PLoS medicine
-
Multicenter Study
Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study.
For virtually every patient with colorectal cancer (CRC), hematoxylin-eosin (HE)-stained tissue slides are available. These images contain quantitative information, which is not routinely used to objectively extract prognostic biomarkers. In the present study, we investigated whether deep convolutional neural networks (CNNs) can extract prognosticators directly from these widely available images. ⋯ In our retrospective study, we show that a CNN can assess the human tumor microenvironment and predict prognosis directly from histopathological images.
-
Randomized Controlled Trial Multicenter Study
Raltegravir-intensified initial antiretroviral therapy in advanced HIV disease in Africa: A randomised controlled trial.
In sub-Saharan Africa, individuals infected with HIV who are severely immunocompromised have high mortality (about 10%) shortly after starting antiretroviral therapy (ART). This group also has the greatest risk of morbidity and mortality associated with immune reconstitution inflammatory syndrome (IRIS), a paradoxical response to successful ART. Integrase inhibitors lead to significantly more rapid declines in HIV viral load (VL) than all other ART classes. We hypothesised that intensifying standard triple-drug ART with the integrase inhibitor, raltegravir, would reduce HIV VL faster and hence reduce early mortality, although this strategy could also risk more IRIS events. ⋯ Although 12 weeks of raltegravir intensification was well tolerated and reduced HIV viraemia significantly faster than standard triple-drug ART during the time of greatest risk for early death, this strategy did not reduce mortality or clinical events in this group and is not warranted. There was no excess of IRIS-compatible events, suggesting that integrase inhibitors can be used safely as part of standard triple-drug first-line therapy in severely immunocompromised individuals.
-
Multicenter Study
Metabolic syndrome in pregnancy and risk for adverse pregnancy outcomes: A prospective cohort of nulliparous women.
Obesity increases the risk for developing gestational diabetes mellitus (GDM) and preeclampsia (PE), which both associate with increased risk for type 2 diabetes mellitus (T2DM) and cardiovascular disease (CVD) in women in later life. In the general population, metabolic syndrome (MetS) associates with T2DM and CVD. The impact of maternal MetS on pregnancy outcomes, in nulliparous pregnant women, has not been investigated. ⋯ We did not compare the impact of individual metabolic components with that of MetS as a composite, and therefore cannot conclude that MetS is better at identifying women at risk. However, more than half of the women who had MetS in early pregnancy developed a pregnancy complication compared with just over a third of women who did not have MetS. Furthermore, while increasing BMI increases the probability of GDM, the addition of MetS exacerbates this probability. Further studies are required to determine if individual MetS components act synergistically or independently.
-
Electronic medical records provide large-scale real-world clinical data for use in developing clinical decision systems. However, sophisticated methodology and analytical skills are required to handle the large-scale datasets necessary for the optimisation of prediction accuracy. Myopia is a common cause of vision loss. Current approaches to control myopia progression are effective but have significant side effects. Therefore, identifying those at greatest risk who should undergo targeted therapy is of great clinical importance. The objective of this study was to apply big data and machine learning technology to develop an algorithm that can predict the onset of high myopia, at specific future time points, among Chinese school-aged children. ⋯ To our knowledge, this study, for the first time, used large-scale data collected from electronic health records to demonstrate the contribution of big data and machine learning approaches to improved prediction of myopia prognosis in Chinese school-aged children. This work provides evidence for transforming clinical practice, health policy-making, and precise individualised interventions regarding the practical control of school-aged myopia.
-
Multicenter Study
Characterising risk of in-hospital mortality following cardiac arrest using machine learning: A retrospective international registry study.
Resuscitated cardiac arrest is associated with high mortality; however, the ability to estimate risk of adverse outcomes using existing illness severity scores is limited. Using in-hospital data available within the first 24 hours of admission, we aimed to develop more accurate models of risk prediction using both logistic regression (LR) and machine learning (ML) techniques, with a combination of demographic, physiologic, and biochemical information. ⋯ ML approaches significantly enhance predictive discrimination for mortality following cardiac arrest compared to existing illness severity scores and LR, without the use of pre-hospital data. The discriminative ability of these ML models requires validation in external cohorts to establish generalisability.