Extubation failure is a complex and ongoing problem in the intensive care unit (ICU). It refers to the patients who require re-intubation after extubation (namely disconnection from mechanical ventilation). In these patients, extubation failure leads to severe risks associated with re-intubation and is associated with increased mortalities, longer stay in ICU and also higher health care costs. Many studies have been proposed to analyze the problem of extubation failure and identify possible factors or indices that may predict extubation failure. However, these studies used a small number of patients for extubation failure and limited their features to several vital signs or main characteristics. We argue that these are insufficient and less accurate for the prediction of extubation failure. In this paper, we analyze 3636 adult patient records in the MIMIC-III clinical database and apply the Light Gradient Boosting Machine (LightGBM) to predict extubation failure. Also, we perform feature importance analysis according to the result of LightGBM and interpret these features using SHapley Additive exPlanations (SHAP). Experimental results show that our LightGBM method is effective in predicting extubation failure and outperform other machine learning methods such as artificial neural network (ANN), logistic regression (LR) and support vector machine (SVM). The results of feature importance and SHAP analysis are also proved effective and accurate.

No abstract

Motivation Drug response prediction (DRP) plays an important role in precision medicine (e.g., for cancer analysis and treatment). Recent advances in deep learning algorithms make it possible to predict drug responses accurately based on genetic profiles. However, existing methods ignore the potential relationships among genes. In addition, similarity among cell lines/drugs was rarely considered explicitly. Results We propose a novel DRP framework, called TGSA, to make better use of prior domain knowledge. TGSA consists of Twin Graph neural networks for Drug Response Prediction (TGDRP) and a Similarity Augmentation (SA) module to fuse fine-grained and coarse-grained information. Specifically, TGDRP abstracts cell lines as graphs based on STRING protein-protein association networks and employs Graph Neural Networks (GNNs) for representation learning. SA views DRP as an edge regression problem on a heterogeneous graph and utilizes GNNs to smooth the representations of similar cell lines/drugs. Besides, we introduce an auxiliary pre-training strategy to remedy the identified limitations of scarce data and poor out-of-distribution generalization. Extensive experiments on the GDSC2 dataset demonstrate that our TGSA consistently outperforms all the state-of-the-art baselines under various experimental settings. We further evaluate the effectiveness and contributions of each component of TGSA via ablation experiments. The promising performance of TGSA shows enormous potential for clinical applications in precision medicine. Availability The source code is available at https://github.com/violet-sto/TGSA. Supplementary information Supplementary data are available at Bioinformatics online.

No abstract

In clinical practice, medical image interpretation often involves multi-labeled classification, since the affected parts of a patient tend to present multiple symptoms or comorbidities. Recently, deep learning based frameworks have attained expertlevel performance on medical image interpretation, which can be attributed partially to large amounts of accurate annotations. However, manually annotating massive amounts of medical images is impractical, while automatic annotation is fast but imprecise (possibly introducing corrupted labels). In this work, we propose a new regularization approach, called Flow-Mixup, for multi-labeled medical image classification with corrupted labels. Flow-Mixup guides the models to capture robust features for each abnormality, thus helping handle corrupted labels effectively and making it possible to apply automatic annotation. Specifically, Flow-Mixup decouples the extracted features by adding constraints to the hidden states of the models. Also, Flow-Mixup is more stable and effective comparing to other known regularization methods, as shown by theoretical and empirical analyses. Experiments on two electrocardiogram datasets and a chest X-ray dataset containing corrupted labels verify that Flow-Mixup is effective and insensitive to corrupted labels.