RESUMO
Deep learning (DL) models have provided state-of-the-art performance in various medical imaging benchmarking challenges, including the Brain Tumor Segmentation (BraTS) challenges. However, the task of focal pathology multi-compartment segmentation (e.g., tumor and lesion sub-regions) is particularly challenging, and potential errors hinder translating DL models into clinical workflows. Quantifying the reliability of DL model predictions in the form of uncertainties could enable clinical review of the most uncertain regions, thereby building trust and paving the way toward clinical translation. Several uncertainty estimation methods have recently been introduced for DL medical image segmentation tasks. Developing scores to evaluate and compare the performance of uncertainty measures will assist the end-user in making more informed decisions. In this study, we explore and evaluate a score developed during the BraTS 2019 and BraTS 2020 task on uncertainty quantification (QU-BraTS) and designed to assess and rank uncertainty estimates for brain tumor multi-compartment segmentation. This score (1) rewards uncertainty estimates that produce high confidence in correct assertions and those that assign low confidence levels at incorrect assertions, and (2) penalizes uncertainty measures that lead to a higher percentage of under-confident correct assertions. We further benchmark the segmentation uncertainties generated by 14 independent participating teams of QU-BraTS 2020, all of which also participated in the main BraTS segmentation task. Overall, our findings confirm the importance and complementary value that uncertainty estimates provide to segmentation algorithms, highlighting the need for uncertainty quantification in medical image analyses. Finally, in favor of transparency and reproducibility, our evaluation code is made publicly available at https://github.com/RagMeh11/QU-BraTS.
RESUMO
Tacrolimus exhibits high inter-patient pharmacokinetics (PK) variability, as well as a narrow therapeutic index, and therefore requires therapeutic drug monitoring. Germline mutations in cytochrome P450 isoforms 4 and 5 genes (CYP3A4/5) and the ATP-binding cassette B1 gene (ABCB1) may contribute to interindividual tacrolimus PK variability, which may impact clinical outcomes among allogeneic hematopoietic stem cell transplantation (HSCT) patients. In this study, 252 adult patients who received tacrolimus for acute graft versus host disease (aGVHD) prophylaxis after allogeneic HSCT were genotyped to evaluate if germline genetic variants associated with tacrolimus PK and pharmacodynamic (PD) variability. Significant associations were detected between germline variants in CYP3A4/5 and ABCB1 and PK endpoints (e.g., median steady-state tacrolimus concentrations and time to goal tacrolimus concentration). However, significant associations were not observed between CYP3A4/5 or ABCB1 germline variants and PD endpoints (e.g., aGVHD and treatment-emergent nephrotoxicity). Decreased age and CYP3A5*1/*1 genotype were independently associated with subtherapeutic tacrolimus trough concentrations while CYP3A5*1*3 or CYP3A5*3/*3 genotypes, myeloablative allogeneic HSCT conditioning regimen (MAC) and increased weight were independently associated with supratherapeutic tacrolimus trough concentrations. Future lines of prospective research inquiry are warranted to use both germline genetic and clinical data to develop precision dosing tools that will optimize both tacrolimus dosing and clinical outcomes among adult HSCT patients.