Search for: All records

Abstract Radiogenomics uses machine-learning (ML) to directly connect the morphologic and physiological appearance of tumors on clinical imaging with underlying genomic features. Despite extensive growth in the area of radiogenomics across many cancers, and its potential role in advancing clinical decision making, no published studies have directly addressed uncertainty in these model predictions. We developed a radiogenomics ML model to quantify uncertainty using transductive Gaussian Processes (GP) and a unique dataset of 95 image-localized biopsies with spatially matched MRI from 25 untreated Glioblastoma (GBM) patients. The model generated predictions for regional EGFR amplification status (a common and important target in GBM) to resolve the intratumoral genetic heterogeneity across each individual tumor—a key factor for future personalized therapeutic paradigms. The model used probability distributions for each sample prediction to quantify uncertainty, and used transductive learning to reduce the overall uncertainty. We compared predictive accuracy and uncertainty of the transductive learning GP model against a standard GP model using leave-one-patient-out cross validation. Additionally, we used a separate dataset containing 24 image-localized biopsies from 7 high-grade glioma patients to validate the model. Predictive uncertainty informed the likelihood of achieving an accurate sample prediction. When stratifying predictions based on uncertainty, we observed substantially higher performance in the group cohort (75% accuracy, n = 95) and amongst sample predictions with the lowest uncertainty (83% accuracy, n = 72) compared to predictions with higher uncertainty (48% accuracy, n = 23), due largely to data interpolation (rather than extrapolation). On the separate validation set, our model achieved 78% accuracy amongst the sample predictions with lowest uncertainty. We present a novel approach to quantify radiogenomics uncertainty to enhance model performance and clinical interpretability. This should help integrate more reliable radiogenomics models for improved medical decision-making. 

Recent analyses of global climate models suggest that uncertainty in the coupling between midlatitude clouds and the atmospheric circulation contributes to uncertainty in climate sensitivity. However, the reasons behind model differences in the cloud–circulation coupling have remained unclear. Here, we use a global climate model in an idealized aquaplanet setup to show that the Southern Hemisphere climatological circulation, which in many models is biased equatorward, contributes to the model differences in the cloud–circulation coupling. For the same poleward shift of the Hadley cell (HC) edge, models with narrower climatological HCs exhibit stronger midlatitude cloud-induced shortwave warming than models with wider climatological HCs. This cloud-induced radiative warming results predominantly from a subsidence warming that decreases cloud fraction and is stronger for narrower HCs because of a larger meridional gradient in the vertical velocity. A comparison of our aquaplanet results with comprehensive climate models suggests that about half of the model uncertainty in the midlatitude cloud–circulation coupling stems from this impact of the circulation on the large-scale temperature structure of the atmosphere, and thus could be removed by improving the climatological circulation in models. This illustrates how understanding of large-scale dynamics can help reduce uncertainty in clouds and their response to climate change.