Search

Missing data is a major problem in real-world datasets, which hinders the performance of data analytics. Conventional data imputation schemes such as univariate single imputation replace missing values in each column with the same approximated value. These univariate single imputation techniques underestimate the variance of the imputed values. On the other hand, multivariate imputation explores the relationships between different columns of data, to impute the missing values. Reinforcement Learning (RL) is a machine learning paradigm where the agent learns by taking actions and receiving rewards in response, to achieve its goal. In this work, we propose an RL-based approach to impute missing data by learning a policy to impute data through an action-reward-based experience. Our approach imputes missing values in a column by working only on the same column (similar to univariate single imputation) but imputes the missing values in the column with different values thus keeping the variance in the imputed values. We report superior performance of our approach, compared with other imputation techniques, on a number of datasets.

In the past decade, deep learning (DL) has achieved unprecedented success in numerous fields, such as computer vision and healthcare. Particularly, DL is experiencing an increasing development in advanced medical image analysis applications in terms of segmentation, classification, detection, and other tasks. On the one hand, tremendous needs that leverage DL’s power for medical image analysis arise from the research community of a medical, clinical, and informatics background to share their knowledge, skills, and experience jointly. On the other hand, barriers between disciplines are on the road for them, often hampering a full and efficient collaboration. To this end, we propose our novel open-source platform, i.e., MEDAS–the MEDical open-source platform As Service. To the best of our knowledge, MEDAS is the first open-source platform providing collaborative and interactive services for researchers from a medical background using DL-related toolkits easily and for scientists or engineers from informatics modeling faster. Based on tools and utilities from the idea of RINV (Rapid Implementation aNd Verification), our proposed platform implements tools in pre-processing, post-processing, augmentation, visualization, and other phases needed in medical image analysis. Five tasks, concerning lung, liver, brain, chest, and pathology, are validated and demonstrated to be efficiently realizable by using MEDAS.

Alzheimer’s disease (AD) is the most common type of dementia that still has no effective treatment. Accurate classification of AD can help in its diagnosis and selection of the most effective treatment options. In the last decade, several studies have proven the effectiveness of deep learning algorithms for AD diagnosis. In this paper, we propose a U-net style model for AD diagnosis using 3D T1-weighted magnetic resonance images (MRI). Combining with deep supervision has been proved to be effective in improving the performance of the model. Our method has been tested on a subset of ADNI dataset and AIBL dataset and achieves a superior average accuracy of 95.71±1.36% for AD versus NC (normal control), 90.14±3.66% for EMCI (early mild cognitive impairment) versus LMCI (late mild cognitive impairment), 90.05±2.63% for AD versus LMCI, and 87.98±4.54% for NC versus EMCI, respectively. Besides these binary-classification tasks, we also test this model for multi-class classification task (AD vs. NC vs. EMCI vs. LMCI) and it achieves an accuracy of 86.47±9.60%. Furthermore, 3D-Grad-CAM method is used to visualize the focused areas of the proposed model. We find that the proposed model pays more attention to the characteristics of the ventricles, hippocampus, and some regions of cortex, which have been proven to be affected by AD.