EAPR: explainable and augmented patient representation learning for disease prediction.

Patient representation learning aims to encode meaningful information about the patient's Electronic Health Records (EHR) in the form of a mathematical representation. Recent advances in deep learning have empowered Patient representation learning methods with greater representational power, allowing the learned representations to significantly improve the performance of disease prediction models. However, the inherent shortcomings of deep learning models, such as the need for massive amounts of labeled data and inexplicability, limit the performance of deep learning-based Patient representation learning methods to further improvements. In particular, learning robust patient representations is challenging when patient data is missing or insufficient. Although data augmentation techniques can tackle this deficiency, the complex data processing further weakens the inexplicability of patient representation learning models. To address the above challenges, this paper proposes an Explainable and Augmented Patient Representation Learning for disease prediction (EAPR). EAPR utilizes data augmentation controlled by confidence interval to enhance patient representation in the presence of limited patient data. Moreover, EAPR proposes to use two-stage gradient backpropagation to address the problem of unexplainable patient representation learning models due to the complex data enhancement process. The experimental results on real clinical data validate the effectiveness and explainability of the proposed approach.

Evolutionary role of chemotherapy in advanced nasopharyngeal carcinoma: a literature-based network meta-analysis.

PURPOSE: The role of chemotherapy has evolved greatly in advanced nasopharyngeal carcinoma (NPC). We undertook this network meta-analysis to establish the optimal chemotherapy strategy in advanced NPC. MATERIALS AND METHODS: This network meta-analysis recruited randomized clinical trials involving patients with advanced NPC randomly allocated to induction chemotherapy plus concurrent chemoradiotherapy (CRT; induction + CRT), CRT plus adjuvant chemotherapy (CRT + adjuvant), CRT or radiotherapy (RT) alone. Pairwise meta-analysis was first conducted, then network meta-analysis was performed using the frequentist approach. Effect size was expressed as HR and 95% CI. RESULTS: In total, we analyzed 15 studies involving 4,067 patients with 880 (21.6%) patients receiving induction + CRT, 897 (22.1%) receiving CRT + adjuvant, 1,421 (34.9%) receiving CRT, and 869 (21.4%) receiving RT alone. Induction + CRT achieved significantly better distant failure-free survival (HR, 0.67; 95% CI, 0.53-0.86) and locoregional failure-free survival (HR, 0.69; 95% CI, 0.54-0.89) than CRT, and CRT + adjuvant achieved better overall survival than CRT (HR, 0.82; 95% CI, 0.67-1.00). However, no significant survival difference was found between the induction + CRT and CRT + adjuvant groups. Additionally, RT alone is always worse than the other three treatments. In terms of P-score, induction + CRT ranked best for distant and locoregional failure-free survival, while CRT + adjuvant ranked best for overall survival. CONCLUSION: Both induction + CRT and CRT + adjuvant were equally effective and feasible choices for patients with advanced NPC.

Multi-Instance Metric Transfer Learning for Genome-Wide Protein Function Prediction.

Multi-Instance (MI) learning has been proven to be effective for the genome-wide protein function prediction problems where each training example is associated with multiple instances. Many studies in this literature attempted to find an appropriate Multi-Instance Learning (MIL) method for genome-wide protein function prediction under a usual assumption, the underlying distribution from testing data (target domain, i.e., TD) is the same as that from training data (source domain, i.e., SD). However, this assumption may be violated in real practice. To tackle this problem, in this paper, we propose a Multi-Instance Metric Transfer Learning (MIMTL) approach for genome-wide protein function prediction. In MIMTL, we first transfer the source domain distribution to the target domain distribution by utilizing the bag weights. Then, we construct a distance metric learning method with the reweighted bags. At last, we develop an alternative optimization scheme for MIMTL. Comprehensive experimental evidence on seven real-world organisms verifies the effectiveness and efficiency of the proposed MIMTL approach over several state-of-the-art methods.