SlideShare a Scribd company logo
© 2020 www.waylay.io | page/1
Remote Patient & Elderly Care
Monitoring
Veselin Pizurica, CTO
AI Session #6 Biotech & Pharma THU 5 November - Online session
© 2020 www.waylay.io | page/2
How is COVID changing the perspective?
© 2020 www.waylay.io | page/3
Remote patient monitoring
© 2020 www.waylay.io | page/4
Sepsis Watch - Duke University
https://physicians.dukehealth.org/articles/dukes-augmented-intelligence-system-helps-prevent-sepsis-ed
Following a successful pilot phase from November 2018 through June 2019, Sepsis
Watch, an augmented intelligence (AI) system developed by Duke, has now been
deployed in the emergency departments (ED) at all three Duke University Health
System hospitals to identify patients in the early stages of developing life-threatening
sepsis. The system signals an AI-computed risk of sepsis so that care teams can
quickly begin treatment.
© 2020 www.waylay.io | page/5
Explainability or not?
Dataset: Data were curated from the local quarternary academic hospital with over 1,000 beds and over
40,000 inpatient admissions per year. In total, the model development and evaluation dataset contained over
32 million data points.
Model explainability was not prioritized, because regulations promote standardized treatment of sepsis,
regardless of cause. As previously described, human experts often disagree about sepsis diagnoses and major
medical and public health organizations publicly promote distinct disease definitions.
The model generates risk scores every hour for every adult patient to detect sepsis. The model extended prior
work using recurrent neural networks (RNNs) for clinical event detection by coupling a RNN with a multi-task
gaussian processes (MGPs).
Although there are emerging methods to improve explainability of RNNs, end users cannot contemplate the
entire model and cannot reliably understand the relationships between model inputs and outputs. As such,
there was not a deliberate effort to explain the MGP-RNN model output.
© 2020 www.waylay.io | page/6
Using DL models in Waylay
© 2020 www.waylay.io | page/7
Can we blindly trust DL models?
https://www.nabla.com/blog/gpt-3/
© 2020 www.waylay.io | page/8
Do we need human interpretation?
https://aaas.confex.com/aaas/2017/webprogram/Paper19142.html
http://nautil.us/issue/40/learning/is-artificial-intelligence-permanently-inscrutable#disqus_thread
© 2020 www.waylay.io | page/9
Sepsis shock diagnosis
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4958885/
https://onlinelibrary.wiley.com/doi/full/10.1111/imj.14199
© 2020 www.waylay.io | page/10
Use case: Sepsis Shock
Sepsis is an abnormal inflammatory response to infection and is the leading cause of inpatient mortality within hospitals in
the United States (*). Possible complications include:
● heart failure
● abnormal blood clotting
● kidney failure
● respiratory failure
● stroke
● liver failure
● loss of a portion of the bowel
● loss of portions of the extremities
* Rhee, C. et al. 2017. Incidence and Trends of Sepsis in US Hospitals Using Clinical vs Claims Data, 2009-2014. JAMA. 318, 13 (Oct. 2017), 1241–1249.
© 2020 www.waylay.io | page/11
Sepsis Shock as a rule
https://www.mdcalc.com/sirs-sepsis-septic-shock-criteria
© 2020 www.waylay.io | page/12
Explainability : Decision trees
© 2020 www.waylay.io | page/13
Branching hell - the curse of dimensionality
The depth of the tree grows linearly with the number of variables, but the number of branches grows exponentially with
the number of states. Let’s consider X and Y variables
with three possible outcomes:
As the figure above shows, this leads to 18 leaf nodes (red/green dots) and overall 31 nodes for only two variables! We
have simplified the example by assuming only subset of AND Boolean functions for a given combination. For instance, we
didn't model (X is low) OR (Y is either medium or high) -> as that would be mean connecting some other branches. How
many distinct decision trees do we have with 'n' Boolean attributes? Answer = #number of Boolean functions = #number
of distinct truth tables with 2^n rows = 2^2^n
e.g.: With 6 Boolean attributes we get 18,446,744,073,709,551,616 (2^64 which is the upper limit of an unsigned long)
13
https://www.waylay.io/blog/the-curse-of-dimensionality-in-decision-trees-branching-problem/
© 2020 www.waylay.io | page/14
Dealing with time and majority voting?
14
This is a majority voting problem (impossible to express with decision trees)
© 2020 www.waylay.io | page/15
Is there another way? Waylay
15
© 2020 www.waylay.io | page/16
Demo: Sepsis shock monitoring
16
© 2020 www.waylay.io | page/17
Demo setup
17
Veselin
Steve
Paul
Piet
© 2020 www.waylay.io | page/18
Demo
18
© 2020 www.waylay.io | page/19
Learn more
BLOG | DATA SHEETS | WHITE PAPERS | VIDEOS | WEBINARS | SERVICES
For a free demo of the Waylay platform, reach out to sales@waylay.io or fill-in the form at
waylay.io/get-demo/ and we’ll get in touch with you.

More Related Content

What's hot

Unified Approach to Interpret Machine Learning Model: SHAP + LIME
Unified Approach to Interpret Machine Learning Model: SHAP + LIMEUnified Approach to Interpret Machine Learning Model: SHAP + LIME
Unified Approach to Interpret Machine Learning Model: SHAP + LIME
Databricks
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
Sri Ambati
 
Machine Learning part1 - Introduction to Data Science
Machine Learning part1 - Introduction to Data Science Machine Learning part1 - Introduction to Data Science
Machine Learning part1 - Introduction to Data Science
Frank Kienle
 
AI: The next frontier by Amparo Alonso at Big Data Spain 2017
AI: The next frontier by Amparo Alonso at Big Data Spain 2017AI: The next frontier by Amparo Alonso at Big Data Spain 2017
AI: The next frontier by Amparo Alonso at Big Data Spain 2017
Big Data Spain
 
DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictions
Anton Kulesh
 
Artificial intelligence and IoT
Artificial intelligence and IoTArtificial intelligence and IoT
Artificial intelligence and IoT
Veselin Pizurica
 
Machine Learning Interpretability
Machine Learning InterpretabilityMachine Learning Interpretability
Machine Learning Interpretability
inovex GmbH
 
Think Big | Enterprise Artificial Intelligence
Think Big | Enterprise Artificial IntelligenceThink Big | Enterprise Artificial Intelligence
Think Big | Enterprise Artificial Intelligence
Data Science Milan
 
DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...
DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...
DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...
Dataconomy Media
 
Inteligent computing relating to cloud computing.final
Inteligent computing relating to cloud computing.finalInteligent computing relating to cloud computing.final
Inteligent computing relating to cloud computing.final
Er. rahul abhishek
 
A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...
A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...
A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...
Amélie Gyrard
 
Interpretable machine learning : Methods for understanding complex models
Interpretable machine learning : Methods for understanding complex modelsInterpretable machine learning : Methods for understanding complex models
Interpretable machine learning : Methods for understanding complex models
Manojit Nandi
 
Eclipse IoT - Day 0 of thingmonk 2016
Eclipse IoT - Day 0 of  thingmonk 2016Eclipse IoT - Day 0 of  thingmonk 2016
Eclipse IoT - Day 0 of thingmonk 2016
Boris Adryan
 
EclipseCon France 2015 - Science Track
EclipseCon France 2015 - Science TrackEclipseCon France 2015 - Science Track
EclipseCon France 2015 - Science Track
Boris Adryan
 
Architecting AI Applications
Architecting AI ApplicationsArchitecting AI Applications
Architecting AI Applications
Mikio L. Braun
 
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
Dataconomy Media
 
Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...
Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...
Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...
Big Data Spain
 
Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...
Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...
Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...
Big Data Spain
 
Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...
Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...
Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...
Hong-Linh Truong
 
Efficient Data Labelling for Ocular Imaging
Efficient Data Labelling for Ocular ImagingEfficient Data Labelling for Ocular Imaging
Efficient Data Labelling for Ocular Imaging
PetteriTeikariPhD
 

What's hot (20)

Unified Approach to Interpret Machine Learning Model: SHAP + LIME
Unified Approach to Interpret Machine Learning Model: SHAP + LIMEUnified Approach to Interpret Machine Learning Model: SHAP + LIME
Unified Approach to Interpret Machine Learning Model: SHAP + LIME
 
Interpretable Machine Learning
Interpretable Machine LearningInterpretable Machine Learning
Interpretable Machine Learning
 
Machine Learning part1 - Introduction to Data Science
Machine Learning part1 - Introduction to Data Science Machine Learning part1 - Introduction to Data Science
Machine Learning part1 - Introduction to Data Science
 
AI: The next frontier by Amparo Alonso at Big Data Spain 2017
AI: The next frontier by Amparo Alonso at Big Data Spain 2017AI: The next frontier by Amparo Alonso at Big Data Spain 2017
AI: The next frontier by Amparo Alonso at Big Data Spain 2017
 
DC02. Interpretation of predictions
DC02. Interpretation of predictionsDC02. Interpretation of predictions
DC02. Interpretation of predictions
 
Artificial intelligence and IoT
Artificial intelligence and IoTArtificial intelligence and IoT
Artificial intelligence and IoT
 
Machine Learning Interpretability
Machine Learning InterpretabilityMachine Learning Interpretability
Machine Learning Interpretability
 
Think Big | Enterprise Artificial Intelligence
Think Big | Enterprise Artificial IntelligenceThink Big | Enterprise Artificial Intelligence
Think Big | Enterprise Artificial Intelligence
 
DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...
DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...
DN18 | The Evolution and Future of Graph Technology: Intelligent Systems | Ax...
 
Inteligent computing relating to cloud computing.final
Inteligent computing relating to cloud computing.finalInteligent computing relating to cloud computing.final
Inteligent computing relating to cloud computing.final
 
A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...
A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...
A Unified Semantic Engine for Internet of Things and Smart Cities: From Senso...
 
Interpretable machine learning : Methods for understanding complex models
Interpretable machine learning : Methods for understanding complex modelsInterpretable machine learning : Methods for understanding complex models
Interpretable machine learning : Methods for understanding complex models
 
Eclipse IoT - Day 0 of thingmonk 2016
Eclipse IoT - Day 0 of  thingmonk 2016Eclipse IoT - Day 0 of  thingmonk 2016
Eclipse IoT - Day 0 of thingmonk 2016
 
EclipseCon France 2015 - Science Track
EclipseCon France 2015 - Science TrackEclipseCon France 2015 - Science Track
EclipseCon France 2015 - Science Track
 
Architecting AI Applications
Architecting AI ApplicationsArchitecting AI Applications
Architecting AI Applications
 
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
DN18 | Applied Machine Learning in Cybersecurity: Detect malicious DGA Domain...
 
Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...
Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...
Meme Index. Analyzing fads and sensations on the Internet by Miguel Romero at...
 
Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...
Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...
Feature selection for Big Data: advances and challenges by Verónica Bolón-Can...
 
Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...
Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...
Emerging Dynamic TUW-ASE Summer 2015 - Distributed Systems and Challenges for...
 
Efficient Data Labelling for Ocular Imaging
Efficient Data Labelling for Ocular ImagingEfficient Data Labelling for Ocular Imaging
Efficient Data Labelling for Ocular Imaging
 

Similar to Remote Patient & Elderly Care Monitoring

Detecting outliers and anomalies in data streams
Detecting outliers and anomalies in data streamsDetecting outliers and anomalies in data streams
Detecting outliers and anomalies in data streams
fatimabenjelloun1
 
PREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHES
PREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHESPREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHES
PREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHES
IRJET Journal
 
Freek bomhof tno
Freek bomhof tnoFreek bomhof tno
Freek bomhof tno
BigDataExpo
 
Beyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysis
Beyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysisBeyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysis
Beyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysis
PetteriTeikariPhD
 
Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...
Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...
Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...
multimediaeval
 
The big data challenge in healthcare and how can business intelligence best d...
The big data challenge in healthcare and how can business intelligence best d...The big data challenge in healthcare and how can business intelligence best d...
The big data challenge in healthcare and how can business intelligence best d...
HealthXn
 
Fake News and Message Detection
Fake News and Message DetectionFake News and Message Detection
Fake News and Message Detection
IRJET Journal
 
Data science landscape in the insurance industry
Data science landscape in the insurance industryData science landscape in the insurance industry
Data science landscape in the insurance industry
Stefano Perfetti
 
Capstone Project.pptx
Capstone Project.pptxCapstone Project.pptx
Capstone Project.pptx
ARESProject1
 
Multipleregression covidmobility and Covid-19 policy recommendation
Multipleregression covidmobility and Covid-19 policy recommendationMultipleregression covidmobility and Covid-19 policy recommendation
Multipleregression covidmobility and Covid-19 policy recommendation
Kan Yuenyong
 
RCA CERN Grand Challenge 10 dec 2018
RCA CERN Grand Challenge 10 dec 2018RCA CERN Grand Challenge 10 dec 2018
RCA CERN Grand Challenge 10 dec 2018
Future Agenda
 
Single view vs. multiple views scatterplots
Single view vs. multiple views scatterplotsSingle view vs. multiple views scatterplots
Single view vs. multiple views scatterplots
IJECEIAES
 
IRJET- Breast Cancer Relapse Prognosis by Classic and Modern Structures o...
IRJET-  	  Breast Cancer Relapse Prognosis by Classic and Modern Structures o...IRJET-  	  Breast Cancer Relapse Prognosis by Classic and Modern Structures o...
IRJET- Breast Cancer Relapse Prognosis by Classic and Modern Structures o...
IRJET Journal
 
Sensitivity Analysis
Sensitivity AnalysisSensitivity Analysis
Sensitivity Analysis
Beth Johnson
 
Wireless Communication, Sensing and REM: A Security Perspective
Wireless Communication, Sensing and REM: A Security PerspectiveWireless Communication, Sensing and REM: A Security Perspective
Wireless Communication, Sensing and REM: A Security Perspective
IRJET Journal
 
Healthcare deserts: How accessible is US healthcare?
Healthcare deserts: How accessible is US healthcare?Healthcare deserts: How accessible is US healthcare?
Healthcare deserts: How accessible is US healthcare?
Data Con LA
 
Overall presentation Matram project
Overall presentation Matram project Overall presentation Matram project
Overall presentation Matram project RaphaelGirod
 
A Review on Credit Card Default Modelling using Data Science
A Review on Credit Card Default Modelling using Data ScienceA Review on Credit Card Default Modelling using Data Science
A Review on Credit Card Default Modelling using Data Science
YogeshIJTSRD
 
A SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNING
A SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNINGA SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNING
A SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNING
IRJET Journal
 
A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...
A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...
A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...
IJECEIAES
 

Similar to Remote Patient & Elderly Care Monitoring (20)

Detecting outliers and anomalies in data streams
Detecting outliers and anomalies in data streamsDetecting outliers and anomalies in data streams
Detecting outliers and anomalies in data streams
 
PREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHES
PREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHESPREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHES
PREDICTION OF COVID-19 USING MACHINE LEARNING APPROACHES
 
Freek bomhof tno
Freek bomhof tnoFreek bomhof tno
Freek bomhof tno
 
Beyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysis
Beyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysisBeyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysis
Beyond Broken Stick Modeling: R Tutorial for interpretable multivariate analysis
 
Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...
Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...
Efficient Supervision Net: Polyp Segmentation using EfficientNet and Attentio...
 
The big data challenge in healthcare and how can business intelligence best d...
The big data challenge in healthcare and how can business intelligence best d...The big data challenge in healthcare and how can business intelligence best d...
The big data challenge in healthcare and how can business intelligence best d...
 
Fake News and Message Detection
Fake News and Message DetectionFake News and Message Detection
Fake News and Message Detection
 
Data science landscape in the insurance industry
Data science landscape in the insurance industryData science landscape in the insurance industry
Data science landscape in the insurance industry
 
Capstone Project.pptx
Capstone Project.pptxCapstone Project.pptx
Capstone Project.pptx
 
Multipleregression covidmobility and Covid-19 policy recommendation
Multipleregression covidmobility and Covid-19 policy recommendationMultipleregression covidmobility and Covid-19 policy recommendation
Multipleregression covidmobility and Covid-19 policy recommendation
 
RCA CERN Grand Challenge 10 dec 2018
RCA CERN Grand Challenge 10 dec 2018RCA CERN Grand Challenge 10 dec 2018
RCA CERN Grand Challenge 10 dec 2018
 
Single view vs. multiple views scatterplots
Single view vs. multiple views scatterplotsSingle view vs. multiple views scatterplots
Single view vs. multiple views scatterplots
 
IRJET- Breast Cancer Relapse Prognosis by Classic and Modern Structures o...
IRJET-  	  Breast Cancer Relapse Prognosis by Classic and Modern Structures o...IRJET-  	  Breast Cancer Relapse Prognosis by Classic and Modern Structures o...
IRJET- Breast Cancer Relapse Prognosis by Classic and Modern Structures o...
 
Sensitivity Analysis
Sensitivity AnalysisSensitivity Analysis
Sensitivity Analysis
 
Wireless Communication, Sensing and REM: A Security Perspective
Wireless Communication, Sensing and REM: A Security PerspectiveWireless Communication, Sensing and REM: A Security Perspective
Wireless Communication, Sensing and REM: A Security Perspective
 
Healthcare deserts: How accessible is US healthcare?
Healthcare deserts: How accessible is US healthcare?Healthcare deserts: How accessible is US healthcare?
Healthcare deserts: How accessible is US healthcare?
 
Overall presentation Matram project
Overall presentation Matram project Overall presentation Matram project
Overall presentation Matram project
 
A Review on Credit Card Default Modelling using Data Science
A Review on Credit Card Default Modelling using Data ScienceA Review on Credit Card Default Modelling using Data Science
A Review on Credit Card Default Modelling using Data Science
 
A SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNING
A SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNINGA SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNING
A SURVEY ON BLOOD DISEASE DETECTION USING MACHINE LEARNING
 
A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...
A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...
A Novel Integrated Framework to Ensure Better Data Quality in Big Data Analyt...
 

More from Veselin Pizurica

Has serverless adoption hit a roadblock?
Has serverless adoption hit a roadblock?Has serverless adoption hit a roadblock?
Has serverless adoption hit a roadblock?
Veselin Pizurica
 
How to use probabilistic inference programming for application orchestration ...
How to use probabilistic inference programming for application orchestration ...How to use probabilistic inference programming for application orchestration ...
How to use probabilistic inference programming for application orchestration ...
Veselin Pizurica
 
Waylay - We are hiring
Waylay - We are hiringWaylay - We are hiring
Waylay - We are hiring
Veselin Pizurica
 
Waylay building smart city solution: IoT convention Antwerp
Waylay building smart city solution: IoT convention AntwerpWaylay building smart city solution: IoT convention Antwerp
Waylay building smart city solution: IoT convention Antwerp
Veselin Pizurica
 
Google Cloud infrastructure in Conrad Connect by Google & waylay
Google Cloud infrastructure in Conrad Connect by Google & waylayGoogle Cloud infrastructure in Conrad Connect by Google & waylay
Google Cloud infrastructure in Conrad Connect by Google & waylay
Veselin Pizurica
 
Artificial intelligence by Aleksandra Pizurica
Artificial intelligence by Aleksandra PizuricaArtificial intelligence by Aleksandra Pizurica
Artificial intelligence by Aleksandra Pizurica
Veselin Pizurica
 
Automation is eating the world
Automation is eating the worldAutomation is eating the world
Automation is eating the world
Veselin Pizurica
 
LPWAN - IoT (Platform) Killer Application
LPWAN - IoT (Platform) Killer ApplicationLPWAN - IoT (Platform) Killer Application
LPWAN - IoT (Platform) Killer Application
Veselin Pizurica
 
Automation, intelligence and knowledge modelling
Automation, intelligence and knowledge modellingAutomation, intelligence and knowledge modelling
Automation, intelligence and knowledge modelling
Veselin Pizurica
 
When IoT Meets Artificial Intelligence
 When IoT Meets Artificial Intelligence When IoT Meets Artificial Intelligence
When IoT Meets Artificial Intelligence
Veselin Pizurica
 
Internet of Things introduction
Internet of Things introductionInternet of Things introduction
Internet of Things introduction
Veselin Pizurica
 
My life in one picture
My life in one pictureMy life in one picture
My life in one picture
Veselin Pizurica
 
WebRTC presentation
WebRTC presentationWebRTC presentation
WebRTC presentation
Veselin Pizurica
 
A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...
A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...
A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...
Veselin Pizurica
 

More from Veselin Pizurica (14)

Has serverless adoption hit a roadblock?
Has serverless adoption hit a roadblock?Has serverless adoption hit a roadblock?
Has serverless adoption hit a roadblock?
 
How to use probabilistic inference programming for application orchestration ...
How to use probabilistic inference programming for application orchestration ...How to use probabilistic inference programming for application orchestration ...
How to use probabilistic inference programming for application orchestration ...
 
Waylay - We are hiring
Waylay - We are hiringWaylay - We are hiring
Waylay - We are hiring
 
Waylay building smart city solution: IoT convention Antwerp
Waylay building smart city solution: IoT convention AntwerpWaylay building smart city solution: IoT convention Antwerp
Waylay building smart city solution: IoT convention Antwerp
 
Google Cloud infrastructure in Conrad Connect by Google & waylay
Google Cloud infrastructure in Conrad Connect by Google & waylayGoogle Cloud infrastructure in Conrad Connect by Google & waylay
Google Cloud infrastructure in Conrad Connect by Google & waylay
 
Artificial intelligence by Aleksandra Pizurica
Artificial intelligence by Aleksandra PizuricaArtificial intelligence by Aleksandra Pizurica
Artificial intelligence by Aleksandra Pizurica
 
Automation is eating the world
Automation is eating the worldAutomation is eating the world
Automation is eating the world
 
LPWAN - IoT (Platform) Killer Application
LPWAN - IoT (Platform) Killer ApplicationLPWAN - IoT (Platform) Killer Application
LPWAN - IoT (Platform) Killer Application
 
Automation, intelligence and knowledge modelling
Automation, intelligence and knowledge modellingAutomation, intelligence and knowledge modelling
Automation, intelligence and knowledge modelling
 
When IoT Meets Artificial Intelligence
 When IoT Meets Artificial Intelligence When IoT Meets Artificial Intelligence
When IoT Meets Artificial Intelligence
 
Internet of Things introduction
Internet of Things introductionInternet of Things introduction
Internet of Things introduction
 
My life in one picture
My life in one pictureMy life in one picture
My life in one picture
 
WebRTC presentation
WebRTC presentationWebRTC presentation
WebRTC presentation
 
A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...
A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...
A Cloud-Based Bayesian Smart Agent Architecture for Internet-of-Things Applic...
 

Recently uploaded

263778731218 Abortion Clinic /Pills In Harare ,
263778731218 Abortion Clinic /Pills In Harare ,263778731218 Abortion Clinic /Pills In Harare ,
263778731218 Abortion Clinic /Pills In Harare ,
sisternakatoto
 
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...
VarunMahajani
 
Surat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model Safe
Surat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model SafeSurat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model Safe
Surat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model Safe
Savita Shen $i11
 
ARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTS
ARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTSARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTS
ARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTS
Dr. Vinay Pareek
 
ANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptx
ANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptxANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptx
ANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptx
Swetaba Besh
 
KDIGO 2024 guidelines for diabetologists
KDIGO 2024 guidelines for diabetologistsKDIGO 2024 guidelines for diabetologists
KDIGO 2024 guidelines for diabetologists
د.محمود نجيب
 
ACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIOND
ACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIONDACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIOND
ACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIOND
DR SETH JOTHAM
 
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
i3 Health
 
Triangles of Neck and Clinical Correlation by Dr. RIG.pptx
Triangles of Neck and Clinical Correlation by Dr. RIG.pptxTriangles of Neck and Clinical Correlation by Dr. RIG.pptx
Triangles of Neck and Clinical Correlation by Dr. RIG.pptx
Dr. Rabia Inam Gandapore
 
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 UpakalpaniyaadhyayaCharaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Dr KHALID B.M
 
Superficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptxSuperficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptx
Dr. Rabia Inam Gandapore
 
The Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of IIThe Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of II
MedicoseAcademics
 
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?Report Back from SGO 2024: What’s the Latest in Cervical Cancer?
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?
bkling
 
Are There Any Natural Remedies To Treat Syphilis.pdf
Are There Any Natural Remedies To Treat Syphilis.pdfAre There Any Natural Remedies To Treat Syphilis.pdf
Are There Any Natural Remedies To Treat Syphilis.pdf
Little Cross Family Clinic
 
POST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its managementPOST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its management
touseefaziz1
 
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists  Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists
Saeid Safari
 
Physiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of TastePhysiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of Taste
MedicoseAcademics
 
Flu Vaccine Alert in Bangalore Karnataka
Flu Vaccine Alert in Bangalore KarnatakaFlu Vaccine Alert in Bangalore Karnataka
Flu Vaccine Alert in Bangalore Karnataka
addon Scans
 
Evaluation of antidepressant activity of clitoris ternatea in animals
Evaluation of antidepressant activity of clitoris ternatea in animalsEvaluation of antidepressant activity of clitoris ternatea in animals
Evaluation of antidepressant activity of clitoris ternatea in animals
Shweta
 
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdfAlcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Dr Jeenal Mistry
 

Recently uploaded (20)

263778731218 Abortion Clinic /Pills In Harare ,
263778731218 Abortion Clinic /Pills In Harare ,263778731218 Abortion Clinic /Pills In Harare ,
263778731218 Abortion Clinic /Pills In Harare ,
 
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...
Pulmonary Thromboembolism - etilogy, types, medical- Surgical and nursing man...
 
Surat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model Safe
Surat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model SafeSurat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model Safe
Surat @ℂall @Girls ꧁❤8527049040❤꧂@ℂall @Girls Service Vip Top Model Safe
 
ARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTS
ARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTSARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTS
ARTHROLOGY PPT NCISM SYLLABUS AYURVEDA STUDENTS
 
ANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptx
ANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptxANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptx
ANATOMY AND PHYSIOLOGY OF URINARY SYSTEM.pptx
 
KDIGO 2024 guidelines for diabetologists
KDIGO 2024 guidelines for diabetologistsKDIGO 2024 guidelines for diabetologists
KDIGO 2024 guidelines for diabetologists
 
ACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIOND
ACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIONDACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIOND
ACUTE SCROTUM.....pdf. ACUTE SCROTAL CONDITIOND
 
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
New Directions in Targeted Therapeutic Approaches for Older Adults With Mantl...
 
Triangles of Neck and Clinical Correlation by Dr. RIG.pptx
Triangles of Neck and Clinical Correlation by Dr. RIG.pptxTriangles of Neck and Clinical Correlation by Dr. RIG.pptx
Triangles of Neck and Clinical Correlation by Dr. RIG.pptx
 
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 UpakalpaniyaadhyayaCharaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
Charaka Samhita Sutra sthana Chapter 15 Upakalpaniyaadhyaya
 
Superficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptxSuperficial & Deep Fascia of the NECK.pptx
Superficial & Deep Fascia of the NECK.pptx
 
The Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of IIThe Normal Electrocardiogram - Part I of II
The Normal Electrocardiogram - Part I of II
 
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?Report Back from SGO 2024: What’s the Latest in Cervical Cancer?
Report Back from SGO 2024: What’s the Latest in Cervical Cancer?
 
Are There Any Natural Remedies To Treat Syphilis.pdf
Are There Any Natural Remedies To Treat Syphilis.pdfAre There Any Natural Remedies To Treat Syphilis.pdf
Are There Any Natural Remedies To Treat Syphilis.pdf
 
POST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its managementPOST OPERATIVE OLIGURIA and its management
POST OPERATIVE OLIGURIA and its management
 
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists  Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists
Ozempic: Preoperative Management of Patients on GLP-1 Receptor Agonists
 
Physiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of TastePhysiology of Special Chemical Sensation of Taste
Physiology of Special Chemical Sensation of Taste
 
Flu Vaccine Alert in Bangalore Karnataka
Flu Vaccine Alert in Bangalore KarnatakaFlu Vaccine Alert in Bangalore Karnataka
Flu Vaccine Alert in Bangalore Karnataka
 
Evaluation of antidepressant activity of clitoris ternatea in animals
Evaluation of antidepressant activity of clitoris ternatea in animalsEvaluation of antidepressant activity of clitoris ternatea in animals
Evaluation of antidepressant activity of clitoris ternatea in animals
 
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdfAlcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
Alcohol_Dr. Jeenal Mistry MD Pharmacology.pdf
 

Remote Patient & Elderly Care Monitoring

  • 1. © 2020 www.waylay.io | page/1 Remote Patient & Elderly Care Monitoring Veselin Pizurica, CTO AI Session #6 Biotech & Pharma THU 5 November - Online session
  • 2. © 2020 www.waylay.io | page/2 How is COVID changing the perspective?
  • 3. © 2020 www.waylay.io | page/3 Remote patient monitoring
  • 4. © 2020 www.waylay.io | page/4 Sepsis Watch - Duke University https://physicians.dukehealth.org/articles/dukes-augmented-intelligence-system-helps-prevent-sepsis-ed Following a successful pilot phase from November 2018 through June 2019, Sepsis Watch, an augmented intelligence (AI) system developed by Duke, has now been deployed in the emergency departments (ED) at all three Duke University Health System hospitals to identify patients in the early stages of developing life-threatening sepsis. The system signals an AI-computed risk of sepsis so that care teams can quickly begin treatment.
  • 5. © 2020 www.waylay.io | page/5 Explainability or not? Dataset: Data were curated from the local quarternary academic hospital with over 1,000 beds and over 40,000 inpatient admissions per year. In total, the model development and evaluation dataset contained over 32 million data points. Model explainability was not prioritized, because regulations promote standardized treatment of sepsis, regardless of cause. As previously described, human experts often disagree about sepsis diagnoses and major medical and public health organizations publicly promote distinct disease definitions. The model generates risk scores every hour for every adult patient to detect sepsis. The model extended prior work using recurrent neural networks (RNNs) for clinical event detection by coupling a RNN with a multi-task gaussian processes (MGPs). Although there are emerging methods to improve explainability of RNNs, end users cannot contemplate the entire model and cannot reliably understand the relationships between model inputs and outputs. As such, there was not a deliberate effort to explain the MGP-RNN model output.
  • 6. © 2020 www.waylay.io | page/6 Using DL models in Waylay
  • 7. © 2020 www.waylay.io | page/7 Can we blindly trust DL models? https://www.nabla.com/blog/gpt-3/
  • 8. © 2020 www.waylay.io | page/8 Do we need human interpretation? https://aaas.confex.com/aaas/2017/webprogram/Paper19142.html http://nautil.us/issue/40/learning/is-artificial-intelligence-permanently-inscrutable#disqus_thread
  • 9. © 2020 www.waylay.io | page/9 Sepsis shock diagnosis https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4958885/ https://onlinelibrary.wiley.com/doi/full/10.1111/imj.14199
  • 10. © 2020 www.waylay.io | page/10 Use case: Sepsis Shock Sepsis is an abnormal inflammatory response to infection and is the leading cause of inpatient mortality within hospitals in the United States (*). Possible complications include: ● heart failure ● abnormal blood clotting ● kidney failure ● respiratory failure ● stroke ● liver failure ● loss of a portion of the bowel ● loss of portions of the extremities * Rhee, C. et al. 2017. Incidence and Trends of Sepsis in US Hospitals Using Clinical vs Claims Data, 2009-2014. JAMA. 318, 13 (Oct. 2017), 1241–1249.
  • 11. © 2020 www.waylay.io | page/11 Sepsis Shock as a rule https://www.mdcalc.com/sirs-sepsis-septic-shock-criteria
  • 12. © 2020 www.waylay.io | page/12 Explainability : Decision trees
  • 13. © 2020 www.waylay.io | page/13 Branching hell - the curse of dimensionality The depth of the tree grows linearly with the number of variables, but the number of branches grows exponentially with the number of states. Let’s consider X and Y variables with three possible outcomes: As the figure above shows, this leads to 18 leaf nodes (red/green dots) and overall 31 nodes for only two variables! We have simplified the example by assuming only subset of AND Boolean functions for a given combination. For instance, we didn't model (X is low) OR (Y is either medium or high) -> as that would be mean connecting some other branches. How many distinct decision trees do we have with 'n' Boolean attributes? Answer = #number of Boolean functions = #number of distinct truth tables with 2^n rows = 2^2^n e.g.: With 6 Boolean attributes we get 18,446,744,073,709,551,616 (2^64 which is the upper limit of an unsigned long) 13 https://www.waylay.io/blog/the-curse-of-dimensionality-in-decision-trees-branching-problem/
  • 14. © 2020 www.waylay.io | page/14 Dealing with time and majority voting? 14 This is a majority voting problem (impossible to express with decision trees)
  • 15. © 2020 www.waylay.io | page/15 Is there another way? Waylay 15
  • 16. © 2020 www.waylay.io | page/16 Demo: Sepsis shock monitoring 16
  • 17. © 2020 www.waylay.io | page/17 Demo setup 17 Veselin Steve Paul Piet
  • 18. © 2020 www.waylay.io | page/18 Demo 18
  • 19. © 2020 www.waylay.io | page/19 Learn more BLOG | DATA SHEETS | WHITE PAPERS | VIDEOS | WEBINARS | SERVICES For a free demo of the Waylay platform, reach out to sales@waylay.io or fill-in the form at waylay.io/get-demo/ and we’ll get in touch with you.