1 / 21

What Makes Healthcare Data Science so Hard & Interesting - Data Science Pop-up Seattle

For more information on Data science and healthcare, Check out our website on www.dezyre.com

DeZyre
Download Presentation

What Makes Healthcare Data Science so Hard & Interesting - Data Science Pop-up Seattle

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. What Makes Healthcare Data Science so Hard & Interesting David Talby SVP Engineering, Atigeo davidtalby antigeo #datapopupseattle

  2. UNSTRUCTURED Data Science POP-UP in Seattle D Produced by Domino Data Lab Domino’s enterprise data science platform is used by leading analytical organizations to increase productivity, enable collaboration, and publish models into production faster. www.dominodatalab.com #datapopupseattle

  3. WHAT&MAKES&HEALTHCARE&DATA&SCIENCE
 SO&HARD&&&INTERESTING David&Talby& SVP&Engineering,&Atigeo& @davidtalby

  4. “TECHNOLOGY&WILL&REPLACE&80%&OF&WHAT&DOCTORS&DO”&
 
 VINOD&KHOSLA,&2012 2 2

  5. “as&many&as&40,500&patients&die&each&year&in&an&ICU&in&
 the&U.S.&due&to&misdiagnosis”&Winters&et&al.,&2012,&John&Hopkins&& “Combining&estimates&from&3&studies&yielded&a&rate&of& outpatient&diagnostic&errors&of&5.08%,&or&12&million&US& adults&every&year.”&Singhet&al.,&2014,&VA&Medical&Center Root$cause:$being$human Premature(closure Overconfidence Snap(judgment Fatigue Team&dynamics Prejudice 3

  6. THE&PROMISE Real&time&monitoring Always&up&to&date&with&science Large&sample&size Works&24x7& No&trip,&no&waiting Cheaper More&accurate More&objective 4

  7. Big Challenge #1 5

  8. HUMAN&NUANCES Can&you&distinguish&between&real&smiles&of&happiness&and&fake&smiles& trying&to&mask&frustration? “The&algorithm&was&able& to&identify&the&fake&smiles& 92%&of&the&time.&
 Humans,&on&the&other& hand,&performed&no&better& than&chance.”& MIT,&2012 6

  9. MENTAL&HEALTH “Algorithms&correctly& predicted&which&atcrisk& youth&would&go&on&to& develop&psychosis&over&a& 2.5cyear&period&with&
 100%&accuracy.”& Bedi&et&al.,&Nature'Schizophrenia,'2015 7

  10. SAMPLE&HYBRID&ANALYTICS&PIPELINE Direct&&&ambient&Feedback Imagery,&
 drugs,&labs,&… Train&&&test&ensembles NLP&Features Freectext& clinical&notes Train&&&test&Classifiers Graph&Features Relationships& &&ontologies& Time&Series& Features Sensors&&& wearables 8

  11. THE&OPEN&PROBLEM:&EXPLAINABILITY @DavidJBianco,&http://www2.mlsecproject.org/blog/oncexplainabilitycincmachineclearning 9

  12. Big Challenge #2 10

  13. THE&MOMENT&YOU&PUT&A&MODEL&IN&PRODUCTION,&
 IT&STARTS&DEGRADING Never&Changing Always&Changing Political$&$ Economic$Models Banking$&$
 eCommerce$fraud& Cyber$Security Natural$Language,$ Social$Behavior$ Models Google/Amazon& Search$models Physical$models:& Face$recognition& Voice$recognition$ Climate$models Online$Social$ Networking$Models/ Rules$ Automated$trading& RealAtime$ad$bidding [Gunjan&Gupta,&Atigeo,&2014]

  14. SO&PUT&THE&RIGHT&MACHINERY&IN&PLACE 100%&Offcline 100%&Online Automated$ ‘challenger’$online$ evaluation$&$ deployment Traditional
 Scientific$Method:
 Test$a$Hypothesis HandAcrafted$ machine$ learned$ models Daily/weekly$ batch$retraining RealAtime$online$ learning$via$ passive$feedback Hard$Crafted$Rules Active$learning$via& Active$feedback Automated$ensemble,$ boosting$&$feature$ selection$techniques

  15. STATE&OF&THE&PRACTICE&IN&HEALTHCARE 100%&Offcline 100%&Online Automated$ ‘challenger’$online$ evaluation$&$ deployment Traditional
 Scientific$Method:
 Test$a$Hypothesis HandAcrafted$ machine$ learned$ models Daily/weekly$ batch$retraining RealAtime$online$ learning$via$ passive$feedback Hard$Crafted$Rules Active$learning$via& Active$feedback Automated$ensemble,$ boosting$&$feature$ selection$techniques

  16. THE&OPEN&PROBLEM:&MODEL&EVALUATION Evaluate&models&that&are:& • Personalized& • Localized& • Evolve&over&time& • Regulatory&acceptable& ?,'? 14

  17. Big Challenge #3 15

  18. @datapopup #datapopupseattle #datapopupseattle

  19. Thank You To Our Sponsors #datapopupseattle

  20. @Atigeo& @davidtalby 16 ‹ #›

  21. ©&2015&Atigeo,&Corporation.&All&rights&reserved.&&Atigeo&and&the&xPatterns&logo&are&trademarks&of&Atigeo.&The&information&herein&is&for&informational&purposes&only&and&represents&the&current&view&of&Atigeo&as&of&the&date&of&this&presentation.&&Because&Atigeo&must&©&2015&Atigeo,&Corporation.&All&rights&reserved.&&Atigeo&and&the&xPatterns&logo&are&trademarks&of&Atigeo.&The&information&herein&is&for&informational&purposes&only&and&represents&the&current&view&of&Atigeo&as&of&the&date&of&this&presentation.&&Because&Atigeo&must& respond&to&changing&market&conditions,&it&should&not&be&interpreted&to&be&a&commitment&on&the&part&of&Atigeo,&and&Atigeo&cannot&guarantee&the&accuracy&of&any&information&provided&after&the&date&of&this&presentation.&&ATIGEO&MAKES&NO&WARRANTIES,&EXPRESS,& IMPLIED&OR&STATUTORY,&AS&TO&THE&INFORMATION&IN&THIS&PRESENTATION.

More Related