KDD18: Showcasing Ways Machine Learning is Improving Healthcare

April 25, 2018
Fred Rahmanian, Chief Technology Officer


Banner image of a doctor pointing to icons.

You’re invited to join me at the annual conference of the Association of Computing Machinery’s Special Interest Group for Knowledge Discovery and Data Mining (KDD), the world’s oldest and largest community for data mining, data science and analytics and the conference at which many of the new data science advancements are published. It’s the ultimate Geekfest, one I very much look forward to attending all year for the opportunity to reconnect with old friends and network with young data scientists, but also for the energy and enthusiasm about new applications and innovations.

KDD18 will be held in London, August 19-23. Even more exciting to me than the location is this year’s focus on healthcare. The goal of KDD18’s Heath Day at KDD18 is to bring together researchers from machine learning, computational linguistics, medical informatics and large healthcare systems to bridge the theory of machine learning, natural language processing, data mining and the needs of the healthcare community. 

Undoubtedly, it’s an exciting time to work in data science and healthcare. The proliferation of new data sources, including clinical, genomic, financial, social determinants of health (SDoH) and Internet of Things (IoT), coupled with improvements in computing power and better algorithms, offers the very real possibility of addressing healthcare’s most confounding issues. 

At Geneia, we’re using artificial intelligence throughout our Theon® analytics, insights and care management platform in the form of risk factors, risk numbers, propensity to engage in clinical programs, and propensity to enroll in health insurance plans. Data generated by AI is used to produce suspect and missing hierarchical condition category (HCC) codes, among other things. 

Image of a doctor using the Theon platform on a tablet. 

Other potential uses for AI include:

  • Patient stratification - Can we train an AI to identify patients before they become high risk or chronic?
  • Treatment variation – Can we train an AI to identify variation in treatment using just claims data?
  • 30-day hospital readmission – Can we train an AI to predict the probability a patient will be readmitted for any reason within 30 days of discharge?
  • Medication adherence – Using only prescription claims, can we train an AI to identify:
    1. The mode of engagement that will improve adherence for each patient
    2. Patients who will stop taking their medications
  • Patient engagement – Can we train an AI to identify who responds best to a particular mode of engagement or care management?

Calling for Papers Using Machine Learning in Precision Medicine & Healthcare Informatics

At this juncture, the possibilities seem endless. Perhaps that’s why I’m so excited about the KDD18 Workshop on Machine Learning for Medicine and Healthcare that I am helping to organize. 

Currently, we’re calling for papers on the application of machine learning for precision medicine and healthcare informatics. Topics may include, but are not limited to:

  • Data standards for translational medicine informatics
  • Analysis of large-scale electronic health records (EHR) or patient-generated health data records
  • Visualization of complex and dynamic biomedical networks
  • Disease subtype discovery for precision medicine
  • Interpretable machine learning for biomedicine and healthcare
  • Deep learning for biomedicine

Submit papers, which can be full or in-progress, to: https://easychair.org/account/signin.cgi?key=69144636.53eUTmy6lR4QDlvo

Papers must be formatted according to the new Standard ACM Conference Proceedings Template, be formatted as a PDF, and a maximum of four pages, including references. The best student paper will be given a $1,000 travel grant.

Act now. The deadline for abstract submission is May 8. 


Related Blogs