Evaluating batch correction methods for image-based cell profiling.

High-throughput image-based profiling platforms are powerful technologies capable of collecting data from billions of cells exposed to thousands of perturbations in a time- and cost-effective manner. Therefore, image-based profiling data has been increasingly used for diverse biological applications, such as predicting drug mechanism of action or gene function. However, batch effects pose severe limitations to community-wide efforts to integrate and interpret image-based profiling data collected across different laboratories and equipment. To address this problem, we benchmarked seven high-performing scRNA-seq batch correction techniques, representing diverse approaches, using a newly released Cell Painting dataset, the largest publicly accessible image-based dataset. We focused on five different scenarios with varying complexity, and we found that Harmony, a mixture-model based method, consistently outperformed the other tested methods. Our proposed framework, benchmark, and metrics can additionally be used to assess new batch correction methods in the future. Overall, this work paves the way for improvements that allow the community to make best use of public Cell Painting data for scientific discovery.

A novel multichannel deep learning model for fast denoising of Monte Carlo dose calculations: preclinical applications.

Objective.In preclinical radiotherapy with kilovolt (kV) x-ray beams, accurate treatment planning is needed to improve the translation potential to clinical trials. Monte Carlo based radiation transport simulations are the gold standard to calculate the absorbed dose distribution in external beam radiotherapy. However, these simulations are notorious for their long computation time, causing a bottleneck in the workflow. Previous studies have used deep learning models to speed up these simulations for clinical megavolt (MV) beams. For kV beams, dose distributions are more affected by tissue type than for MV beams, leading to steep dose gradients. This study aims to speed up preclinical kV dose simulations by proposing a novel deep learning pipeline.Approach.A deep learning model is proposed that denoises low precision (∼106simulated particles) dose distributions to produce high precision (109simulated particles) dose distributions. To effectively denoise the steep dose gradients in preclinical kV dose distributions, the model uses the novel approach to use the low precision Monte Carlo dose calculation as well as the Monte Carlo uncertainty (MCU) map and the mass density map as additional input channels. The model was trained on a large synthetic dataset and tested on a real dataset with a different data distribution. To keep model inference time to a minimum, a novel method for inference optimization was developed as well.Main results.The proposed model provides dose distributions which achieve a median gamma pass rate (3%/0.3 mm) of 98% with a lower bound of 95% when compared to the high precision Monte Carlo dose distributions from the test set, which represents a different dataset distribution than the training set. Using the proposed model together with the novel inference optimization method, the total computation time was reduced from approximately 45 min to less than six seconds on average.Significance.This study presents the first model that can denoise preclinical kV instead of clinical MV Monte Carlo dose distributions. This was achieved by using the MCU and mass density maps as additional model inputs. Additionally, this study shows that training such a model on a synthetic dataset is not only a viable option, but even increases the generalization of the model compared to training on real data due to the sheer size and variety of the synthetic dataset. The application of this model will enable speeding up treatment plan optimization in the preclinical workflow.

Diabetes, pulse pressure and cardiovascular mortality: the Hoorn Study.

OBJECTIVE: Type 2 diabetic patients have an increased arterial stiffness and a very high risk of cardiovascular death. The present study investigated the relationship between pulse pressure, an indicator of vascular stiffness, and risk of cardiovascular mortality among type 2 diabetic and non-diabetic individuals. Second, we determined the relationship between pulse pressure and its main determinant (i.e. age), and the influence of diabetes and mean arterial pressure on this relationship. DESIGN AND METHODS: We studied a cohort of 2484 individuals including 208 type 2 diabetic patients. Mean age and median follow-up for non-diabetic and diabetic individuals, respectively, were 61 and 66 years, and 8.8 and 8.6 years. One-hundred and sixteen non-diabetic and 34 diabetic individuals died of cardiovascular causes. Relative risks of cardiovascular mortality were estimated by Cox proportional hazards regression adjusted for age, gender and mean arterial pressure. RESULTS: Pulse pressure was associated with cardiovascular mortality among the diabetic, but not among the non-diabetic individuals [adjusted relative risk (95% confidence interval) per 10 mmHg increase, 1.27 (1.00-1.61) and 0.98 (0.85-1.13), P interaction = 0.07]. Further adjustment for other risk factors gave similar results. The association, at baseline, between age and pulse pressure was dependent on the presence of diabetes (P interaction = 0.03) and on the mean arterial pressure (P interaction< 0.001) (i.e. there was a stronger association when diabetes was present and when mean arterial pressure was higher). CONCLUSIONS: We conclude that, in type 2 diabetes, pulse pressure is positively associated with cardiovascular mortality. The association between age and pulse pressure is influenced by the presence of type 2 diabetes and by the height of the mean arterial pressure. These findings support the concept of accelerated vascular aging in type 2 diabetes.

Determinants of brachial artery mean 24 h pulse pressure in individuals with Type II diabetes mellitus and untreated mild hypertension.

Brachial artery pulse pressure is a predictor of (cardiovascular) morbidity, but its determinants in individuals with Type II diabetes and untreated mild hypertension have not been elucidated. We therefore cross-sectionally investigated determinants of brachial artery mean 24 h pulse pressure in 60 individuals (40 males; age, mean +/- S.D., 57.8 +/- 7.5 years) with Type II diabetes [median diabetes duration (interquartile range), 6.3 (3.6-10.1) years] and untreated mild hypertension [sitting blood pressure >140/90 mmHg and <190/120 mmHg (mean of two consecutive auscultatory office measurements after 5 min of rest)]. We measured (1) three potential determinants reflecting different aspects of central artery stiffness [the overall systemic arterial compliance, the aortic augmentation index and 1/(regional carotido-femoral transit time)], (2) structural and functional changes of the circulatory system often observed in Type II diabetes, and (3) diabetes-associated metabolic variables. After adjustment for age, gender and mean arterial pressure, brachial artery pulse pressure was associated with autonomic function [standardized regression coefficient (beta), -0.27 (P=0.01)], blood pressure decline during sleep [standardized beta, -0.32 (P=0.002)], fasting glucose concentration [standardized beta, 0.26 (P=0.01)], HbA(1c) concentration [standardized beta, 0.27 (P=0.003)] and diabetes duration [standardized beta, 0.28 (P=0.002)] in linear regression analyses. In a combined multivariate model, brachial artery pulse pressure was independently determined by gender [1=male, 2=female; standardized beta, 0.24 (P=0.01)], diabetes duration [standardized beta, 0.18 (P=0.03)], mean arterial pressure [standardized beta, 0.32 (P=0.002)], systemic arterial compliance [standardized beta, -0.23 (P=0.02)] and fasting glucose concentration [standardized beta, 0.20 (P=0.02)]. Aortic augmentation index and 1/(carotido-femoral transit time) were not independently associated with pulse pressure. In conclusion, in individuals with Type II diabetes and untreated mild hypertension, brachial artery pulse pressure is determined mainly by proximal aortic stiffness in a way which is not strongly influenced by peripheral pulse wave reflection. Approx. 60% of the variance in brachial artery pulse pressure could be explained by potentially modifiable determinants.