Smart Data for Smart Doctors: How Algorithms Support ... · Tools Real-time Analysis App-spanning...
Transcript of Smart Data for Smart Doctors: How Algorithms Support ... · Tools Real-time Analysis App-spanning...
Smart Data for Smart Doctors:
How Algorithms Support Fighting Complex Diseases
Dr.-Ing. Matthieu-P. Schapranow
3rd Int’l Symposium Big Data in Medicine, Potsdam
Nov 20, 2017
■ Can we enable clinicians to take their therapy decisions:
□ Incorporating all available patient specifics,
□ Referencing latest lab results and worldwide medical knowledge, and
□ In an interactive manner during their ward round?
Our Motivation
Turn Precision Medicine Into Clinical Routine
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
2
Schapranow, Big Data
in Medicine, Nov 20,
2017
The Challenge
Distributed Heterogeneous Data Sources
3
Human genome/biological data
600GB per full genome
15PB+ in databases of leading institutes
Prescription data
1.5B records from 10,000 doctors and
10M Patients (100 GB)
Clinical trials
Currently more than 30k
recruiting on
ClinicalTrials.gov
Human proteome
160M data points (2.4GB) per sample
>3TB raw proteome data in ProteomicsDB
PubMed database
>23M articlesHospital information systems
Often more than 50GB
Medical sensor data
Scan of a single organ in 1s
creates 10GB of raw data
Cancer patient records
>160k records at NCT
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
Schapranow, Big Data
in Medicine, Nov 20,
2017Routine Data
Treatments and cost
reimbursements
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
4
Schapranow, Big Data
in Medicine, Nov 20,
2017
Our Approach: AnalyzeGenomes.com
In-Memory Computing Platform for Big Medical Data
5In-Memory Computing Platform
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
+
++
++
v
+++
t
SQL
xx
T
Schapranow, Big Data
in Medicine, Nov 20,
2017
Our Approach: AnalyzeGenomes.com
In-Memory Computing Platform for Big Medical Data
6In-Memory Computing Platform
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
Schapranow, Big Data
in Medicine, Nov 20,
2017
Our Approach: AnalyzeGenomes.com
In-Memory Computing Platform for Big Medical Data
7In-Memory Computing Platform
Combined and Linked Data
Genome
Data
Cellular
Pathways
Genome
MetadataResearch
Publications
Pipeline and
Analysis Models
Drugs and
Interactions
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
IndexedSources
Schapranow, Big Data
in Medicine, Nov 20,
2017
Our Approach: AnalyzeGenomes.com
In-Memory Computing Platform for Big Medical Data
8In-Memory Computing Platform
Extensions for Life Sciences
Data Exchange,
App Store
Access Control,
Data Protection
Fair Use
Statistical
Tools
Real-time
Analysis
App-spanning
User Profiles
Combined and Linked Data
Genome
Data
Cellular
Pathways
Genome
MetadataResearch
Publications
Pipeline and
Analysis Models
Drugs and
Interactions
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
IndexedSources
Schapranow, Big Data
in Medicine, Nov 20,
2017
Our Approach: AnalyzeGenomes.com
In-Memory Computing Platform for Big Medical Data
9In-Memory Computing Platform
Extensions for Life Sciences
Data Exchange,
App Store
Access Control,
Data Protection
Fair Use
Statistical
Tools
Real-time
Analysis
App-spanning
User Profiles
Combined and Linked Data
Genome
Data
Cellular
Pathways
Genome
MetadataResearch
Publications
Pipeline and
Analysis Models
Drugs and
Interactions
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
Drug Response
AnalysisPathway Topology
AnalysisMedical
Knowledge
Cockpit
Oncolyzer
Clinical Trial
Recruitment
Cohort
Analysis
...
IndexedSources
Heart Failure
Sleeping disorder
Fibrosis
Blood pressure
Blood volume
Gene ex-pression
Hyper-trophyCalcium
meta-bolism
Energy meta-bolism
Iron deficiency
Vitamin-D deficiency
Gender
Epi-genetics
■ Integrated systems medicine based on
real-time analysis of healthcare data
■ Funding period: Mar 2015 – Dec 2018
■ Funded consortium partners:
Use Case:
Systems Medicine Model of Heart Failure (SMART)
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
10
A R T
+
T R A M
S
+
S M
■ Patient: 63 years, male, smoker, chronic heart insufficiency, stage III-IV
1. Appointment I: Acquire systemic patient details, e.g. physiological and
blood markers
2. Predict patient-specific surgery outcome using clinical model
3. Select adequate surgery option and conduct valve replacement
4. Equip patient with sensors to allow regular monitoring
5. Appointment II: 6 weeks after surgery to validate outcome
Establish Systems Medicine Model for
Improved Treatment of Heart Failure
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
11
■ Process definition together with
subject matter experts
■ Identification of time-consuming steps
■ Key contributions
□ Sharing of data
□ Improved communication
□ Reproducible data processing
□ Data analysis applications, e.g. for
interactive hypothesis validation
Requirements Engineering for System Medicine
Computer-aided Systems Medicine Process
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
12
■ In-Memory Database (IMDB)
as data integration platform
■ Integration of partner data to
provide a holistic patient view
■ Event-driven process
notifications
■ Specific real-time analysis
workflows, e.g. for
RNAseq data
Integrated Systems Medicine
Data Processing Platform
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
SMART IT Platform
Event
Notification
Data
AcquisitionData artifactData Artifacts
Statistical Data Analysis
Methods, Machine Learning,
Bioinformatics Tools
Analysis
Applications
Sync Client
Researcher/
Clinician
Specific Analysis
Applications
In-Memory Database
User and Event
Data
Hemodynamic
Parameters
Clinical Data
Omics Data
Sync Client
Sync Server
Importer for
non-NGS Data
Importer for
FASTQ Data
Modeling
Parameters
13
SMART IT Platform
Event
Notification
Data
AcquisitionData artifactData Artifacts
Statistical Data Analysis
Methods, Machine Learning,
Bioinformatics Tools
Analysis
Applications
Sync Client
Researcher/
Clinician
Specific Analysis
Applications
In-Memory Database
User and Event
Data
Hemodynamic
Parameters
Clinical Data
Omics Data
Sync Client
Sync Server
Importer for
non-NGS Data
Importer for
FASTQ Data
Modeling
Parameters
Reproducible Data Processing:
Transfer of Processing Logic to Platform
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
■ Example: Flexible protocol for processing of RNAseq data
■ Standardized model using Business Process Modeling and Notation (BPMN)
■ Use of standardized pipeline models enable reproducible researchRNA Seq Analysis_V2
TopHat
Trimmomatic
FASTQC
STAR
featureCounts
Counts Matrix
BAM-FileAligned Reads
FASTQC 2
FASTQ -Trimmed Reads
Pre-TrimmingQC-Report
FASTQ - Reads
Post-AlignmentQC-Report
14
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
15
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
16
■ Interdisciplinary partners collaborate on enabling interactive health research
■ Current funding period: Aug 2015 – July 2018
■ Funded consortium partners:
□ AOK
German healthcare insurance company
□ data experts group
Technology operations
□ Hasso Plattner Institute
Real-time data analysis, in-memory database technology
□ Technology, Methods, and Infrastructure for Networked Medical Research
Legal and data protection
Use Case:
Smart Analysis Health Research Access (SAHRA)
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
17
Elderly Care Planning in Mecklenburg-Vorpommern
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
18
Geospational Data Exploration
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
19
■ Stratification of patient cohorts using patient specifics
■ Automatic matching of similar patients and patient anamnesis
■ Interactive graphical exploration of longitudinal patient data
Stratification of Hypertension Patients
and Longitudinal Data Analysis
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
20
■ For patients
□ Support education and sovereignty of patients
□ Identify patient-specific details, e.g. clinical trials or medical experts
■ For clinicians
□ Identify similar patient cases, e.g. to find adequate therapy
□ Analyze pharmacokinetic correlations
■ For researchers
□ Enable real-time analysis of medical data, e.g. assess biological
pathways to evaluate impact of genetic variants
□ Analysis of structured and unstructured data, e.g. publications,
diagnosis, and EMR data
What to Take Home?
Test-drive it yourself: https://we.AnalyzeGenomes.com
Schapranow, Big Data
in Medicine, Nov 20,
2017
21
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
Stay in Contact!
Dr.-Ing. Matthieu-P. Schapranow
Program Manager E-Health & Life Sciences
Hasso Plattner Institute
August-Bebel-Str. 88
14482 Potsdam, Germany
we.analyzegenomes.com
@AnalyzeGenomes
Schapranow, Big Data
in Medicine, Nov 20,
2017
Smart Data for Smart
Doctors: How
Algorithms Support
Fighting Complex
Diseases
22