Semantic Deep Learning: Prior Knowledge and a Type of Four-Term Embedding Analogy to Acquire Treatments for Well-Known Diseases.

BACKGROUND: How to treat a disease remains to be the most common type of clinical question. Obtaining evidence-based answers from biomedical literature is difficult. Analogical reasoning with embeddings from deep learning (embedding analogies) may extract such biomedical facts, although the state-of-the-art focuses on pair-based proportional (pairwise) analogies such as man:woman::king:queen ("queen = -man +king +woman"). OBJECTIVE: This study aimed to systematically extract disease treatment statements with a Semantic Deep Learning (SemDeep) approach underpinned by prior knowledge and another type of 4-term analogy (other than pairwise). METHODS: As preliminaries, we investigated Continuous Bag-of-Words (CBOW) embedding analogies in a common-English corpus with five lines of text and observed a type of 4-term analogy (not pairwise) applying the 3CosAdd formula and relating the semantic fields person and death: "dagger = -Romeo +die +died" (search query: -Romeo +die +died). Our SemDeep approach worked with pre-existing items of knowledge (what is known) to make inferences sanctioned by a 4-term analogy (search query -x +z1 +z2) from CBOW and Skip-gram embeddings created with a PubMed systematic reviews subset (PMSB dataset). Stage1: Knowledge acquisition. Obtaining a set of terms, candidate y, from embeddings using vector arithmetic. Some n-gram pairs from the cosine and validated with evidence (prior knowledge) are the input for the 3cosAdd, seeking a type of 4-term analogy relating the semantic fields disease and treatment. Stage 2: Knowledge organization. Identification of candidates sanctioned by the analogy belonging to the semantic field treatment and mapping these candidates to unified medical language system Metathesaurus concepts with MetaMap. A concept pair is a brief disease treatment statement (biomedical fact). Stage 3: Knowledge validation. An evidence-based evaluation followed by human validation of biomedical facts potentially useful for clinicians. RESULTS: We obtained 5352 n-gram pairs from 446 search queries by applying the 3CosAdd. The microaveraging performance of MetaMap for candidate y belonging to the semantic field treatment was F-measure=80.00% (precision=77.00%, recall=83.25%). We developed an empirical heuristic with some predictive power for clinical winners, that is, search queries bringing candidate y with evidence of a therapeutic intent for target disease x. The search queries -asthma +inhaled_corticosteroids +inhaled_corticosteroid and -epilepsy +valproate +antiepileptic_drug were clinical winners, finding eight evidence-based beneficial treatments. CONCLUSIONS: Extracting treatments with therapeutic intent by analogical reasoning from embeddings (423K n-grams from the PMSB dataset) is an ambitious goal. Our SemDeep approach is knowledge-based, underpinned by embedding analogies that exploit prior knowledge. Biomedical facts from embedding analogies (4-term type, not pairwise) are potentially useful for clinicians. The heuristic offers a practical way to discover beneficial treatments for well-known diseases. Learning from deep learning models does not require a massive amount of data. Embedding analogies are not limited to pairwise analogies; hence, analogical reasoning with embeddings is underexploited.

Mobilising computable biomedical knowledge: challenges for clinical decision support from a medical knowledge provider.

INTRODUCTION: This paper summarises a talk given at the first UK workshop on mobilising computable biomedical knowledge on 29 October 2019 in London. It examines challenges in mobilising computable biomedical knowledge for clinical decision support from the perspective of a medical knowledge provider. METHODS: We developed the themes outlined below after personally reflecting on the challenges that we have encountered in this field and after considering the barriers that knowledge providers face in ensuring that their content is accessed and used by healthcare professionals. We further developed the themes after discussing them with delegates at the workshop and listening to their feedback. DISCUSSION: There are many challenges in mobilising computable knowledge for clinical decision support from the perspective of a medical knowledge provider. These include the size of the task at hand, the challenge of creating machine interpretable content, the issue of standards, the need to do better in tracing how computable medical knowledge that is part of clinical decision support impacts patient outcomes, the challenge of comorbidities, the problem of adhering to safety standards and finally the challenge of integrating knowledge with problem solving and procedural skills, healthy attitudes and professional behaviours. Partnership is likely to be essential if we are to make progress in this field. The problems are too complex and interrelated to be solved by any one institution alone.

Exploring semantic deep learning for building reliable and reusable one health knowledge from PubMed systematic reviews and veterinary clinical notes.

BACKGROUND: Deep Learning opens up opportunities for routinely scanning large bodies of biomedical literature and clinical narratives to represent the meaning of biomedical and clinical terms. However, the validation and integration of this knowledge on a scale requires cross checking with ground truths (i.e. evidence-based resources) that are unavailable in an actionable or computable form. In this paper we explore how to turn information about diagnoses, prognoses, therapies and other clinical concepts into computable knowledge using free-text data about human and animal health. We used a Semantic Deep Learning approach that combines the Semantic Web technologies and Deep Learning to acquire and validate knowledge about 11 well-known medical conditions mined from two sets of unstructured free-text data: 300 K PubMed Systematic Review articles (the PMSB dataset) and 2.5 M veterinary clinical notes (the VetCN dataset). For each target condition we obtained 20 related clinical concepts using two deep learning methods applied separately on the two datasets, resulting in 880 term pairs (target term, candidate term). Each concept, represented by an n-gram, is mapped to UMLS using MetaMap; we also developed a bespoke method for mapping short forms (e.g. abbreviations and acronyms). Existing ontologies were used to formally represent associations. We also create ontological modules and illustrate how the extracted knowledge can be queried. The evaluation was performed using the content within BMJ Best Practice. RESULTS: MetaMap achieves an F measure of 88% (precision 85%, recall 91%) when applied directly to the total of 613 unique candidate terms for the 880 term pairs. When the processing of short forms is included, MetaMap achieves an F measure of 94% (precision 92%, recall 96%). Validation of the term pairs with BMJ Best Practice yields precision between 98 and 99%. CONCLUSIONS: The Semantic Deep Learning approach can transform neural embeddings built from unstructured free-text data into reliable and reusable One Health knowledge using ontologies and content from BMJ Best Practice.

The Montagues and the Capulets.

Two households, both alike in dignity, In fair Genomics, where we lay our scene, (One, comforted by its logic's rigour, Claims ontology for the realm of pure, The other, with blessed scientist's vigour, Acts hastily on models that endure), From ancient grudge break to new mutiny, When 'being' drives a fly-man to blaspheme. From forth the fatal loins of these two foes, Researchers to unlock the book of life; Whole misadventured piteous overthrows, Can with their work bury their clans' strife. The fruitful passage of their GO-mark'd love, And the continuance of their studies sage, Which, united, yield ontologies undreamed-of, Is now the hour's traffic of our stage; The which if you with patient ears attend, What here shall miss, our toil shall strive to mend.

Building ontologies in DAML + OIL.

In this article we describe an approach to representing and building ontologies advocated by the Bioinformatics and Medical Informatics groups at the University of Manchester. The hand-crafting of ontologies offers an easy and rapid avenue to delivering ontologies. Experience has shown that such approaches are unsustainable. Description logic approaches have been shown to offer computational support for building sound, complete and logically consistent ontologies. A new knowledge representation language, DAML + OIL, offers a new standard that is able to support many styles of ontology, from hand-crafted to full logic-based descriptions with reasoning support. We describe this language, the OilEd editing tool, reasoning support and a strategy for the language's use. We finish with a current example, in the Gene Ontology Next Generation (GONG) project, that uses DAML + OIL as the basis for moving the Gene Ontology from its current hand-crafted, form to one that uses logical descriptions of a concept's properties to deliver a more complete version of the ontology.

Scale and context: issues in ontologies to link health- and bio-informatics.

Bridging levels of scale and context are key problems for integrating Bio- and Health Informatics. Formal, logic-based ontologies using expressive formalisms are naturally "fractal" and provide new methods to support these aims. The basic notion of composition can be used to bridge scales; axioms can be used to carry implicit information; specific context markers can be included in definitions; and a hierarchy of semantic links can be used to represent subtle differences in point of view. Experience with OpenGALEN, the UK Drug Ontology and new experiments with the Gene Ontology and Foundational Model of Anatomy suggest that these are powerful tools provide practical solutions.