow therapeutic window and a terminal half-life of 45 to 55 days. These characteristics make it necessary to control accurately the serum concentrations of the drug. Therefore, the aim of the present study was to develop a rapid loading regimen, followed by weekly administration of suramin to maintain serum concentrations of between 150 and 300 micrograms/mL for 8 weeks. PATIENTS AND METHODS: Eligible patients were treated with five different loading regimens. Initially, weekly maintenance doses were estimated manually by the treating physician. Subsequently, computer-assisted dosing that used Bayesian pharmacokinetic modeling was used. RESULTS: Thirty-eight courses of suramin that were administered to 35 patients were studied. The optimal loading regimen consisted of a continuous infusion of 600 mg/m2 during a 24-hour period, which resulted in a mean serum concentration of 319 micrograms/mL. Potentially toxic concentrations that were observed with shorter infusions were avoided. Maintenance treatment, which used the weekly administration of suramin during a 6-hour period, seemed to be able to maintain mean suramin serum trough concentrations of 150 micrograms/mL, while preventing mean peak concentrations of more than 300 micrograms/mL. The use of Bayesian pharmacokinetics was superior to manual estimation in tailoring the optimal dose to the therapeutic window. CONCLUSIONS: Continuous infusion is the optimal way of delivering suramin during the loading phase. To maintain trough levels and peak levels within a narrower therapeutic window, suramin will have to be administered more frequently than once a week. Bayesian modeling based on individual serum levels and population pharmacokinetics allows accurate dosing to maintain suramin levels within the therapeutic window.

Adult↗