Article
Cognition Guided Surgery: Towards comprehensive integrated information processing as clinical decision support in malignant neoplasms of the liver
Search Medline for
Authors
-
Hannes Götz Kenngott
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
-
Mohammadreza Hafezi
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
-
Keno März
- Deutsches Krebsforschungszentrum Heidelberg, Junior group computer-assisted interventions, Heidelberg, Deutschland
-
Arash Saffari
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
-
Matthias Eisenmann
- Deutsches Krebsforschungszentrum Heidelberg, Junior group computer-assisted interventions, Heidelberg, Deutschland
-
Sandy Engelhardt
- Deutsches Krebsforschungszentrum Heidelberg, Junior group computer-assisted interventions, Heidelberg, Deutschland
-
Rudi Studer
- Karlsruher Institut für Technologie, Institut für Angewandte Informatik und Formale Beschreibungsverfahren, Karlsruhe, Deutschland
-
Alfred Franz
- Deutsches Krebsforschungszentrum Heidelberg, Junior group computer-assisted interventions, Heidelberg, Deutschland
-
Achim Rettinger
- Karlsruher Institut für Technologie, Institut für Angewandte Informatik und Formale Beschreibungsverfahren, Karlsruhe, Deutschland
-
Marco Nolden
- Deutsches Krebsforschungszentrum Heidelberg, Junior group computer-assisted interventions, Heidelberg, Deutschland
-
Maria Maleshkova
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
-
Lena Maier-Hein
- Deutsches Krebsforschungszentrum Heidelberg, Junior group computer-assisted interventions, Heidelberg, Deutschland
-
Martin Wagner
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
-
Beat Peter Müller
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
-
Arianeb Mehrabi
- Universitätsklinikum Heidelberg, Allgemein-, Viszeral- und Transplantationschirurgie, Heidelberg, Deutschland
| Published: | April 21, 2016 |
|---|---|
| Published with erratum: | September 13, 2016 |
Outline
Text
Background: Malignant neoplasms of the liver are among the most frequent cancers worldwide. It is hard to select the right therapy because the ideal treatment depends on a multitude of parameters including patient condition, tumor size and location, liver function, and previous interventions. Computer-assisted treatment strategy planning and risk stratification based on holistic processing of patient-individual data, practical knowledge (i.e., case knowledge), and factual knowledge (e.g., clinical guidelines and studies) could help physicians to find the right treatment for each individual patient.
Materials and methods: First a technical infrastructure that enables storing, accessing, and processing of heterogeneous data to support clinical decision making was built. Then a formalized dynamic patient model and framework that incorporates all the heterogeneous data acquired for a specific patient in the whole course of disease treatment was established. All was integrated in a comprehensive framework with the additional concept for formalizing factual knowledge. 213 relevant publications (studies, guidelines, meta-analyses) on hepatic liver resections were exemplarily extracted searching medline. Rules were extracted on the basis of the major findings in these publications and formalized in the framework. In addition 184 exemplary cases of liver resection were analyzed and uploaded into the system as structured and semantically annotated datasets.
Results: A patient model that covered 602 patient-individual parameters was successfully instantiated for 184 patients. It served as basis for the formalization of 72 rules extracted from studies on patients with colorectal liver metastases or hepatocellular carcinoma. For a subset of 70 patients with these diagnoses, the system derived an average of 37 ± 15 assertions per patient. Patient individual prognosis and recommendation was possible to obtain using this system.
Conclusion: The proposed concept paves the way for integrated, automated computer-assisted treatment strategy planning by enabling joint storing and processing of heterogeneous data from various information sources. The individual patient and its treatment strategy could be valued according to guidelines, studies and meta-analyses as well as a hospital specific cohort of patients with similar operations in terms of prognosis and assessment of best surgical strategy.
