Hans-Peter Eich

Decision support in acute abdominal pain using an expert system for different knowledge bases

This paper describes a knowledge-based system for the diagnosis of acute abdominal pain, in which scores and rule sets are integrated. The system is linked to a documentation program via a medical data dictionary and allows an on-line application of knowledge modules to clinical data. Different rule sets were generated by automatic rule generation (C4.5) from a prospective database. The rule sets and two published diagnostic scores were evaluated on a test set, resulting in a diagnostic accuracy of 57% for a general knowledge module and between 44 and 88% for specific knowledge modules. The program is fully functioning and has been evaluated carefully in 14 German hospitals.

Internet-Based Decision-Support Server for Acute Abdominal Pain

The paper describes conception and prototypical design of a decision-support server for acute abdominal pain. A user survey was initiated in three surgical departments to assess the user requirements concerning formal decision-aids. The results of this survey are presented. For scoring systems a work-up to separate terminological information from structure is described. The terminology is separately stored in a data dictionary and the structure in the knowledge base. This procedure enables a reuse of terminology for documentation and decision-support. The whole system covers a decision-support server written in C++ with underlying data dictionary and knowledge base, a documentation module written in Java and a Corba middleware (ORBacus 3.1) that establishes a connection via internet.

Does Size Really Matter—Using a Decision Tree Approach for Comparison of Three Different Databases from the Medical Field of Acute Appendicitis

Decision trees have been successfully used for years in many medical decision making applications. Transparent representation of acquired knowledge and fast algorithms made decision trees one of the most often used symbolic machine learning approaches. This paper concentrates on the problem of separating acute appendicitis, which is a special problem of acute abdominal pain, from other diseases that cause acute abdominal pain by use of an decision tree approach. Early and accurate diagnosing of acute appendicitis is still a difficult and challenging problem in everyday clinical routine. An important factor in the error rate is poor discrimination between acute appendicitis and other diseases that cause acute abdominal pain. This error rate is still high, despite considerable improvements in history-taking and clinical examination, computer-aided decision-support, and special investigation such as ultrasound. We investigated three databases of different size with ca ses of acute abdominal pain to complete this task as successful as possible. The results show that the size of the database does not necessary directly influence the success of the decision tree built on it. Surprisingly we got the best results from the decision trees built on the smallest and the biggest database, where the database with medium size (relative to the other two) was not so successful. Despite this we were able to produce decision tree classifiers that were capable of producing correct decisions on test data sets with accuracy up to 84%, sensitivity to acute appendicitis up to 90%, and specificity up to 80% on the same test set.

Internet-based decision-support server for acute abdominal pain

The paper describes conception and prototypical design of a decision-support server for acute abdominal pain. Existing formal methods to develop and exchange scores, guidelines and algorithms are used for integration. For scoring systems a work-up to separate terminological information from structure is described. The terminology is separately stored in a data dictionary and the structure in a knowledge base. This procedure enables a reuse of terminology for documentation and decision-support. The whole system covers a decision-support server written in C++ with underlying data dictionary and knowledge base, a documentation module written in Java and a CORBA middleware that establishes a connection via Internet.

Cross-learning on multiple databases in the case of acute appendicitis

We study the cross-learning approach on multiple databases to predict acute appendicitis. For the machine learning algorithm, our evolutionary method for inducing decision trees is used. The results of cross-learning are presented for the three different databases obtained in international projects regarding acute abdominal pain.

Smartie: Smart Medical Applications Repository of Tools for Informed Expert Decision

SMARTIE is part of the Information Society Technologies research program funded by the European Commission. The project has a total duration of 30 months, ending June 2003. Project partners include experts in all aspects of computerised decision support development, from research and clinical, through software and systems development, to quality management and dissemination. Decision support is being developed in several medical specialties: Gastroenterology, Endocrinology, Emergency Medicine and Intensive Care Medicine, Paediatrics, Nutrition and Quality of Life with plans to expand to other specialties in the future.

Entscheidungshilfen für die Kitteltasche: Neue Strategien zur Implementierung computergestützte Entscheidungshilfen in der Gastroenterologie, Endokrinologie und Intensivstation

Einleitung: Ein Vielzahl von Werkzeugen zur Unterstutzung der medizinischen Entscheidungsfindung wurden bisher entwickelt. Diese reichen von einfachen Scores zur Risikoabschatzung bis hin zu umfangreichen Programmen (z. B. fur akute Bauchschmerzen). Trotz eines nachgewiesenen klinischen Nutzens konnten sich diese Werkzeuge bisher nicht in der klinischen Routine durchsetzen. Ein Hauptgrund hierfur ist, dass diese Werkzeuge dem Arzt nicht zur Verfugung stehen, wenn sie benotigt werden. In einem durch die Europaische Kommission geforderten Projekt (SMARTIE — Smart Medical Applications Repository for Informed Expert Decision) wurde ein neuer Ansatz gewahlt, um Werkzeuge zur Unterstutzung der medizinischen Entscheidungsfindung in den Bereichen Gastroenterologie, Endokrinologie und Intensivstation einzufuhren und den Arzt mit kritischen Informationen zur Behandlung des Patienten zu versorgen. Methodik: Die Wissensakquisition in SMARTIE erlaubt eine Auswahl der best moglichen Werkzeuge. Sie basiert auf einer umfangreichen Literaturrecherche und Konsultation von fuhrenden klinischen Experten. Internationale Experten in speziellen klinischen Gebieten aus weit-weiten und europaischen Organisationen zertifizieren das verwendete Wissen (klinische Studien, klinische Leitlinien, Uberlebenstabellen, Prognosescores, Risikofaktor-Analyse, Anwendungen zur Entscheidungsunterstutzung). Eine einheitliche Benutzerlobflache wird uber unterschiedliche Computersysteme und Plattformen beibehalten. Dies sind Anwendungen fur PDA (z.B. Palm, Compaq), webbasierte und PC-basierte Anwendungen. Nach einer Evaluation der Funktionalitat und der Nutzlichkeit werden medizinische Nutzergruppen diese Programme (genannt MedNotes) bewerten. Klinische Studien werden in ausgewahlten Zentren durchgefuhrt. Alle entwickelten Programme werden uber Internet kostenlos zur Verfugung gestellt (Open-Source Lizenz). Ergebnisse: Das Projekt hat eine Laufzeit von 30 Monaten (01.01.2001 bis 31.06.2003). Bis jetzt wurde eine Liste von Werkzeugen zur medizinischen Entscheidungsfindung zusammengestellt und von klinischen Experten bewertet. Erste Prototypen wurden entwickelt und befinden sich in der Testung durch klinische Nutzergruppen. Ein besonderer Schwerpunkt wird auf die Benutzerfreundlichkeit der Anwendungen gelegt. Ein Uberblick uber die Wissensakquisition, Entwicklung und Bewertung durch klinische Nutzer wird in der Prasentation gegeben. Informationen zum Projekt SMARTIE sind uber http://www.smartie-ist.org ab November 2001 abrufbar. Danksagung: Das Projekt wird gefordert durch die Europaische Kommission (IST-2000-25429 im funften Rahmenprogramm).