TY - CHAP A1 - Jannidis, Fotis A1 - Reger, Isabella A1 - Weimer, Lukas A1 - Krug, Markus A1 - Puppe, Frank T1 - Automatische Erkennung von Figuren in deutschsprachigen Romanen N2 - Eine wichtige Grundlage für die quantitative Analyse von Erzähltexten, etwa eine Netzwerkanalyse der Figurenkonstellation, ist die automatische Erkennung von Referenzen auf Figuren in Erzähltexten, ein Sonderfall des generischen NLP-Problems der Named Entity Recognition. Bestehende, auf Zeitungstexten trainierte Modelle sind für literarische Texte nur eingeschränkt brauchbar, da die Einbeziehung von Appellativen in die Named Entity-Definition und deren häufige Verwendung in Romantexten zu einem schlechten Ergebnis führt. Dieses Paper stellt eine anhand eines manuell annotierten Korpus auf deutschsprachige Romane des 19. Jahrhunderts angepasste NER-Komponente vor. KW - Digital Humanities KW - Figurenerkennung KW - Named-Entity-Recognition KW - Domänenadaption KW - Literatur Y1 - 2015 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-143332 UR - https://dhd2015.uni-graz.at/ ER - TY - JOUR A1 - Gehrke, Alexander A1 - Balbach, Nico A1 - Rauch, Yong-Mi A1 - Degkwitz, Andreas A1 - Puppe, Frank T1 - Erkennung von handschriftlichen Unterstreichungen in Alten Drucken JF - Bibliothek Forschung und Praxis N2 - Die Erkennung handschriftlicher Artefakte wie Unterstreichungen in Buchdrucken ermöglicht Rückschlüsse auf das Rezeptionsverhalten und die Provenienzgeschichte und wird auch für eine OCR benötigt. Dabei soll zwischen handschriftlichen Unterstreichungen und waagerechten Linien im Druck (z. B. Trennlinien usw.) unterschieden werden, da letztere nicht ausgezeichnet werden sollen. Im Beitrag wird ein Ansatz basierend auf einem auf Unterstreichungen trainierten Neuronalen Netz gemäß der U-Net Architektur vorgestellt, dessen Ergebnisse in einem zweiten Schritt mit heuristischen Regeln nachbearbeitet werden. Die Evaluationen zeigen, dass Unterstreichungen sehr gut erkannt werden, wenn bei der Binarisierung der Scans nicht zu viele Pixel der Unterstreichung wegen geringem Kontrast verloren gehen. Zukünftig sollen die Worte oberhalb der Unterstreichung mit OCR transkribiert werden und auch andere Artefakte wie handschriftliche Notizen in alten Drucken erkannt werden. N2 - The recognition of handwritten artefacts like underlines in historical printings allows inference on the reception and provenance history and is necessary for OCR (optical character recognition). In this context it is important to differentiate between handwritten and printed lines, since the latter are common in printings, but should be ignored. We present an approach based on neural nets with the U-Net architecture, whose segmentation results are post processed with heuristic rules. The evaluations show that handwritten underlines are very well recognized if the binarisation of the scans is adequate. Future work includes transcription of the underlined words with OCR and recognition of other artefacts like handwritten notes in historical printings. T2 - Recognition of handwritten underlines in historical printings KW - Brüder Grimm Privatbibliothek KW - Erkennung handschriftlicher Artefakte KW - Convolutional Neural Network KW - regelbasierte Nachbearbeitung KW - Grimm brothers personal library KW - handwritten artefact recognition KW - convolutional neural network KW - rule based post processing Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-193377 SN - 1865-7648 SN - 0341-4183 N1 - Dieser Beitrag ist mit Zustimmung des Rechteinhabers aufgrund einer (DFG-geförderten) Allianz- bzw. Nationallizenz frei zugänglich. VL - 43 IS - 3 SP - 447 EP - 452 ER - TY - JOUR A1 - Toepfer, Martin A1 - Corovic, Hamo A1 - Fette, Georg A1 - Klügl, Peter A1 - Störk, Stefan A1 - Puppe, Frank T1 - Fine-grained information extraction from German transthoracic echocardiography reports JF - BMC Medical Informatics and Decision Making N2 - Background Information extraction techniques that get structured representations out of unstructured data make a large amount of clinically relevant information about patients accessible for semantic applications. These methods typically rely on standardized terminologies that guide this process. Many languages and clinical domains, however, lack appropriate resources and tools, as well as evaluations of their applications, especially if detailed conceptualizations of the domain are required. For instance, German transthoracic echocardiography reports have not been targeted sufficiently before, despite of their importance for clinical trials. This work therefore aimed at development and evaluation of an information extraction component with a fine-grained terminology that enables to recognize almost all relevant information stated in German transthoracic echocardiography reports at the University Hospital of Würzburg. Methods A domain expert validated and iteratively refined an automatically inferred base terminology. The terminology was used by an ontology-driven information extraction system that outputs attribute value pairs. The final component has been mapped to the central elements of a standardized terminology, and it has been evaluated according to documents with different layouts. Results The final system achieved state-of-the-art precision (micro average.996) and recall (micro average.961) on 100 test documents that represent more than 90 % of all reports. In particular, principal aspects as defined in a standardized external terminology were recognized with f 1=.989 (micro average) and f 1=.963 (macro average). As a result of keyword matching and restraint concept extraction, the system obtained high precision also on unstructured or exceptionally short documents, and documents with uncommon layout. Conclusions The developed terminology and the proposed information extraction system allow to extract fine-grained information from German semi-structured transthoracic echocardiography reports with very high precision and high recall on the majority of documents at the University Hospital of Würzburg. Extracted results populate a clinical data warehouse which supports clinical research. Y1 - 2015 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-125509 VL - 15 IS - 91 ER - TY - JOUR A1 - Hörnlein, Alexander A1 - Mandel, Alexander A1 - Ifland, Marianus A1 - Lüneberg, Edeltraud A1 - Deckert, Jürgen A1 - Puppe, Frank T1 - Akzeptanz medizinischer Trainingsfälle als Ergänzung zu Vorlesungen T1 - Acceptance of medical training cases as supplement to lectures JF - GMS Zeitschrift für Medizinische Ausbildung N2 - Introduction: Medical training cases (virtual patients) are in widespread use for student education. Most publications report about development and experiences in one course with training cases. In this paper we compare the acceptance of different training case courses with different usages deployed as supplement to lectures of the medical faculty of Wuerzburg university during a period of three semesters. Methods: The training cases were developed with the authoring tool CaseTrain and are available for students via the Moodle-based eLearning platform WueCampus at Wuerzburg university. Various data about usage and acceptance is automatically collected. Results: From WS (winter semester) 08/09 till WS 09/10 19 courses with about 200 cases were available. In each semester, about 550 different medical students from Würzburg and 50 students from other universities processed about 12000 training cases and filled in about 2000 evaluation forms. In different courses, the usage varied between less than 50 and more than 5000 processed cases. Discussion: Although students demand training cases as supplement to all lectures, the data show that the usage does not primarily depend on the quality of the available training cases. Instead, the training cases of nearly all case collections were processed extremely often shortly before the examination. It shows that the degree of usage depends primarily on the perceived relevance of the training cases for the examination." N2 - Einleitung: Medizinische Trainingsfälle sind in der studentischen Ausbildung inzwischen weit verbreitet. In den meisten Publikationen wird über die Entwicklung und die Erfahrungen in einem Kurs mit Trainingsfällen berichtet. In diesem Beitrag vergleichen wir die Akzeptanz von verschiedenen Trainingsfallkursen, die als Ergänzung zu zahlreichen Vorlesungen der Medizinischen Fakultät der Universität Würzburg mit sehr unterschiedlichen Nutzungsraten eingesetzt wurden, über einen Zeitraum von drei Semestern. Methoden: Die Trainingsfälle wurden mit dem Autoren- und Ablaufsystem CaseTrain erstellt und über die Moodle-basierte Würzburger Lernplattform WueCampus den Studierenden verfügbar gemacht. Dabei wurden umfangreiche Daten über die Nutzung und Akzeptanz erhoben. Ergebnisse: Im Zeitraum vom WS 08/09 bis zum WS 09/10 waren 19 Kurse mit insgesamt ca. 200 Fällen für die Studierenden verfügbar, die pro Semester von ca. 550 verschiedenen Medizinstudenten der Universität Würzburg und weiteren 50 Studierenden anderer bayerischer Universitäten genutzt wurden. Insgesamt wurden pro Semester ca. 12000 Mal Trainingsfälle vollständig durchgespielt zu denen ca. 2000 Evaluationen von den Studierenden ausgefüllt wurden. In den verschiedenen Kursen variiert die Nutzung zwischen unter 50 Bearbeitungen in wenig frequentierten Fallsammlungen und über 5000 Bearbeitungen in stark frequentierten Fallsammlungen. Diskussion: Auch wenn Studierende wünschen, dass zu allen Vorlesungen Trainingsfälle angeboten werden, zeigen die Daten, dass der Umfang der Nutzung nicht primär von der Qualität der verfügbaren Trainingsfälle abhängt. Dagegen werden die Trainingsfälle in fast allen Fallsammlungen kurz vor den Klausuren extrem häufig bearbeitet. Dies zeigt, dass die Nutzung von Trainingsfällen im Wesentlichen von der wahrgenommenen Klausurrelevanz der Fälle abhängt. KW - Blended Learning KW - Trainingsfall KW - Problembasiertes Lernen KW - Akzeptanz-Evaluation KW - Autorensystem KW - CaseTrain KW - Teaching KW - Patient Simulation KW - Problem Based Learning KW - Acceptance Evaluation KW - Authoring Tools Y1 - 2011 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-133569 VL - 28 IS - 3 ER - TY - JOUR A1 - Mandel, Alexander A1 - Hörnlein, Alexander A1 - Ifland, Marianus A1 - Lüneburg, Edeltraud A1 - Deckert, Jürgen A1 - Puppe, Frank T1 - Aufwandsanalyse für computerunterstützte Multiple-Choice Papierklausuren T1 - Cost analysis for computer supported multiple-choice paper examinations JF - GMS Journal for Medical Education N2 - Introduction: Multiple-choice-examinations are still fundamental for assessment in medical degree programs. In addition to content related research, the optimization of the technical procedure is an important question. Medical examiners face three options: paper-based examinations with or without computer support or completely electronic examinations. Critical aspects are the effort for formatting, the logistic effort during the actual examination, quality, promptness and effort of the correction, the time for making the documents available for inspection by the students, and the statistical analysis of the examination results. Methods: Since three semesters a computer program for input and formatting of MC-questions in medical and other paper-based examinations is used and continuously improved at Wuerzburg University. In the winter semester (WS) 2009/10 eleven, in the summer semester (SS) 2010 twelve and in WS 2010/11 thirteen medical examinations were accomplished with the program and automatically evaluated. For the last two semesters the remaining manual workload was recorded. Results: The cost of the formatting and the subsequent analysis including adjustments of the analysis of an average examination with about 140 participants and about 35 questions was 5-7 hours for exams without complications in the winter semester 2009/2010, about 2 hours in SS 2010 and about 1.5 hours in the winter semester 2010/11. Including exams with complications, the average time was about 3 hours per exam in SS 2010 and 2.67 hours for the WS 10/11. Discussion: For conventional multiple-choice exams the computer-based formatting and evaluation of paper-based exams offers a significant time reduction for lecturers in comparison with the manual correction of paper-based exams and compared to purely electronically conducted exams it needs a much simpler technological infrastructure and fewer staff during the exam." N2 - Einleitung: Multiple-Choice-Klausuren spielen immer noch eine herausragende Rolle für fakultätsinterne medizinische Prüfungen. Neben inhaltlichen Arbeiten stellt sich die Frage, wie die technische Abwicklung optimiert werden kann. Für Dozenten in der Medizin gibt es zunehmend drei Optionen zur Durchführung von MC-Klausuren: Papierklausuren mit oder ohne Computerunterstützung oder vollständig elektronische Klausuren. Kritische Faktoren sind der Aufwand für die Formatierung der Klausur, der logistische Aufwand bei der Klausurdurchführung, die Qualität, Schnelligkeit und der Aufwand der Klausurkorrektur, die Bereitstellung der Dokumente für die Einsichtnahme, und die statistische Analyse der Klausurergebnisse. Methoden: An der Universität Würzburg wird seit drei Semestern ein Computerprogramm zur Eingabe und Formatierung der MC-Fragen in medizinischen und anderen Papierklausuren verwendet und optimiert, mit dem im Wintersemester (WS) 2009/2010 elf, im Sommersemester (SS) 2010 zwölf und im WS 2010/11 dreizehn medizinische Klausuren erstellt und anschließend die eingescannten Antwortblätter automatisch ausgewertet wurden. In den letzten beiden Semestern wurden die Aufwände protokolliert. Ergebnisse: Der Aufwand der Formatierung und der Auswertung einschl. nachträglicher Anpassung der Auswertung einer Durchschnittsklausur mit ca. 140 Teilnehmern und ca. 35 Fragen ist von 5-7 Stunden für Klausuren ohne Komplikation im WS 2009/2010 über ca. 2 Stunden im SS 2010 auf ca. 1,5 Stunden im WS 2010/11 gefallen. Einschließlich der Klausuren mit Komplikationen bei der Auswertung betrug die durchschnittliche Zeit im SS 2010 ca. 3 Stunden und im WS 10/11 ca. 2,67 Stunden pro Klausur. Diskussion: Für konventionelle Multiple-Choice-Klausuren bietet die computergestützte Formatierung und Auswertung von Papierklausuren einen beträchtlichen Zeitvorteil für die Dozenten im Vergleich zur manuellen Korrektur von Papierklausuren und benötigt im Vergleich zu rein elektronischen Klausuren eine deutlich einfachere technische Infrastruktur und weniger Personal bei der Klausurdurchführung. KW - Multiple-Choice Prüfungen KW - Automatisierte Prüfungskorrektur KW - Aufwandsanalyse KW - Educational Measurement (I2.399) KW - Self-Evaluation Programs (I2.399.780) KW - Multiple-Choice Examination KW - Cost Analysis Y1 - 2011 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-134386 VL - 28 IS - 4 ER - TY - JOUR A1 - Krenzer, Adrian A1 - Makowski, Kevin A1 - Hekalo, Amar A1 - Fitting, Daniel A1 - Troya, Joel A1 - Zoller, Wolfram G. A1 - Hann, Alexander A1 - Puppe, Frank T1 - Fast machine learning annotation in the medical domain: a semi-automated video annotation tool for gastroenterologists JF - BioMedical Engineering OnLine N2 - Background Machine learning, especially deep learning, is becoming more and more relevant in research and development in the medical domain. For all the supervised deep learning applications, data is the most critical factor in securing successful implementation and sustaining the progress of the machine learning model. Especially gastroenterological data, which often involves endoscopic videos, are cumbersome to annotate. Domain experts are needed to interpret and annotate the videos. To support those domain experts, we generated a framework. With this framework, instead of annotating every frame in the video sequence, experts are just performing key annotations at the beginning and the end of sequences with pathologies, e.g., visible polyps. Subsequently, non-expert annotators supported by machine learning add the missing annotations for the frames in-between. Methods In our framework, an expert reviews the video and annotates a few video frames to verify the object’s annotations for the non-expert. In a second step, a non-expert has visual confirmation of the given object and can annotate all following and preceding frames with AI assistance. After the expert has finished, relevant frames will be selected and passed on to an AI model. This information allows the AI model to detect and mark the desired object on all following and preceding frames with an annotation. Therefore, the non-expert can adjust and modify the AI predictions and export the results, which can then be used to train the AI model. Results Using this framework, we were able to reduce workload of domain experts on average by a factor of 20 on our data. This is primarily due to the structure of the framework, which is designed to minimize the workload of the domain expert. Pairing this framework with a state-of-the-art semi-automated AI model enhances the annotation speed further. Through a prospective study with 10 participants, we show that semi-automated annotation using our tool doubles the annotation speed of non-expert annotators compared to a well-known state-of-the-art annotation tool. Conclusion In summary, we introduce a framework for fast expert annotation for gastroenterologists, which reduces the workload of the domain expert considerably while maintaining a very high annotation quality. The framework incorporates a semi-automated annotation system utilizing trained object detection models. The software and framework are open-source. KW - object detection KW - machine learning KW - deep learning KW - annotation KW - endoscopy KW - gastroenterology KW - automation Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-300231 VL - 21 IS - 1 ER - TY - JOUR A1 - Kaspar, Mathias A1 - Fette, Georg A1 - Hanke, Monika A1 - Ertl, Maximilian A1 - Puppe, Frank A1 - Störk, Stefan T1 - Automated provision of clinical routine data for a complex clinical follow-up study: A data warehouse solution JF - Health Informatics Journal N2 - A deep integration of routine care and research remains challenging in many respects. We aimed to show the feasibility of an automated transformation and transfer process feeding deeply structured data with a high level of granularity collected for a clinical prospective cohort study from our hospital information system to the study's electronic data capture system, while accounting for study-specific data and visits. We developed a system integrating all necessary software and organizational processes then used in the study. The process and key system components are described together with descriptive statistics to show its feasibility in general and to identify individual challenges in particular. Data of 2051 patients enrolled between 2014 and 2020 was transferred. We were able to automate the transfer of approximately 11 million individual data values, representing 95% of all entered study data. These were recorded in n = 314 variables (28% of all variables), with some variables being used multiple times for follow-up visits. Our validation approach allowed for constant good data quality over the course of the study. In conclusion, the automated transfer of multi-dimensional routine medical data from HIS to study databases using specific study data and visit structures is complex, yet viable. KW - clinical data warehouse KW - clinical study KW - electronic data capture KW - electronic health records KW - secondary data usage Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-260828 VL - 28 IS - 1 ER - TY - JOUR A1 - Wick, Christoph A1 - Hartelt, Alexander A1 - Puppe, Frank T1 - Staff, symbol and melody detection of Medieval manuscripts written in square notation using deep Fully Convolutional Networks JF - Applied Sciences N2 - Even today, the automatic digitisation of scanned documents in general, but especially the automatic optical music recognition (OMR) of historical manuscripts, still remains an enormous challenge, since both handwritten musical symbols and text have to be identified. This paper focuses on the Medieval so-called square notation developed in the 11th–12th century, which is already composed of staff lines, staves, clefs, accidentals, and neumes that are roughly spoken connected single notes. The aim is to develop an algorithm that captures both the neumes, and in particular its melody, which can be used to reconstruct the original writing. Our pipeline is similar to the standard OMR approach and comprises a novel staff line and symbol detection algorithm based on deep Fully Convolutional Networks (FCN), which perform pixel-based predictions for either staff lines or symbols and their respective types. Then, the staff line detection combines the extracted lines to staves and yields an F\(_1\) -score of over 99% for both detecting lines and complete staves. For the music symbol detection, we choose a novel approach that skips the step to identify neumes and instead directly predicts note components (NCs) and their respective affiliation to a neume. Furthermore, the algorithm detects clefs and accidentals. Our algorithm predicts the symbol sequence of a staff with a diplomatic symbol accuracy rate (dSAR) of about 87%, which includes symbol type and location. If only the NCs without their respective connection to a neume, all clefs and accidentals are of interest, the algorithm reaches an harmonic symbol accuracy rate (hSAR) of approximately 90%. In general, the algorithm recognises a symbol in the manuscript with an F\(_1\) -score of over 96%. KW - optical music recognition KW - historical document analysis KW - medieval manuscripts KW - neume notation KW - fully convolutional neural networks Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-197248 SN - 2076-3417 VL - 9 IS - 13 ER - TY - JOUR A1 - Reul, Christian A1 - Christ, Dennis A1 - Hartelt, Alexander A1 - Balbach, Nico A1 - Wehner, Maximilian A1 - Springmann, Uwe A1 - Wick, Christoph A1 - Grundig, Christine A1 - Büttner, Andreas A1 - Puppe, Frank T1 - OCR4all—An open-source tool providing a (semi-)automatic OCR workflow for historical printings JF - Applied Sciences N2 - Optical Character Recognition (OCR) on historical printings is a challenging task mainly due to the complexity of the layout and the highly variant typography. Nevertheless, in the last few years, great progress has been made in the area of historical OCR, resulting in several powerful open-source tools for preprocessing, layout analysis and segmentation, character recognition, and post-processing. The drawback of these tools often is their limited applicability by non-technical users like humanist scholars and in particular the combined use of several tools in a workflow. In this paper, we present an open-source OCR software called OCR4all, which combines state-of-the-art OCR components and continuous model training into a comprehensive workflow. While a variety of materials can already be processed fully automatically, books with more complex layouts require manual intervention by the users. This is mostly due to the fact that the required ground truth for training stronger mixed models (for segmentation, as well as text recognition) is not available, yet, neither in the desired quantity nor quality. To deal with this issue in the short run, OCR4all offers a comfortable GUI that allows error corrections not only in the final output, but already in early stages to minimize error propagations. In the long run, this constant manual correction produces large quantities of valuable, high quality training material, which can be used to improve fully automatic approaches. Further on, extensive configuration capabilities are provided to set the degree of automation of the workflow and to make adaptations to the carefully selected default parameters for specific printings, if necessary. During experiments, the fully automated application on 19th Century novels showed that OCR4all can considerably outperform the commercial state-of-the-art tool ABBYY Finereader on moderate layouts if suitably pretrained mixed OCR models are available. Furthermore, on very complex early printed books, even users with minimal or no experience were able to capture the text with manageable effort and great quality, achieving excellent Character Error Rates (CERs) below 0.5%. The architecture of OCR4all allows the easy integration (or substitution) of newly developed tools for its main components by standardized interfaces like PageXML, thus aiming at continual higher automation for historical printings. KW - optical character recognition KW - document analysis KW - historical printings Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-193103 SN - 2076-3417 VL - 9 IS - 22 ER - TY - JOUR A1 - Djebko, Kirill A1 - Puppe, Frank A1 - Kayal, Hakan T1 - Model-based fault detection and diagnosis for spacecraft with an application for the SONATE triple cube nano-satellite JF - Aerospace N2 - The correct behavior of spacecraft components is the foundation of unhindered mission operation. However, no technical system is free of wear and degradation. A malfunction of one single component might significantly alter the behavior of the whole spacecraft and may even lead to a complete mission failure. Therefore, abnormal component behavior must be detected early in order to be able to perform counter measures. A dedicated fault detection system can be employed, as opposed to classical health monitoring, performed by human operators, to decrease the response time to a malfunction. In this paper, we present a generic model-based diagnosis system, which detects faults by analyzing the spacecraft’s housekeeping data. The observed behavior of the spacecraft components, given by the housekeeping data is compared to their expected behavior, obtained through simulation. Each discrepancy between the observed and the expected behavior of a component generates a so-called symptom. Given the symptoms, the diagnoses are derived by computing sets of components whose malfunction might cause the observed discrepancies. We demonstrate the applicability of the diagnosis system by using modified housekeeping data of the qualification model of an actual spacecraft and outline the advantages and drawbacks of our approach. KW - fault detection KW - model-based diagnosis KW - nano-satellite Y1 - 2019 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-198836 SN - 2226-4310 VL - 6 IS - 10 ER -