medexter clinical decision support ARDENSUITE: Arden-Syntax-based genuine technology platform for clinical decision support (CDS)

Size: px

Start display at page:

Download "medexter clinical decision support ARDENSUITE: Arden-Syntax-based genuine technology platform for clinical decision support (CDS)"

Gwendolyn Perry
10 years ago
Views:

1 medexter clinical decision support ARDENSUITE: Arden-Syntax-based genuine technology platform for clinical decision support (CDS) I: Web services for MLM calling and for data transfer II: Web services for MLM calling and server/database connector for data access III: Data warehouse + Arden Syntax server = autonomous CDS system

calling and for data transfer II: Web services for MLM calling and

2 Integration into i.s.h.med at the Vienna General Hospital SOP checking in melanoma patients receiving chemotherapy

3 medexter clinical decision support University of Colorado Health with Epic EHR Heart failure readmission risk score (HFRRS) Input: vital signs lab data demographics ATD info ICD codes 2014 Epic Systems Corporation. Used with permission. Example of from HFRRS MLM to HF nurse practitioners: patient follow-up and authorization of additional inpatient services (e.g., occupational and physical therapy)

Systems Corporation. Used with permission.

4 medexter clinical decision support Moni-ICU and Moni-NICU Monitoring (for surveillance and alerts) of healthcare-associated infections (HAIs) in ICUs with adult patients and in NICUs with neonatal patients Characteristics (1) PDMSs and LISs as electronic data sources provide structured medical data (2) medical knowledge bases containing computerized knowledge of every clinical entity involved (3) processing algorithms evaluate, aggregate, and interpret clinical data stepwise until data can be mapped into the given HAI definitions

data sources provide structured medical data (2) medical knowledge bases containing computerized knowledge of every clinical entity

5 Translation of HAI definitions into IT terminology example: bloodstream infections (BSIs) HELICS-protocol HAI in ICU, version 6.1, Sep IT statement in free language Recognized pathogen OR clinical signs AND growth of same skin contaminant from two separate blood samples OR clinical signs AND growth of same skin contaminant from blood AND intravascular line OR clinical signs AND positive antigen test from blood

2004 IT statement in free language Recognized pathogen OR clinical signs AND growth of same skin

6 medexter clinical decision support A bloodstream infection with clinical signs and growth of same skin contaminant from two separate blood samples BSI-A2 1 clinical_signs_of_bsi (t-1d, t, t+1d) same_skin_contaminant_from_two_separate_blood_samples

two separate blood samples BSI-A2 1 clinical_signs_of_bsi

7 Decomposition clinical signs clinical_signs_of_bsi (t-1d, t, t+1d)[yesterday, today, tomorrow] = fevert (t-1d) hypotension (t-1d) clinical_signs_of_bsi (t-1d) = leucopenia (t-1d) leucocytosis (t-1d) CRP increased (t-1d) fevert (t) hypotension (t) clinical_signs_of_bsi (t) = leucopenia (t) leucocytosis (t) CRP increased (t) fevert (t+1d) hypotension (t+1d) clinical_signs_of_bsi (t+1d) = leucopenia (t+1d) leucocytosis (t+1d) CRP increased (t+1d)

fevert (t) hypotension (t) clinical_signs_of_bsi (t) = leucopenia (t) leucocytosis (t) CRP increased (t) fevert

8 Linguistic uncertainty defined by fuzzy sets example: fever medexter clinical decision support fevert (t-1d)... fever 1 data import intensive care unit fevert (t) C maximum value of the day e.g., 38.5 C thermoregulation applied fevert (t+1d)...

.. fever 1 data import intensive care unit fevert (t) 0 37 37.

9 Decomposition skin contaminant first blood culture - coagulase-negative staphylococci - Micrococcus sp. - Propionibacterium acnes - Bacillus sp. - Corynebacterium sp. same_skin_contaminant_from_ two_separate_blood_samples (within 48 hours) data import microbiology second blood culture - coagulase-negative staphylococci - Micrococcus sp. - Propionibacterium acnes - Bacillus sp. - Corynebacterium sp.

same_skin_contaminant_from_ two_separate_blood_samples (within 48 hours) data import microbiology

10 patient-specific cockpit & legal reporting & quality benchmarking layer n (goal) layer n-y layer n-x-y linguistic HAI definitions y inference steps intermediate concepts: pathophysiological states x inference steps basic concepts: symptoms, signs, test results, clinical findings reasoning symbols layer n-x-y-1 layer 2 layer 1 abstraction: rules, type-1 & type-2 fuzzy sets, temporal abstraction feature extraction: mean values, scores, preprocessing: missing data, plausibility, data-to-symbol conversion layer 0 (start) ICU, NICU, and microbiology patient data bases raw data

reasoning symbols layer n-x-y-1 layer 2 layer 1 abstraction: rules, type-1 & type-2 fuzzy sets, temporal abstraction feature extraction: mean

12 Moni-ICU cockpit

14 Automated interpretation of hepatitis serology test results

15 Automated interpretation of hepatitis serology test results includes frequent, rare, as well as inconsistent combinations complete coverage of the problem domains e.g., hepatitis B serology: about 150 rules in 3 layers for more than 61,000 possible combinations

16 ARDENSUITE-Integration am Universitätsklinikum Erlangen Integration der kommerziellen ARDENSUITE-Engine (Medexter Healthcare) MLM-Ausführung daten-, zeit- und benutzergesteuert (Viewer) based on Stefan Kraus

17 by Stefan Kraus

18 Use Case: Hypoglycemia Hypoglycemia may seriously harm If patient is unconscious, it is difficult to notice The PDMS should actively notify the physician: If glucose is less than 50mg/dl, then send an SMS message to the physician. by Stefan Kraus

should actively notify the physician: If glucose is less than

19 Use Case: Hypoglycemia DATA: LET glucose BE READ { glucose }; LET physician_dect BE DESTINATION {sms:26789}; LOGIC: IF LATEST glucose IS LESS THAN 50 THEN CONCLUDE true; ENDIF; CONCLUDE TRUE Do something ACTION: WRITE Warning AT physician_dect; by Stefan Kraus

glucose IS LESS THAN 50 THEN CONCLUDE true; ENDIF; CONCLUDE TRUE

20 Use Case: Hypoglycemia Event monitors are tireless observers, constantly monitoring clinical events (George Hripscak) by Stefan Kraus

21 medexter clinical decision support To summarize ARDENSUITE software: versatile, scalable, data- and knowledge-processing software for CDS and quality measures High integratability through web services and database connectors Cockpit monitoring of and dashboard analytics for adverse events Reporting and quality benchmarking of adverse events Users: patient-care institutions, healthcare and research institutions, health IT companies, and consumers

medexter clinical decision support

medexter clinical decision support medexter Arden Syntax training course Klaus-Peter Adlassnig and Karsten Fehre Medexter Healthcare Borschkegasse 7/5 A-1090 Vienna www.medexter.com Arden Syntax training course, Vienna, 17 June 2015 Computers