Predicting clinical outcomes such as mortality and length of hospital stay during the initial admission to the intensive care unit (ICU) is vital for enhancing patient survival rates and optimizing hospital resource allocation. Accurate diagnostic information is deeply correlated with mortality and hospitalization duration; however, initial admission data often lack sufficient diagnostic details and may even have discrepancies with discharge diagnoses. Previous studies have leveraged Electronic Health Record (EHR) data, encompassing clinical notes and vital sign monitoring data, to boost the performance of clinical outcomes prediction. Nevertheless, they have often overlooked the relevance between sentence information, resorting to simple averages, and have missed the importance of diagnostic information. To overcome this limitation, we propose the Diagnosis-Enhanced Sentence Attention Model for Clinical Prediction (DESAM-cp), a multimodal neural network. We employed an auxiliary loss to ensure that embeddings of clinical note were informative of the patient's discharge diagnosis information and used sentence-level self-attention to account for the relevance between sentences. Ultimately, clinical note embeddings containing diagnostic information were fused with monitoring data embeddings to predict clinical outcomes. Our experiments demonstrate that the methodology of DESAM-cp outperforms previous research methods. Additionally, through ablation tests, visualization analysis, and attention analysis, we have substantiated that our individual modules function effectively. Our analysis also suggests that our model focuses on medically relevant sentences in relation to diagnosis and clinical outcome predictions and can provide supporting evidence to assist clinicians in making informed decisions regarding diagnoses and treatments. The code for DESAM-cp used in this work is available at https://github.com/a11525/DESAM-cp

OBJECTIVE: Applications of machine learning in healthcare are of high interest and have the potential to improve patient care. Yet, the real-world accuracy of these models in clinical practice and on different patient subpopulations remains unclear. To address these important questions, we hosted a community challenge to evaluate methods that predict healthcare outcomes. We focused on the prediction of all-cause mortality as the community challenge question. MATERIALS AND METHODS: Using a Model-to-Data framework, 345 registered participants, coalescing into 25 independent teams, spread over 3 continents and 10 countries, generated 25 accurate models all trained on a dataset of over 1.1 million patients and evaluated on patients prospectively collected over a 1-year observation of a large health system. RESULTS: The top performing team achieved a final area under the receiver operator curve of 0.947 (95% CI, 0.942-0.951) and an area under the precision-recall curve of 0.487 (95% CI, 0.458-0.499) on a prospectively collected patient cohort. DISCUSSION: Post hoc analysis after the challenge revealed that models differ in accuracy on subpopulations, delineated by race or gender, even when they are trained on the same data. CONCLUSION: This is the largest community challenge focused on the evaluation of state-of-the-art machine learning methods in a healthcare system performed to date, revealing both opportunities and pitfalls of clinical AI.

MOTIVATION: Identifying mechanism of actions (MoA) of novel compounds is crucial in drug discovery. Careful understanding of MoA can avoid potential side effects of drug candidates. Efforts have been made to identify MoA using the transcriptomic signatures induced by compounds. However, these approaches fail to reveal MoAs in the absence of actual compound signatures. RESULTS: We present MoAble, which predicts MoAs without requiring compound signatures. We train a deep learning-based coembedding model to map compound signatures and compound structure into the same embedding space. The model generates low-dimensional compound signature representation from the compound structures. To predict MoAs, pathway enrichment analysis is performed based on the connectivity between embedding vectors of compounds and those of genetic perturbation. Results show that MoAble is comparable to the methods that use actual compound signatures. We demonstrate that MoAble can be used to reveal MoAs of novel compounds without measuring compound signatures with the same prediction accuracy as that with measuring them. AVAILABILITY AND IMPLEMENTATION: MoAble is available at https://github.com/dmis-lab/moable. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Recent advances in deep learning have offered solutions to many biomedical tasks. However, there remains a challenge in applying deep learning to survival analysis using human cancer transcriptome data. As the number of genes, the input variables of survival model, is larger than the amount of available cancer patient samples, deep-learning models are prone to overfitting. To address the issue, we introduce a new deep-learning architecture called VAECox. VAECox uses transfer learning and fine tuning. RESULTS: We pre-trained a variational autoencoder on all RNA-seq data in 20 TCGA datasets and transferred the trained weights to our survival prediction model. Then we fine-tuned the transferred weights during training the survival model on each dataset. Results show that our model outperformed other previous models such as Cox Proportional Hazard with LASSO and ridge penalty and Cox-nnet on the 7 of 10 TCGA datasets in terms of C-index. The results signify that the transferred information obtained from entire cancer transcriptome data helped our survival prediction model reduce overfitting and show robust performance in unseen cancer patient samples. AVAILABILITY AND IMPLEMENTATION: Our implementation of VAECox is available at https://github.com/dmis-lab/VAECox. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Biomedical text mining is becoming increasingly important as the number of biomedical documents rapidly grows. With the progress in natural language processing (NLP), extracting valuable information from biomedical literature has gained popularity among researchers, and deep learning has boosted the development of effective biomedical text mining models. However, directly applying the advancements in NLP to biomedical text mining often yields unsatisfactory results due to a word distribution shift from general domain corpora to biomedical corpora. In this article, we investigate how the recently introduced pre-trained language model BERT can be adapted for biomedical corpora. RESULTS: We introduce BioBERT (Bidirectional Encoder Representations from Transformers for Biomedical Text Mining), which is a domain-specific language representation model pre-trained on large-scale biomedical corpora. With almost the same architecture across tasks, BioBERT largely outperforms BERT and previous state-of-the-art models in a variety of biomedical text mining tasks when pre-trained on biomedical corpora. While BERT obtains performance comparable to that of previous state-of-the-art models, BioBERT significantly outperforms them on the following three representative biomedical text mining tasks: biomedical named entity recognition (0.62% F1 score improvement), biomedical relation extraction (2.80% F1 score improvement) and biomedical question answering (12.24% MRR improvement). Our analysis results show that pre-training BERT on biomedical corpora helps it to understand complex biomedical texts. AVAILABILITY AND IMPLEMENTATION: We make the pre-trained weights of BioBERT freely available at https://github.com/naver/biobert-pretrained, and the source code for fine-tuning BioBERT available at https://github.com/dmis-lab/biobert.

MOTIVATION: Traditional drug discovery approaches identify a target for a disease and find a compound that binds to the target. In this approach, structures of compounds are considered as the most important features because it is assumed that similar structures will bind to the same target. Therefore, structural analogs of the drugs that bind to the target are selected as drug candidates. However, even though compounds are not structural analogs, they may achieve the desired response. A new drug discovery method based on drug response, which can complement the structure-based methods, is needed. RESULTS: We implemented Siamese neural networks called ReSimNet that take as input two chemical compounds and predicts the CMap score of the two compounds, which we use to measure the transcriptional response similarity of the two compounds. ReSimNet learns the embedding vector of a chemical compound in a transcriptional response space. ReSimNet is trained to minimize the difference between the cosine similarity of the embedding vectors of the two compounds and the CMap score of the two compounds. ReSimNet can find pairs of compounds that are similar in response even though they may have dissimilar structures. In our quantitative evaluation, ReSimNet outperformed the baseline machine learning models. The ReSimNet ensemble model achieves a Pearson correlation of 0.518 and a precision@1% of 0.989. In addition, in the qualitative analysis, we tested ReSimNet on the ZINC15 database and showed that ReSimNet successfully identifies chemical compounds that are relevant to a prototype drug whose mechanism of action is known. AVAILABILITY AND IMPLEMENTATION: The source code and the pre-trained weights of ReSimNet are available at https://github.com/dmis-lab/ReSimNet. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: Molecular biomarkers that can predict drug efficacy in cancer patients are crucial components for the advancement of precision medicine. However, identifying these molecular biomarkers remains a laborious and challenging task. Next-generation sequencing of patients and preclinical models have increasingly led to the identification of novel gene-mutation-drug relations, and these results have been reported and published in the scientific literature. RESULTS: Here, we present two new computational methods that utilize all the PubMed articles as domain specific background knowledge to assist in the extraction and curation of gene-mutation-drug relations from the literature. The first method uses the Biomedical Entity Search Tool (BEST) scoring results as some of the features to train the machine learning classifiers. The second method uses not only the BEST scoring results, but also word vectors in a deep convolutional neural network model that are constructed from and trained on numerous documents such as PubMed abstracts and Google News articles. Using the features obtained from both the BEST search engine scores and word vectors, we extract mutation-gene and mutation-drug relations from the literature using machine learning classifiers such as random forest and deep convolutional neural networks. Our methods achieved better results compared with the state-of-the-art methods. We used our proposed features in a simple machine learning model, and obtained F1-scores of 0.96 and 0.82 for mutation-gene and mutation-drug relation classification, respectively. We also developed a deep learning classification model using convolutional neural networks, BEST scores, and the word embeddings that are pre-trained on PubMed or Google News data. Using deep learning, the classification accuracy improved, and F1-scores of 0.96 and 0.86 were obtained for the mutation-gene and mutation-drug relations, respectively. CONCLUSION: We believe that our computational methods described in this research could be used as an important tool in identifying molecular biomarkers that predict drug responses in cancer patients. We also built a database of these mutation-gene-drug relations that were extracted from all the PubMed abstracts. We believe that our database can prove to be a valuable resource for precision medicine researchers.

UNLABELLED: We introduce HiPub, a seamless Chrome browser plug-in that automatically recognizes, annotates and translates biomedical entities from texts into networks for knowledge discovery. Using a combination of two different named-entity recognition resources, HiPub can recognize genes, proteins, diseases, drugs, mutations and cell lines in texts, and achieve high precision and recall. HiPub extracts biomedical entity-relationships from texts to construct context-specific networks, and integrates existing network data from external databases for knowledge discovery. It allows users to add additional entities from related articles, as well as user-defined entities for discovering new and unexpected entity-relationships. HiPub provides functional enrichment analysis on the biomedical entity network, and link-outs to external resources to assist users in learning new entities and relations. AVAILABILITY AND IMPLEMENTATION: HiPub and detailed user guide are available at http://hipub.korea.ac.kr CONTACT: [email protected], [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.