1. AI conference Odense 12th of September 2019
Digital solutions at the Emergency rescue department from a patient perspective
Mathias Karlsson, MD PhD
CMO, IBM Nordics
Watson Health © IBM Corporation 2017 1

2. IBM's statements regarding its plans, directions and intent are subject to change or withdrawal without notice at IBM's sole discretion.
Information regarding potential future products is intended to outline our general product direction and it should not be relied on in making a purchasing decision.
The information mentioned regarding potential future products is not a commitment, promise, or legal obligation to deliver any material, code or functionality. Information about potential future products may not be incorporated into any contract. The development, release, and timing of any future features or functionality described for our products remains at our sole discretion.
Watson Health © IBM Corporation 2019

3. 190, The future Growth of global ‘oldest old’ population
Shortage of healthcare workers
More people with chronic diseases
Empowered individuals
Increased healthcare expenditures
Flæskesteg med hvide og brune kartofler, rødkål og sauce ,
190,
Sometimes you might get the impression that the reason for digitalizing healthcare is to digitalize it. That is not the case
3x growth of global ‘oldest old’ population¹ with 1.6 billion People over the age of 60 by 2050
15 million Projected shortage of healthcare workers by 2030
Projected global healthcare spend by 2040 – nearly 2.5x as today
Almost 50% of the Swedish population live with a chronic disease
80% of healthcare expenditures are focused on those individuals.
The rising voice of the consumer is empowering individuals
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation

4. Emergency Rescue Department (ER) Analysis Sweden 2018
2m visits/year
The assignment for ER not clearly defined
Broad competence needed 24/7
60% outside office hrs,
old and sick patients
Very diverse diagnosis
Multiple level of competence behind “The doctor”
ER mirrors how well the whole Healthcare system works
Lack of hospital beds (approx. 50% inpatient care from ER)
Malfunctional primary care
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation

5. E-health or digitalization of health care
A bunch of new tools and new solution that could
Keep wrong patients out of the ER
Increase quality for our Patients at the ER (LoS, less errors)
Equality (psychiatric patients, Women etc)
Decrease the working load for Co-worker at ER
Have impact on costs
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation

6. One of the new tools: AI Watson’s Law
Knowledge
“By far, the greatest danger of Artificial Intelligence is that people conclude too early that they understand it.”
Eliezer Yudkowsky
AI Researcher and writer
Data
Watson Health © IBM Corporation 2019 6

7. Foundation The Evolution of AI Revolutionary Disruptive and Pervasive
General AI
Revolutionary
The Evolution of AI
Broad AI
Disruptive and Pervasive
Foundation
Narrow AI
Emerging
Watson Health © IBM Corporation 2019
We are here
2050 and beyond 7

8. The Foundation
Things that should be in place first

9. No Foundation? No place to put your AI solutions

10. Strategy and roadmap

11. EHR
Interoperability
Security

12. Data
Clinical
and
none-Clinical data

13. None-Clinical data at the ER
Identity and Access management
Robotic Process automation
Analytics
Asset management
Enkla respektive komplexa processer Processer kan delas in i enkla respektive komplexa. För enkla processer går det att i förväg bestämma hur hanteringen ska gå till i varje steg. Genom att utforma program som hanterar informationen på ett förutbestämt sätt enligt tydliga regler kan hanteringen automatiseras. Ett exempel på en enkel process är ansökan om boendeparkering. Om en person 1) är folkbokförd på en viss adress och 2) äger en bil har personen rätt till boendeparkering. Den information som behövs för att kontrollera detta finns lättillgänglig i adressregistret och fordonsregistret. Hanteringen kan automatiseras. I komplexa processer går det inte att i förhand bestämma hur alla steg ska hanteras. Det kan till exempel krävas tolkning av ostrukturerad text, tolkning av regler, bedömningar och avvägningar. Än så länge är sådana processer inte mjöliga att automatisera fullt ut. Ett exempel på en komplex process är ansökan om sjukpenning hos försäkringskassan. Där krävs oftast
läkarintyg där läkaren i fritext beskrivit symptom och arbetsförmåga. För att automatisera en process behöver informationen struktureras på ett tydligare sätt. Några steg i en komplex process kan dock vara av en sådan art att de kan automatiseras. Processen kan då automatiseras till viss del. En process kan vara komplicerad utan att för den skull vara komplex. Det kan till exempel krävas avancerade uträkningar i flera steg. Sådana uträkningar kan vara komplicerade och tidskrävande för en människa. För ett program är sådana beräkningar dock enkla att utföra på ett ögonblick då reglerna/formlerna är tydliga.
Computer-Aided Facility mangement

14. When the digital house is in order

15. AI could be used in the clinical work

16. Kim
DM-I, healthy life One-month history of headache, tired. Have noticed blood in the stool Last 2 days dizziness and fatigue
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation

17. Narrow AI: Initial Value Creation
General AI: Revolutionary
2016
2050–Beyond
Watson Health © IBM Corporation 2019 17

18. more flexible and intelligent
Make existing EHR
more flexible and intelligent
Data extraction from free text
Diagnostic and/or predictive algorithms
Clinical decision support
However, all of these capabilities need to be tightly integrated with EHRs to be effective. Most current AI options are “encapsulated” as standalone offerings and don’t provide as much value as integrated ones, and require time-pressed physicians to learn how to use new interfaces. But mainstream EHR vendors are beginning to add AI capabilities to make their systems easier to use. Firms like Epic, Cerner, Allscripts, and Athena are adding capabilities like natural language processing, machine learning for clinical decision support, integration with telehealth technologies and automated imaging analysis. This will provide integrated interfaces, access to data held within the systems, and multiple other benefits — though it will probably happen slowly.
Hb 104g/L, MCH 23, Ferritin 300
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation
Group Name / DOC ID / Month XX, 2019 / © 2019 IBM Corporation

19. Will Kim suffer from diabetes-related complications within the next 3 years?
Using real-world data from more than 600’000 patients, the Roche/IBM algorithm is able to identify people with diabetes who are at high risk for chronic kidney disease (CKD).
An average AUC = 79% is achieved for predicting the risk for the first 3 years after initial diagnosis of diabetes, outperforming literature algorithms.
The prediction is based on 7 features and very stable with respect to missing data.
Our Roche/IBM algorithm outperforms literature algorithms in a one-to-one comparison with subcohorts as well as the real-world cohort.
Nature Medicine 25, 57–59 (2019), January 7th, (free full text access:
Watson Health © IBM Corporation 2019

20. Patel N, et al. The Oncologist. 2018;23(2):179-185.
The Oncologist: Watson for Genomics found 99% of the actionable mutations identified by the molecular tumor board
Patel N, et al. The Oncologist. 2018;23(2):
< 3 min
99%
32%
Watson for Genomics took less than 3 minutes per case. Found 99% of the actionable mutations identified by UNC’s MTB and identified additional actionable mutations that the MTB did not in 32% of the cases (N = 1,018)
“Molecular tumor boards empowered by cognitive computing can significantly improve patient care by providing a fast, cost-effective, and comprehensive approach for data analysis in the delivery of precision medicine.” -The Oncologist
***DO NOT EXTRAPOLATE MEANING FROM ANY EVIDENCE. THE STATS QUOTED ON THIS SLIDE REPRESENT ONLY THE RESEARCH DISCUSSED IN THE MANUSCRIPT AND DO NOT SUPPORT BROADER PRODUCT CLAIMS***
*Slide previously QMS approved; key points for speaker new; need review*
REFERENCE
Patel N, Michelini V, Snell J , Balu S, Hoyle A, Parker J, Hayward M, Eberhard D, Salazar A, McNeillie P, Xu J, Huettner C, Koyama T, Utro F, Rhrissorrakrai K, Norel R, Bilal E, Royyuru A, Parida L, Earp H, Grilley-Olson J, Hayes D, Harvey S, Sharpless N, Kim W. Enhancing next-generation sequencing-guided cancer care through cognitive computing [published online November 20, 2017]. Oncologist. 2018;23(2): doi: /theoncologist
KEY POINTS:
In a comparative study, Watson for Genomics (WfG) analyzed 1018 patient cases previously sequenced by UNCseq and analyzed at the University of North Carolina through human-led methods
In less than 3 minutes of analysis per case, WfG was able to identify all variants previously defined as actionable by the human-only molecular tumor board as well as detect 323 additional actionable variants not previously identified
These results demonstrate that the interpretation and actionability of somatic NGS results are evolving too rapidly to rely solely on human curation. Molecular tumor boards empowered by cognitive computing may be able to improve patient care by providing a rapid, comprehensive approach for data analysis and consideration of up‐to‐date availability of clinical trials.
ABSTRACT
BACKGROUND:
Using next-generation sequencing (NGS) to guide cancer therapy has created challenges in analyzing and reporting large volumes of genomic data to patients and caregivers. Specifically, providing current, accurate information on newly approved therapies and open clinical trials requires considerable manual curation performed mainly by human "molecular tumor boards" (MTBs). The purpose of this study was to determine the utility of cognitive computing as performed by Watson for Genomics (WfG) compared with a human MTB.
MATERIALS AND METHODS:
One thousand eighteen patient cases that previously underwent targeted exon sequencing at the University of North Carolina (UNC) and subsequent analysis by the UNCseq informatics pipeline and the UNC MTB between November 7, 2011, and May 12, 2015, were analyzed with WfG, a cognitive computing technology for genomic analysis.
RESULTS:
Using a WfG-curated actionable gene list, we identified additional genomic events of potential significance (not discovered by traditional MTB curation) in 323 (32%) patients. The majority of these additional genomic events were considered actionable based upon their ability to qualify patients for biomarker-selected clinical trials. Indeed, the opening of a relevant clinical trial within 1 month prior to WfG analysis provided the rationale for identification of a new actionable event in nearly a quarter of the 323 patients. This automated analysis took <3 minutes per case.
CONCLUSION:
These results demonstrate that the interpretation and actionability of somatic NGS results are evolving too rapidly to rely solely on human curation. Molecular tumor boards empowered by cognitive computing could potentially improve patient care by providing a rapid, comprehensive approach for data analysis and consideration of up-to-date availability of clinical trials.
IMPLICATIONS FOR PRACTICE:
The results of this study demonstrate that the interpretation and actionability of somatic next-generation sequencing results are evolving too rapidly to rely solely on human curation. Molecular tumor boards empowered by cognitive computing can significantly improve patient care by providing a fast, cost-effective, and comprehensive approach for data analysis in the delivery of precision medicine. Patients and physicians who are considering enrollment in clinical trials may benefit from the support of such tools applied to genomic data.
*excerpt from manuscript
Watson Health © IBM Corporation 2019

21. Successes together – matching patients-to-trials
84 % increase in
monthly enrollment
Cognitive technology supports increased enrollment in clinical trials for breast cancer. In July 2016, Mayo Clinic implemented IBM Watson for Clinical Trial Matching with a team of screening clinical research coordinators in its ambulatory practice for patients with breast cancer at the Rochester campus.
In the 18 months after implementation, there was on average an 84 percent increase in enrollment to Mayo’s systemic therapy clinical trials for breast cancer. The time to screen an individual patient for clinical trial matches also fell when compared with traditional manual methods.1
This was further increased to 8.5 patients/month when including accruals to breast cancer cohorts of multi-disease, phase I trials within the experimental cancer therapeutics program.
DISCLAIMER: This slide may not be modified or the presentation of the data altered. The context provided for this study must be kept intact.
1: Haddad T, et al. ASCO 2018
Watson Health © IBM Corporation 2018

22. Schork NJ, Personalized medicine: Time for one-person trials. Nature

23. Patient Similarity Analytics
Extract historical patient event trails & characteristics
Patient Similarity Analytics
Clinical Histories Database
Patient Characteristics
Here is how it can be implemented.
Using patient similarity analytics, cognitive can identify retrospective cohort of similar individuals
Create visual summary of various clinical pathways and outcomes pathways
Diagram; Start with COPD patients on corticosteroids; color represents outcomes (dark blue better, red is worse); march through various states: Dx, lab results, treatments
Not same strength of evidence gained from RCTs that informs guidelines; but, does give clinician additional point of care insight beyond the treatment guidelines that may inform decision making

24. Narrow AI: Initial Value Creation
Broad AI:
Disruptive and Pervasive
General AI: Revolutionary ….
2016
2050–Beyond
We are here
Watson Health © IBM Corporation 2019 24

25. Moving from Narrow AI to Broad AI…  Medtronic’s Sugar
Moving from Narrow AI to Broad AI…  Medtronic’s Sugar.IQ Empowering Patients

26. Broad AI Examples
Watson Health © IBM Corporation 2019

27. It is no longer about Big Data …
It is no longer about Big Data ….. We are already living in the world of Broad Data
Watson Health © IBM Corporation 2019

28. From predicting likelihood to individual management
Evidence-based medicine
Precision Medicine
Disease Management
Preventative Medicine
From predicting likelihood to individual management
Broad AI+ Data
Across individuals
Individual
& sensor data
Data integration
Molecular Data
Big Data
Photo by jesse orrico on Unsplash

29. General AI: Revolutionary
2050–Beyond
Watson Health © IBM Corporation 2019 29 29

31. Legal Disclaimer © IBM Corporation 2019. All Rights Reserved.
The information contained in this publication is provided for informational purposes only. While efforts were made to verify the completeness and accuracy of the information contained in this publication, it is provided AS IS without warranty of any kind, express or implied. In addition, this information is based on IBM’s current product plans and strategy, which are subject to change by IBM without notice. IBM shall not be responsible for any damages arising out of the use of, or otherwise related to, this publication or any other materials. Nothing contained in this publication is intended to, nor shall have the effect of, creating any warranties or representations from IBM or its suppliers or licensors, or altering the terms and conditions of the applicable license agreement governing the use of IBM software.
References in this presentation to IBM products, programs, or services do not imply that they will be available in all countries in which IBM operates. Product release dates and/or capabilities referenced in this presentation may change at any time at IBM’s sole discretion based on market opportunities or other factors, and are not intended to be a commitment to future product or feature availability in any way. Nothing contained in these materials is intended to, nor shall have the effect of, stating or implying that any activities undertaken by you will result in any specific sales, revenue growth or other results.
Performance is based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput or performance that any user will experience will vary depending upon many factors, including considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve results similar to those stated here.
All customer examples described are presented as illustrations of how those customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics may vary by customer.
IBM, the IBM logo, ibm.com, and Watson Health are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at ibm.com/legal/copytrade.
Watson Health © IBM Corporation 2019
IBM Confidential