Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records.

OBJECTIVE: The accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, covariate imbalance and delayed diagnosis when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. The goal of this study is to investigate the extent to which the aforementioned issues influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. METHODS: We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. RESULTS: We developed a novel backward masking scheme to deal with the issue of delayed diagnosis which is very common in EHR data analysis and evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. CONCLUSIONS: The strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.

Deep representation learning of patient data from Electronic Health Records (EHR): A systematic review.

OBJECTIVES: Patient representation learning refers to learning a dense mathematical representation of a patient that encodes meaningful information from Electronic Health Records (EHRs). This is generally performed using advanced deep learning methods. This study presents a systematic review of this field and provides both qualitative and quantitative analyses from a methodological perspective. METHODS: We identified studies developing patient representations from EHRs with deep learning methods from MEDLINE, EMBASE, Scopus, the Association for Computing Machinery (ACM) Digital Library, and the Institute of Electrical and Electronics Engineers (IEEE) Xplore Digital Library. After screening 363 articles, 49 papers were included for a comprehensive data collection. RESULTS: Publications developing patient representations almost doubled each year from 2015 until 2019. We noticed a typical workflow starting with feeding raw data, applying deep learning models, and ending with clinical outcome predictions as evaluations of the learned representations. Specifically, learning representations from structured EHR data was dominant (37 out of 49 studies). Recurrent Neural Networks were widely applied as the deep learning architecture (Long short-term memory: 13 studies, Gated recurrent unit: 11 studies). Learning was mainly performed in a supervised manner (30 studies) optimized with cross-entropy loss. Disease prediction was the most common application and evaluation (31 studies). Benchmark datasets were mostly unavailable (28 studies) due to privacy concerns of EHR data, and code availability was assured in 20 studies. DISCUSSION & CONCLUSION: The existing predictive models mainly focus on the prediction of single diseases, rather than considering the complex mechanisms of patients from a holistic review. We show the importance and feasibility of learning comprehensive representations of patient EHR data through a systematic review. Advances in patient representation learning techniques will be essential for powering patient-level EHR analyses. Future work will still be devoted to leveraging the richness and potential of available EHR data. Reproducibility and transparency of reported results will hopefully improve. Knowledge distillation and advanced learning techniques will be exploited to assist the capability of learning patient representation further.

Gene fingerprint model for literature based detection of the associations among complex diseases: a case study of COPD.

BACKGROUND: Disease comorbidity is very common and has significant impact on disease treatment. Revealing the associations among diseases may help to understand the mechanisms of diseases, improve the prevention and treatment of diseases, and support the discovery of new drugs or new uses of existing drugs. METHODS: In this paper, we introduced a mathematical model to represent gene related diseases with a series of associated genes based on the overrepresentation of genes and diseases in PubMed literature. We also illustrated an efficient way to reveal the implicit connections between COPD and other diseases based on this model. RESULTS: We applied this approach to analyze the relationships between Chronic Obstructive Pulmonary Disease (COPD) and other diseases under the Lung diseases branch in the Medical subject heading index system and detected 4 novel diseases relevant to COPD. As judged by domain experts, the F score of our approach is up to 77.6%. CONCLUSIONS: The results demonstrate the effectiveness of the gene fingerprint model for diseases on the basis of medical literature.

The International Conference on Intelligent Biology and Medicine (ICIBM) 2016: summary and innovation in genomics.

In this editorial, we first summarize the 2016 International Conference on Intelligent Biology and Medicine (ICIBM 2016) that was held on December 8-10, 2016 in Houston, Texas, USA, and then briefly introduce the ten research articles included in this supplement issue. ICIBM 2016 included four workshops or tutorials, four keynote lectures, four conference invited talks, eight concurrent scientific sessions and a poster session for 53 accepted abstracts, covering current topics in bioinformatics, systems biology, intelligent computing, and biomedical informatics. Through our call for papers, a total of 77 original manuscripts were submitted to ICIBM 2016. After peer review, 11 articles were selected in this special issue, covering topics such as single cell RNA-seq analysis method, genome sequence and variation analysis, bioinformatics method for vaccine development, and cancer genomics.

Using Ontology Fingerprints to disambiguate gene name entities in the biomedical literature.

Ambiguous gene names in the biomedical literature are a barrier to accurate information extraction. To overcome this hurdle, we generated Ontology Fingerprints for selected genes that are relevant for personalized cancer therapy. These Ontology Fingerprints were used to evaluate the association between genes and biomedical literature to disambiguate gene names. We obtained 93.6% precision for the test gene set and 80.4% for the area under a receiver-operating characteristics curve for gene and article association. The core algorithm was implemented using a graphics processing unit-based MapReduce framework to handle big data and to improve performance. We conclude that Ontology Fingerprints can help disambiguate gene names mentioned in text and analyse the association between genes and articles. Database URL: http://www.ontologyfingerprint.org