RESUMO
We report domain knowledge-based rules for assigning voxels in brain multiparametric MRI (mpMRI) to distinct tissuetypes based on their appearance on Apparent Diffusion Coefficient of water (ADC) maps, T1-weighted unenhanced and contrast-enhanced, T2-weighted, and Fluid-Attenuated Inversion Recovery images. The development dataset comprised mpMRI of 18 participants with preoperative high-grade glioma (HGG), recurrent HGG (rHGG), and brain metastases. External validation was performed on mpMRI of 235 HGG participants in the BraTS 2020 training dataset. The treatment dataset comprised serial mpMRI of 32 participants (total 231 scan dates) in a clinical trial of immunoradiotherapy in rHGG (NCT02313272). Pixel intensity-based rules for segmenting contrast-enhancing tumor (CE), hemorrhage, Fluid, non-enhancing tumor (Edema1), and leukoaraiosis (Edema2) were identified on calibrated, co-registered mpMRI images in the development dataset. On validation, rule-based CE and High FLAIR (Edema1 + Edema2) volumes were significantly correlated with ground truth volumes of enhancing tumor (R = 0.85;p < 0.001) and peritumoral edema (R = 0.87;p < 0.001), respectively. In the treatment dataset, a model combining time-on-treatment and rule-based volumes of CE and intratumoral Fluid was 82.5% accurate for predicting progression within 30 days of the scan date. An explainable decision tree applied to brain mpMRI yields validated, consistent, intratumoral tissuetype volumes suitable for quantitative response assessment in clinical trials of rHGG.
RESUMO
Background: Overdiagnosis is the leading factor contributing to the rapid increase in thyroid cancer incidence of the last decades. During this period, however, thyroid cancer incidence has not been increasing at a constant pace. We hypothesized that changes in the slope of the incidence trends curve, called joinpoints, could be associated with changes in clinical practice guideline recommendations. Methods: Data were obtained from the initial nine registries of the Surveillance, Epidemiology, and End Results (SEER) Program. Thyroid cancer incidence was analyzed from 1975 to 2016. Joinpoints in thyroid cancer incidence trends and clinical variables were correlated with significant changes in clinical practice recommendations. Results: Incidence rate trends of medullary and anaplastic thyroid cancer were constant during the study period. Among papillary thyroid cancers (PTCs), three main joinpoints were identified, mainly driven by changes in incidence trends of smaller cancers. First, acceleration followed by two deceleration periods in thyroid cancer incidence coincident in time with the release of American Thyroid Association guidelines in 1996, 2009, and 2015. In 1996, the use of thyroid ultrasound and fine needle aspiration biopsy for the evaluation of thyroid nodules was described; and in 2009 and 2015, higher size thresholds for the biopsy of thyroid nodules were set. For the follicular variant of PTC, a joinpoint was observed around 1988, when the histological diagnosis of this entity was revised in the World Health Organization classification; and another one in 2015 coinciding with the proposal to remove the term carcinoma from noninvasive follicular-pattern tumors with papillary-like nuclear features which contributed to drive down the overall thyroid cancer incidence. Follicular thyroid cancer incidence was affected as well by changes in the guidelines, but to a lesser extent, and it was fairly stable during the study period. Conclusions: This study suggests that thyroid cancer incidence trends have been shaped, in large part, but not completely, by changes in professional guideline recommendations.