Artificial Intelligence in Medicine (AIM)


Note, I am just starting this discussion ... and this page is only one step in a million or more steps ... so 'hold your fire' if you (correctly) believe that I hardly touched on anything.  I haven't, ... there's a lot of work to do, many people are working it. MY aim to add different and additional perspectives. 

----


AI in Medicine (AIM) approaches and applications have assisted in both trivial and profound ways, and they hold great promise. We argue that there are even larger systemic benefits when AI enabled medicine is considered at a national level.

This page aims to present relevant information and links to resources useful in furthering AIM objectives.  Some links point to reports,preprints,  papers, and books, other point to active and inactive databases, still others point software repositories and AIM specific software and platforms.  While some of the links point to completed and/or terminated projects, we believe there's much to be learned from the linked resources, and we hope these are used to spark curiosity and further ideas and progress in the spirit of "on the shoulders of Giants".


AIM Related Projects


IBM WATSON / WatsonPaths: IBM's Watson architecture was and is being employed in Medical applications.
WatsonPaths: Scenario-Based Question Answering and Inference over Unstructured Information is a key paper available here. As an illustration, the paper discusses a Patient with Erythropoietin Deficiency. Via the query “A 32-year-old woman with type 1 diabetes mellitus has had progressive renal failure... Her hemoglobin concentration is 9 g/dL... A blood smear shows normochromic, normocytic cells. What is the problem?


Since Watson for Medicine is a major platform, we provide additional discussion and links here.


Hetnets in biomedicine - Hetnets — short for heterogeneous networks with multiple node or relationship types; Useful for data integration, translation, and  biomedical knowledge mining.

Project Rephetio (Drug Repurposing) developed to predict new uses for existing compounds. Utilizes edge prediction method and systematic model of drug efficacy, to predict the probability of treatment between over 1500 approved small molecule compounds and over 130 complex diseases. Documentation states the "This study describes Project Rephetio to systematically model drug efficacy based on 755 existing treatments. First, we constructed Hetionet (neo4j.het.io), an integrative network encoding knowledge from millions of biomedical studies. Hetionet v1.0 consists of 47,031 nodes of 11 types and 2,250,197 relationships of 24 types. Data were integrated from 29 public resources to connect compounds, diseases, genes, anatomies, pathways, biological processes, molecular functions, cellular components, pharmacologic classes, side effects, and symptoms. Next, we identified network patterns that distinguish treatments from non-treatments. Then, we predicted the probability of treatment for 209,168 compound–disease pairs (het.io/repurpose)."

there's a related Disease-Associated Gene Browser.


Ontologies

BioPortal, claims to be "the world's most comprehensive repository of biomedical ontologies". It states there are 779 ontologies in its repository (as of 06/06/19). These contain 9,531,435 classes and 144,789,582,932 Direct Plus Expanded Annotations. Happy Hunting :-)  In the meanwhile, here are some more manageable pieces ...  of course we're going to visit and revisit MeSH (Medical Subject Headings (MeSH);National Library of Medicine -NLM). NLM reports that The MeSH Browser now displays 2019 MeSH and 2018 MeSH vocabularies and that all the 2019 MeSH files are  vailable via FTP download . Nice!

OBI - Ontology for Biomedical Investigations: Described in Bandrowski, A., Brinkman, R., Brochhausen, M., Brush, M. H., Bug, B., Chibucos, M. C., ... & Fan, L. (2016). The ontology for biomedical investigations. PloS one, 11(4), e0154556. -- This paper states "OBI has made this possible by creating terms for entities involved in biological and medical investigations and by importing parts of other biomedical ontologies such as GO, Chemical Entities of Biological Interest (ChEBI) and Phenotype Attribute and Trait Ontology (PATO) without altering their meaning. OBI is being used in a wide range of projects covering genomics, multi-omics, immunology, and catalogs of services" At the time of its publication the paper stated "OBI has created 2366 classes and 40 relations along with textual and formal definitions. The OBI Consortium maintains a web resource (http://obi-ontology.org) providing details on the people, policies, and issues being addressed in association with OBI. The current release of OBI is available at http://purl.obolibrary.org/obo/obi.owl.

OGMS -   The Ontology for General Medical Science - based on the papers Toward an Ontological Treatment of Disease and Diagnosis and On Carcinomas and Other Pathological Entities. The ontology attempts to address some of the issues raised at the Workshop on Ontology of Diseases (Dallas, TX) and the Signs, Symptoms, and Findings Workshop(Milan, Italy). OGMS was formerly called the clinical phenotype ontology. Terms from OGMS hang from the Basic Formal Ontology. See http://ontology.buffalo.edu/medo/Disease_and_Diagnosis.pdf


PATENT  EXPLORATiON EXPERIMENTS (PATEX)


to check reality, I am exploring recent  US Patents Granted (by USPTO) that include "Artificial Intelligence" and Medicine in their discussion.

PATEXs results are discussed at:

PATEX-1: AIM Patent Search Experiment #1

PATEX-2: The Search for Deep Learning Impact


the resuts at this stage are preliminary. As time permits, more insights will be added.

more  more more
to be added












current initiatives    sitemap    about    tools     contact   home

under continuing construction


(c) ,2019 ARCHSIX.COM  (for ARCHSIX CREATED MATERIAL)