Molecular and physiological changes in the SpaceX Inspiration4 civilian crew.

Human spaceflight has historically been managed by government agencies, such as in the NASA Twins Study1, but new commercial spaceflight opportunities have opened spaceflight to a broader population. In 2021, the SpaceX Inspiration4 mission launched the first all-civilian crew to low Earth orbit, which included the youngest American astronaut (aged 29), new in-flight experimental technologies (handheld ultrasound imaging, smartwatch wearables and immune profiling), ocular alignment measurements and new protocols for in-depth, multi-omic molecular and cellular profiling. Here we report the primary findings from the 3-day spaceflight mission, which induced a broad range of physiological and stress responses, neurovestibular changes indexed by ocular misalignment, and altered neurocognitive functioning, some of which match those of long-term spaceflight2, but almost all of which did not differ from baseline (pre-flight) after return to Earth. Overall, these preliminary civilian spaceflight data suggest that short-duration missions do not pose a significant health risk, and moreover present a rich opportunity to measure the earliest phases of adaptation to spaceflight in the human body at anatomical, cellular, physiological and cognitive levels. Finally, these methods and results lay the foundation for an open, rapidly expanding biomedical database for astronauts3, which can inform countermeasure development for both private and government-sponsored space missions.

The Space Omics and Medical Atlas (SOMA) and international astronaut biobank.

Spaceflight induces molecular, cellular and physiological shifts in astronauts and poses myriad biomedical challenges to the human body, which are becoming increasingly relevant as more humans venture into space1-6. Yet current frameworks for aerospace medicine are nascent and lag far behind advancements in precision medicine on Earth, underscoring the need for rapid development of space medicine databases, tools and protocols. Here we present the Space Omics and Medical Atlas (SOMA), an integrated data and sample repository for clinical, cellular and multi-omic research profiles from a diverse range of missions, including the NASA Twins Study7, JAXA CFE study8,9, SpaceX Inspiration4 crew10-12, Axiom and Polaris. The SOMA resource represents a more than tenfold increase in publicly available human space omics data, with matched samples available from the Cornell Aerospace Medicine Biobank. The Atlas includes extensive molecular and physiological profiles encompassing genomics, epigenomics, transcriptomics, proteomics, metabolomics and microbiome datasets, which reveal some consistent features across missions, including cytokine shifts, telomere elongation and gene expression changes, as well as mission-specific molecular responses and links to orthologous, tissue-specific mouse datasets. Leveraging the datasets, tools and resources in SOMA can help to accelerate precision aerospace medicine, bringing needed health monitoring, risk mitigation and countermeasure data for upcoming lunar, Mars and exploration-class missions.

A network-theoretic approach for decompositional translation across Open Biological Ontologies.

Biological ontologies are now being widely used for annotation, sharing and retrieval of the biological data. Many of these ontologies are hosted under the umbrella of the Open Biological Ontologies Foundry. In order to support interterminology mapping, composite terms in these ontologies need to be translated into atomic or primitive terms in other, orthogonal ontologies, for example, gluconeogenesis (biological process term) to glucose (chemical ontology term). Identifying such decompositional ontology translations is a challenging problem. In this paper, we propose a network-theoretic approach based on the structure of the integrated OBO relationship graph. We use a network-theoretic measure, called the clustering coefficient, to find relevant atomic terms in the neighborhood of a composite term. By eliminating the existing GO to ChEBI Ontology mappings from OBO, we evaluate whether the proposed approach can re-identify the corresponding relationships. The results indicate that the network structure provides strong cues for decompositional ontology translation and the existing relationships can be used to identify new translations.

Using semantic and structural properties of the Unified Medical Language System to discover potential terminological relationships.

OBJECTIVE: To use the semantic and structural properties in the Unified Medical Language System (UMLS) Metathesaurus to characterize and discover potential relationships. DESIGN: The UMLS integrates knowledge from several biomedical terminologies. This knowledge can be used to discover implicit semantic relationships between concepts. In this paper, the authors propose a problem-independent approach for discovering potential terminological relationships that employs semantic abstraction of indirect relationship paths to perform classification and analysis of network theoretical measures such as topological overlap, preferential attachment, graph partitioning, and number of indirect paths. Using different versions of the UMLS, the authors evaluate the proposed approach's ability to predict newly added relationships. MEASUREMENTS: Classification accuracy, precision-recall. RESULTS: Strong discriminative characteristics were observed with a semantic abstraction based classifier (classification accuracy of 91%), the average number of indirect paths, preferential attachment, and graph partitioning to identify potential relationships. The proposed relationship prediction algorithm resulted in 56% recall in top 10 results for new relationships added to subsequent versions of the UMLS between 2005 and 2007. CONCLUSIONS: The UMLS has sufficient knowledge to enable discovery of potential terminological relationships.

A scale-free network view of the UMLS to learn terminology translations.

The UMLS Metathesaurus belongs to the class of scale-free networks with few concept hubs possessing a large number of relationships. The hubs provide useful links between the concepts from disparate terminologies in the UMLS; however, they also exponentially increase the number of possible transitive cross-terminology paths. Towards the goal of using machine learning to rank cross-terminology translations, we propose a traversal algorithm that exploits the scale-free property of the UMLS to reduce the number of candidate translations. We characterize the concept hubs into "informational" and "noisy" concept hubs and provide an automated method to detect them. Using gold standard mappings from SNOMED-CT to ICD9CM, we found an average 20-fold reduction in the number of candidate mappings while achieving comparable recall and ranking results. A hub-driven traversal strategy provides a promising approach to generate high quality cross-terminology translations from the UMLS.

Trusted and relevant medical knowledge: the promise of information retrieval in biomedicine.

As the world of medicine becomes increasingly digitized, the Web has become a de facto resource for physicians to quickly glean pertinent clinical information to carry out diagnostic and therapeutic decisions. At present, physicians face the dual challenge of judging the relevance of the information and trusting its Web source. This paper proposes a trust-relevance framework for conceptualizing computer-accessed medical information resources, a set of criteria for evaluating these information resources, and descriptions of a sample of available online resources. It also presents a usable framework for evaluating information retrieval innovations and explains the different capabilities of representative information retrieval tools and applications. By demystifying the concepts associated with information resources, search engines, and retrieval tools, and presenting a reasonable view of current opportunities as well as future possibilities, the authors hope to provide guidance so physicians can more rapidly adopt innovative computer-assisted search tools for acquiring information that facilitate patient care decision-making.

What do patients search for when seeking clinical trial information online?

The Internet has become a common source for consumers to seek health information across a wide range of topics including searching for clinical trials. However, not much is known about what consumers search for in relation to clinical trials and how they formulate their search queries. In this study, we use log file data from TrialX.com, a consumer-centric website that provides clinical trial information to ascertain patterns in consumer queries. We analyzed semantic patterns in the queries by mapping query keywords to the UMLS Semantic Types and performed a manual evaluation of user paths. We found that the queries can be grouped into combinations of information needs related to condition, location and treatment. The results also suggested that the consumers using longer search queries with multiple Semantic Types are more likely to take action to participate in clinical trials. The study provides early insights that can be used to inform changes in website content and information display to improve clinical trials information seeking.

ECRL: an eligibility criteria representation language based on the UMLS Semantic Network.

We propose a formal representation language to represent, share and reuse eligibility criteria in clinical research protocols towards the goal of automated eligibility identification. The language is an extension over the UMLS Semantic Network and can be transformed into other computable representations.

Comparing ICD9-encoded diagnoses and NLP-processed discharge summaries for clinical trials pre-screening: a case study.

The prevalence of electronic medical record (EMR) systems has made mass-screening for clinical trials viable through secondary uses of clinical data, which often exist in both structured and free text formats. The tradeoffs of using information in either data format for clinical trials screening are understudied. This paper compares the results of clinical trial eligibility queries over ICD9-encoded diagnoses and NLP-processed textual discharge summaries. The strengths and weaknesses of both data sources are summarized along the following dimensions: information completeness, expressiveness, code granularity, and accuracy of temporal information. We conclude that NLP-processed patient reports supplement important information for eligibility screening and should be used in combination with structured data.

Decompositional terminology translation using network analysis.

Biomedical terminologies often contain composite concepts that cannot be translated into single unique synonymous concepts in a target controlled terminology. Such composite concepts need to be decomposed into sets of component concepts present in the target terminology that can serve as the proxy for applications in information retrieval, decision support or data analysis. Towards this goal, we use a "clustering coefficient" over the UMLS Metathesaurus to traverse the closely clustered neighbors of the composite source concept to generate a ranked list of possible component concepts. Using the MeSH Associated Expression mappings as the gold-standard, we show that the proposed approach generates relevant component concepts as compared to existing semantic locality based methods. The topological connectivity of the concepts in the UMLS Metathesaurus is a useful feature that can be coupled with existing lexical and semantic locality based approaches towards terminology translation.

Semantic query generation from eligibility criteria in clinical trials.

Towards the goal of automated eligibility determination for clinical trials from electronic health records, we propose a method to formulate Semantic Web based queries using the free-text eligibility criteria on clinicaltrials.gov.

Mining cross-terminology links in the UMLS.

OBJECTIVE: To explore link mining approaches over transitive relationship paths in the Unified Medical Language System (UMLS). The goal is to classify relevant and 'interesting' cross-terminology links/paths for integration of Electronic Health Records (EHRs) and information resources. METHODS: We present approaches for using the link semantics as learning features, sampling the UMLS to create training examples, and ranking the classified links. We use the clinical query and MEDLINE pairs in the OHSUMED dataset to extract 'gold-links' between SNOMED-CT and MeSH respectively, and compare them against corresponding two-step transitive links generated from the UMLS. RESULTS: a). 75.7% increase in reachable MeSH concepts with two-step links as compared to direct one-step links b). 94.08% recall after link classification. CONCLUSION: Using link mining with the UMLS is a promising approach for inter-terminology translation; further research is needed to handle the exponential link growth.

GODSN: Global News Driven Disease Outbreak and Surveillance.

Disease surveillance has evolved dramatically in the last few years, becoming more real-time, comprehensive and technology driven. As the Internet grows and evolves as a powerful information medium, the ability to mine real-time news feeds for disease surveillance has become viable. We propose a system, GODSN that monitors global news for disease outbreaks and surveillance. The system processes real-time news feeds using natural language processing to obtain disease information and the geographical reference to plot them on a geographic information system. GODSN provides an effective approach to visualize the spatial and temporal trends of infectious disease outbreaks or disease specific developments.