mak akin ing sen ense o e of big ig da data in a in hea
play

Mak akin ing sen ense o e of big ig da data in a in hea ealt - PowerPoint PPT Presentation

latrob obe.edu.a .edu.au Mak akin ing sen ense o e of big ig da data in a in hea ealt lthca care re wi with ef effec ectiv ive e an and d in inter erac activ ive e an anal alys ysis is Dr. Li Lianu anuhu hua (Lin


  1. latrob obe.edu.a .edu.au Mak akin ing sen ense o e of big ig da data in a in hea ealt lthca care re wi with ef effec ectiv ive e an and d in inter erac activ ive e an anal alys ysis is Dr. Li Lianu anuhu hua (Lin ina) a) Chi Depar artm tment ent of Comput uter er Scie ienc nce e and Informat ormation ion Techn hnolo ology gy La Trobe University CRICOS Provider Code Number 00115M

  2. latrob obe.edu.a .edu.au A b A bit it ab about out me me Global ally ly Top p You oung ng Global al Romberg mberg 200 You oung ng Change nger by by Exter erna nal l Grant nt “Best Paper Resear searcher hers s Think nk 20 20 Honors 2017 2017 Award” at Aw Awar arded ded by 4 th th Awar Aw ard d by by /G20 by IBM th HLF 4 th PAKDD KDD20 2013 HLF La Trobe University La ersity PhD Period iod at HUST (China na) ) and UTS (Aust stralia ralia) ) IBM M Resear earch ch Aust stral alia ia 02/201 018 2009 2009 05/2015 015 201 2013 Bendigo Data Watson On-Line Big data fast SURO: Data Health: Warehouse Health Analytical response: Surgical Unit Storming (ML) Modeling Tool Processing Resource real-time Survey and (DWDesigner) System (OLAP) Optimisation classification Data AI on Digital ital Health lth (ML Section) of big data Visualisation stream Data a Mining ng Machine ne Learni ning ng Machine ne Learni ning ng Resear search Area Resear search h Area Natura ral Langua nguage Proces essing sing Slide 2 | Version 2

  3. latrob obe.edu.a .edu.au No No ind ndus ustr try cou ount nts more tha han health lthcar care Slide 3 | Version 2

  4. latrob obe.edu.a .edu.au No No ind ndus ustr try cou ount nts less ss tha han he healthc lthcare are Slide 4 | Version 2

  5. latrob obe.edu.a .edu.au Ever ery yea ear 18,000+ 18,000+ Aust stralia ralian de deat aths hs are are cau aused sed by by medica cal error, 12,000 12,000 of of whom wer were e dying becaus ause e of preventabl entable e events nts. (This was published in 1995 and it is likely increasing each year) A survey (publi blish shed ed in in 1999): 1999): 14,000 admissions surveyed 16.6% associated with an “ adverse erse event nt ” 51% of the adverse events considered preventa ntabl ble 4.9% the patient died See MJA: The incidence and cost of adverse events in Victorian hospitals 2003 – 04 * ABS: Transport accidents: 1,402 deaths registered in 2008. Slide 5 | Version 2

  6. latrob obe.edu.a .edu.au The total cost of adverse rse events ts (in the ye year r 2003 2003 – 04 04 to selecte ted d Victori rian an hospita tals ls )”: • $460.311 million • representing 15.7% of the total expenditure on direct hospital costs • or an additional 18.6% of the total inpatient hospital budget. Adverse events are associated with significant costs. Administrative datasets are a cost-effective source of information that can be used for a range of clinical governance activities to prevent adverse events. Slide 6 | Version 2

  7. latrob obe.edu.a .edu.au The he Return turn On On In Invest estment ment (R (ROI OI) on on big big data data approa oaches ches wi will be be even hi high gher er for or he healthcare thcare tha han ot other her ind ndus ustrie tries*. * Source from Harvard Medical School. Slide 7 | Version 2

  8. latrob obe.edu.a .edu.au https://www.aihw.gov.au/reports- statistics/health-welfare-overview/health- welfare-expenditure/overview Slide 8 | Version 2

  9. latrob obe.edu.a .edu.au Necessar essary con onditions ditions for or data to to ta take on on new li life in Health in althcare: are: Ac Access ess to to Incenti entives es Da Data ta Slide 9 | Version 2

  10. latrob obe.edu.a .edu.au Digi gital tal Health th Data in in Aus ustra trali lia Slide 10 | Version 2

  11. latrob obe.edu.a .edu.au Digi gital tal Health th in in Aust stra rali lia My Health TeleHealth Record Health Identifier Service Slide 11 | Version 2

  12. latrob obe.edu.a .edu.au My My Health th Record ord Population ulation registr gistration ation: As of 23 March 2016, there was a total of 2,642,278 active digital records, approximately 11% of the current Australian population. Slide 12 | Version 2 From: Evolution of eHealth in Australia, Achievements, lessons, and opportunities

  13. latrob obe.edu.a .edu.au My My Health th Record ord Organi anisat atio ion regist gistrat ration ion: As of 23 March 2016, a total of 8,139 organisations were registered in the My Health Record system. This number continues to increase steadily Slide 13 | Version 2

  14. latrob obe.edu.a .edu.au Health th Id Ident ntifier ifier (H (HI) I) Servic vice Organisations registered in the HI Service by state / territory and type – 29 Feb 2016 Slide 14 | Version 2

  15. latrob obe.edu.a .edu.au Teleheal ehealth th For the 2014-15 financial year, there were over 84,000 MBS claims for telehealth consultations 1 . A CSIRO telehealth and monitoring study found a 37% reduction in mortality with the use of the Telemedcare Clinical Monitoring Unit in the management of chronic disease in the home. 1 Source: Medicare Statistics, Medicare Item Reports July 2014 – July 2015. 2 Results of the CSIRO multi ‐ site national trial of telehealth for the management of chronic disease in the home, in Health Informatics Conference, Brisbane , 2015. Slide 15 | Version 2

  16. latrob obe.edu.a .edu.au In Incenti entives es in in Aus ustra trali lia Slide 16 | Version 2

  17. latrob obe.edu.a .edu.au The Aus ustralian tralian heal althcare thcare sy syst stem em is s pot otentia ntially lly de deal aling ng with h two o ma main n prob oblems1 lems1 ? Perf rformanc ormance e and and ? Patien Patient t Resource source Out utcomes comes Alloca location tion Impro provements ements 1 A review of the Australian healthcare system: A policy perspective, 2018 Slide 17 | Version 2

  18. latrob obe.edu.a .edu.au Data Ana nalytic ytics Slide 18 | Version 2

  19. latrob obe.edu.a .edu.au Wh Which diabetic etics are are likely ely to to benefi nefit fr from om int nter erventi ention? on? Slide 19 | Version 2

  20. latrob obe.edu.a .edu.au Repor orting ting Which diabetics are likely to Select count(PatientID) from benefit from intervention? PatientTable join PatientID from Claims where ICD9 is (‘250 * ’) or NDC is (‘25545565’, ‘34982738’) and VisitDate between (“1/1/2010”, ’31/12/2010’) = null Slide 20 | Version 2

  21. latrob obe.edu.a .edu.au Risk sk Scor ores es Whi hich diabeti betics cs are likely ely to to bene nefit fit from int nterventi ntion? n? Claim aim data data Logist istic ic Regre ress ssion ion • Leukemi emia • Diabetes abetes Code or or • Heart failure Rules es • Dementia • COPD • etc Slide 21 | Version 2

  22. latrob obe.edu.a .edu.au • He Heavy avy de dependen endence ce on on di dise sease ase co code des • Risk sk sc scor ores es of often on one si size ze fits al all – al all di disease sease, pop opulati ulation, on, etc Slide 22 | Version 2

  23. latrob obe.edu.a .edu.au How to to answer er th these se question estions fr from om Do Docto tors? rs? • I know who cost more last year, I need to know which of my diabetics is most likely to end up in the emergency department soonest. • I already know these five people I am going to see them in any way. Tell me who’s pre - diabetics that’s heading toward diabetes that I can do something about that’s very different. Slide 23 | Version 2

  24. latrob obe.edu.a .edu.au How Su Super pervised vised Machine chine Learning arning Wor orks ks Whi hich h of my y diabetics betics is most t likely ly to to end nd up i in th n the emergen rgency y department? tment? Diabetic abetics Diabetic abetics Classif assific icatio ation without hout ED ED with ED ED Model el Mach chine ine Learning ning Slide 24 | Version 2

  25. latrob obe.edu.a .edu.au How Su Super pervised vised Machine chine Learning arning Wor orks ks Clas assifi ificat cation ion Model Diabetic abetics 95% 95% 90% 90% 89% 89% Slide 25 | Version 2

  26. latrob obe.edu.a .edu.au Machin chine Learning arning con onsiders siders mo more re data ta Risk of of Diabeti betic likely ely to to be be re re-admission admission admi mitted tted • Length of Stay • Length of Stay • Admitted via ED • Admitted via ED • # ED visits • # ED visits • All co-morbidities • Substance Issues • Specialty Provider Types • Behavioral Health • Nurses’ impressions • Etc …… Slide 26 | Version 2

  27. latrob obe.edu.a .edu.au Machine chine Learning arning • More tolerant erant to to missing, g, inaccurat curate data • Models ls learn with new data Slide 27 | Version 2

  28. latrob obe.edu.a .edu.au Resource source Allocatio location – Ho Hospit spital al Waiting ting Time mes Problem: blem: How many patients should we operate on every day for the next month in order to increase the hospital revenue, in respect to the limited number of beds and operating theatres? Slide 28 | Version 2

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend