Artificial-intelligence-augmented clinical medicine

Transcription

1 Artificial-intelligence-augmented clinical medicine Klaus-Peter Adlassnig Section for Medical Expert and Knowledge-Based Systems Center for Medical Statistics, Informatics, and Intelligent Systems Medical University of Vienna Spitalgasse 23 A-1090 Vienna, Austria Einführung in Medizinische Informatik, WS 2012/13, 24. Oktober 2012

2 Artificial Intelligence (AI) Definition 1: AI is a field of science and engineering concerned with the computational understanding of what is commonly called intelligent behavior, and with the creation of artifacts that exhibit such behavior. from: Shapiro, S.C. (1992) Artificial Intelligence. In Shapiro, S.C. (ed.) Encyclopedia of Artificial Intelligence, 2nd ed., vol. 1, Wiley, New York, Definition 2: AI is the science of artificial simulation of human thought processes with computers. from: Feigenbaum, E.A. & Feldman, J. (eds.) (1995) Computers & Thought. AAAI Press, Menlo Park, back cover.

3 Artificial Intelligence applicable to clinical medicine It is the decomposition of an entire clinical thought process and its separate artificial simulation also of simple instances of clinical thought that make the task of AI in clinical medicine manageable. A functionally-driven science of AI that extends clinicians through computer systems step by step can immediately be established. artificial-intelligence-augmented clinical medicine

4 Computational intelligence in medical research medical knowledge definitional, causal, statistical, and heuristic knowledge facts consensus generalization molecular biomedicine evidence-based medicine medical studies biomolecular research medical statistics clustering and classification data and knowledge mining consensus conferences

5 Computational intelligence in patient care human-to-human patientphysician encounter AI-augmentation decision-oriented analysis and interpretation of patient data human-to-human patientphysician follow-up medical knowledge modules definitional, causal, statistical, and heuristic knowledge medical knowledge

6 Computers in clinical medicine steps of natural progression step 1: patient administration admission, transfer, discharge, and billing of medical services step 2: computerized documentation of patients medical data electronic health record: life-long, multimedia step 3: patient data retrieval and analysis at the medical institution data warehouse, research databases, study support systems, patient recruitment quality assurance and reporting step 4: knowledge-based software systems for clinical decision support safety net, quality assurance and improvement: for the individual patient and the physician and the medical institution

7 Clinical medicine medication history symptomatic therapy history data symptoms signs symptoms patient laboratory test results laboratory diagnosis signs test results differential diagnosis differential therapy prognosis patient biosignals findings images genetic data radiological diagnosis medical guidelines examination subspecialities clinic

8 Clinical medicine medication history symptomatic therapy patient history data symptoms signs MYCIN laboratory test results laboratory diagnosis QMR DxPlain CADIAG symptoms signs test results differential diagnosis + differential therapy prognosis patient biosignalsanns findings images SVMs radiological diagnosis genetic data personalized medicine + medical guidelines examination subspecialities clinic

9 Clinical medicine: high complexity sources of medical knowledge definitional causal statistical heuristic layers of medical knowledge observational and measurement level interpretation, abstraction, aggregation, summation pathophysiological states diseases/diagnoses, therapies, prognoses, management decisions imprecision, uncertainty, and incompleteness imprecision (=fuzziness) of medical concepts due to the unsharpness of boundaries of linguistic concepts uncertainty of medical conclusions due to the uncertainty of the occurrence and co-occurrence of imprecise medical concepts incompleteness of medical data and medical theory due to only partially known data and partially known explanations for medical phenomena gigantic amount of medical data and medical knowledge patient history, physical examination, laboratory test results, clinical findings symptom-disease relationships, disease-therapy relationships, terminologies, ontologies: SNOMED CT, LOINC, UMLS, spezialisation, teamwork, quality management, computer support

10 studies in Colorado and Utah and in New York (1997) errors in the delivery of health care leading to the death of as many as 98,000 US citizens annually causes of errors error or delay in diagnosis failure to employ indicated tests errors tests or therapy use of outmoded failure to act on results of testing or monitoring error in the performance of a test, procedure, or operation error in administering the treatment error in the dose or method of using a drug avoidable delay in treatment or in responding to an abnormal test prevention inappropriate (not indicated) care failure of communication equipment failure prevention of errors we must systematically design safety into processes of care

11 Medical information and knowledge-based systems patient s medical data symptoms signs test results clinical findings biosignals images diagnoses therapies nursing data standardization telecommunication chip cards information systems medical statistics clustering & classification data & knowledge mining machine learning induction single patient many patients general knowledge clinical decision support medical expert systems deduction general knowledge diagnosis therapy prognosis management physician s medical knowledge anatomy biochemistry physiology pathophysiology pathology nosology therapeutic knowledge disease management medical knowledge acquisition by clinicians + knowledge engineers subjective experience intuition knowledge-based systems telemedicine integration telemedicine

12 Clinical decision support systems patients structured medical data diagnostic support clinical alerts, reminders, calculations data interpretation, (tele)monitoring differential diagnostic consultation rare diseases, rare syndromes further or redundant investigations pathological signs accounted for consensus-criteria-based evaluation definitions classification criteria prognostic prediction illness severity scores, prediction rules trend detection and visualization therapy advice drug alerts, reminders, calculations indication, contraindications, redundant medications, substitutions adverse drug events, interactions, dosage calculations, consequent orders management of antimicrobial therapies (open-loop) control systems patient management guidelines guideline-based reminders computerized clinical guidelines, protocols, SOPs high-level patient and hospital analytics

13 according to Kensaku Kawamoto, University of Utah, 2012: A Holy Grail of clinical informatics is scalable, interoperable clinical decision support. What have we done?

14 Interpretation von Hepatitis-Serologie-Befunden

15 test results interpretation

16 ORBIS Experter: Hepatitis serology diagnostics

17 Interpretation of hepatitis serology test results

18

19 Differentialdiagnose rheumatischer Erkrankungen

20

21 Computergestützte Entwöhnung vom Respirator

22

23 Personalized clinical decision support patient medical data history physical signs lab tests clinical findings + genomic data present (personalized) CDS future personalized CDS unavoidable, more specific diagnostics, extends the realm of therapy

24 Solution at the Vienna General Hospital hospital wards HIS departments knowledge Arden Syntax development & test environment Arden Syntax server HIS PDMS units clinical laboratories LIS Extended documentation & research data base interfaces data & knowledge services center knowledge results data Arden Syntax rule engine genomic data laboratories LIS data & concept mining

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

26 Arden Syntax, Arden Syntax server, and health care information systems integration HIS, MIS, PDMS, LIS, medical practice SW, web-based EHR, telemedicine applications, health portals, service-oriented *data & knowledge services center operational: - harmonized input data - Arden Syntax MLMs - collected reasoning data exploratory: - rule learning/tuning - data and concept mining * web-based functionality reminders and alerts, monitoring, surveillance, diagnostic and therapeutic decision support,

27 Arden Syntax server and software components Arden Syntax integrated development and test environment (IDE) including - Medical logic module (MLM) editor and authoring tool - Arden Syntax compiler (syntax versions 2.1, 2.5, 2.6, 2.7, and 2.8) - Arden Syntax engine - MLM test environment - MLM export component command-line Arden Syntax compiler health care information system results reporting tools knowledge administration data Arden Syntax development & test environment knowledge Arden Syntax server 1) interfaces 2) data & knowledge services center 1) integrated, local, or remote 2) local and web services, web frontend knowledge results Arden Syntax data rule engine web-services-based Arden Syntax server including - Arden Syntax engine - MLM manager - XML-protocol-based interfaces, e.g., SOAP, REST, and HL7 - a project-specific data and knowledge services center may be hosted Java libraries - Arden Syntax compiler - Arden Syntax engine Fuzzy Arden Syntax (version 2.9) extension to fuzzy sets, operators, statements, and parallel execution

28 Crisp sets vs. fuzzy sets Y young threshold age U = [0, 120] yes/no decision Y U with Y = {( Y (x)/ x) x U} Y : U {0, 1} Y (x) = 0 x > threshold 1 x threshold x U Y young threshold age gradual transition U = [0, 120] Y U with Y = {(µ Y (x)/ x) x U} µ Y : U [0, 1] 1 x > threshold µ 1 + (0.04 x) 2 Y (x) = x U 1 x threshold

29 Is everything a question of definition? Y 1 young arbitrary yes/no decisions cause of unfruitful discussions often simply wrong 0 0 threshold age Y 1 young intuitive gradual transitions 0 0 threshold age

30 Degree of compatibility [= degree of membership] A (x) 1.00 highly decreased decreased normal increased highly increased µ (x) = µ (x) = 0.18 µ (x) = 0.00 µ (x) = 0.00 µ (x) = x [mg/dl] glucose level in serum of 130 mg/dl

31 Service-oriented, scalable, and interoperable clinical decision support

32 Towards a science of clinical medicine patient s medical data and healthcare processes for human processing observations e.g., temperature chart skin color (jaundice, livid, ) patient s medical data and healthcare processes for machine processing measurements e.g., CRP color measurement Measure what is measurable, and make measurable what is not so. Crucial point in clinical medicine: Digitize what is digitizable, and make digitizable what is not so. Galileo Galilei Klaus-Peter Adlassnig

33 The medical world becomes flat after Thomas L. Friedman The World is Flat. Penguin Books, in a local world decision support will be part of clinical information systems in a global world any activity where we can digitize and decompose the value chain*, and move the work around will get moved around * patient value chain: patient examination, diagnosis, therapy, prognosis, health care decisions, patient care

34 Future of artificial-intelligence-augmented clinical medicine today tomorrow predictable future medical data collection, storage, & distribution clinical decision support personalized medicine implants, prostheses, robotics HIS 1.0 HIS 2.0

35 Closing remark: formalism vs. reality Pure mathematics is much easier to understand, much simpler, than the messy real world! Gregory Chaitin (2005) Meta Math!: The Quest for Omega, Pantheon Books, New York. Clinical informatics deals with the messy real patient.