Refine
Year of publication
Document Type
- Article (19)
- Working Paper (5)
- Part of a Book (4)
- Conference Proceeding (4)
- Preprint (3)
- Report (3)
Has Fulltext
- yes (38)
Is part of the Bibliography
- no (38)
Keywords
Institute
- Informatik (38)
- Medizin (3)
- E-Finance Lab e.V. (1)
In contrast to the symbolic approach, neural networks seldom are designed to explain what they have learned. This is a major obstacle for its use in everyday life. With the appearance of neuro-fuzzy systems which use vague, human-like categories the situation has changed. Based on the well-known mechanisms of learning for RBF networks, a special neuro-fuzzy interface is proposed in this paper. It is especially useful in medical applications, using the notation and habits of physicians and other medically trained people. As an example, a liver disease diagnosis system is presented.
The efficient management of large multimedia databases requires the development of new techniques to process, characterize, and search for multimedia objects. Especially in the case of image data, the rapidly growing amount of documents prohibits a manual description of the images’ content. Instead, the automated characterization is highly desirable to support annotation and retrieval of digital images. However, this is a very complex and still unsolved task. To contribute to a solution of this problem, we have developed a mechanism for recognizing objects in images based on the query by example paradigm. Therefore, the most salient image features of an example image representing the searched object are extracted to obtain a scale-invariant object model. The use of this model provides an efficient and robust strategy for recognizing objects in images independently of their size. Further applications of the mechanism are classical recognition tasks such as scene decomposition or object tracking in video sequences.
Erkennung kritischer Zustände von Patienten mit der Diagnose "Septischer Schock" mit einem RBF-Netz
(2000)
Es wurde gezeigt, dass der Arzt mit dem wachsenden RBF-Netz durch die Ausgabe von verlässlichen Warnungen unterstützt werden kann. Wie in der Clusteranalyse erläutert, leiden die Ergebnisse jedoch unter den wenigen Patienten und unter der ungenauen zeitlichen Erfassung der Daten. Da jeder Patient sehr individuelle Zustände annimmt, ist ein größeres Patientenkollektiv notwendig, um eine umfassende Wissensbasis zu lernen. Eine medizinische Nachbearbeitung der Wissensbasis durch die Analyse der Fälle ließe eine weitere Verbesserung des Ergebnisses erwarten. Somit könnten unbekannte Zusammenhänge durch das Lernen aus Beispielen und medizinisches Fachwissen kombiniert werden. Abstraktere Merkmale, die weniger abhängig von individuellen Zuständen sind, könnten eine Klassifikation noch weiter verbessern. Ein Ansatzpunkt ist z.B. die Abweichung der Messwerte vom gleitenden Mittelwert. Dieses Maß ist unempfindlicher gegenüber den individuellen Arbeitspunkten der Patienten und bildet auch die Basis von relativen Abhängigkeiten zwischen zwei Variablen, die in einem weiteren Schritt ebenfalls als Merkmal herangezogen wurden. Obwohl die Verwendung der relativen Abhängigkeiten zwischen zwei Variablen als Merkmal nicht deutlichere oder häufigere Warnungen hervorbringen konnte, weist doch die Clusteranalyse auf eine bessere Verteilung der Patienten hin. Einige Cluster sind besser für die Vorhersage geeignet, als dieses bei einer Clusterung auf Basis der Zustände erreicht werden kann. Unterstützt wird dieses Ergebnis auch durch den größeren Unterschied der Sicherheiten von falschen und richtigen Klassifikationen. Neben den bisher untersuchten Merkmalen scheinen auch die Variablen interessant zu sein, bei denen festgestellt wurde, dass sie sich trotz Medikamentengabe und adäquater Behandlung schwer stabilisieren lassen. Durch den behandelnden Arzt werden diese Werte üblicherweise in einem gewissen Bereich gehalten. Falls sich das Paar Medikament/physiologischer Parameter nicht mehr in einem sinnvollen Verhältnis befindet, kann dieses ein wichtiger Indikator sein. Nach dem Aufbau der grundlegenden Funktionalität der hier untersuchten Methoden ist die Suche nach geeigneten Merkmalen als Eingabe für ein neuronales Netz ein wesentlicher Bestandteil folgender Arbeiten. Abgesehen von dem generell anspruchsvollen Vorhaben aus Klinikdaten deutliche Hinweise für die Mortalität septischer-Schock-Patienten zu erhalten, liegen die wesentlichen Probleme in dem Umfang und der Messhäufigkeit der Frankfurter Vorstudie begründet, so dass eine Anwendung von Klassifikationsverfahren auf das umfassendere Patientenkollektiv der MEDAN Multicenter-Studie klarere Ergebnisse erwarten lässt. Eine weitere, für medizinische Anwendungen interessante, Analysemöglichkeit ist die Regelgenerierung, die zur Zeit in einem anderen Teilprojekt in der MEDAN-Arbeitsgruppe bearbeitet wird. Hier können im Fall metrischer Daten zusätzliche Hinweise für die Leistung eines reinen Klassifikationsverfahrens gewonnen werden mit dem Vorteil einer expliziten Regelausgabe. Zum anderen werden in diesem Teilprojekt auch Verfahren zur Regelgenerierung eingesetzt, die ordinale und nominale Variablen wie Diagnosen, Operationen, Therapien und Medikamentenangaben (binär, ohne genaue Dosis) auswerten können. Diese werden in den Multicenter-Daten vorhanden sein. Durch Kopplung der Regelgeneratoren für metrische Daten auf der einen Seite und für diskrete Variablen auf der anderen Seite, besteht durchaus die Hoffnung bessere Ergebnisse zu erzielen. Da der Regelgenerator für metrische Daten auf dem RBF-DDA (Abk. für: Dynamic Decay Adjustment)-Netz [BERTHOLD und DIAMOND, 1995] beruht, bietet es sich innerhalb des MEDAN-Projekts an, einen (bislang nicht durchgeführten) Vergleich mit dem hier verwendeten Netztyp durchzuführen. Der Vergleich ist allerdings nur von prinzipiellem Interesse und kann auf den hier betrachteten Daten kein grundsätzlich besseres Ergebnis liefern als die bislang durchgeführten Analysen; er kann aber zu einer umfangreichen Bewertung der Ergebnisse beitragen.
In its first part, this contribution reviews shortly the application of neural network methods to medical problems and characterizes its advantages and problems in the context of the medical background. Successful application examples show that human diagnostic capabilities are significantly worse than the neural diagnostic systems. Then, paradigm of neural networks is shortly introduced and the main problems of medical data base and the basic approaches for training and testing a network by medical data are described. Additionally, the problem of interfacing the network and its result is given and the neuro-fuzzy approach is presented. Finally, as case study of neural rule based diagnosis septic shock diagnosis is described, on one hand by a growing neural network and on the other hand by a rule based system. Keywords: Statistical Classification, Adaptive Prediction, Neural Networks, Neurofuzzy, Medical Systems
In diesem Bericht wurde das in [Pae02] eingeführte Verfahren "GenDurchschnitt" auf die symbolischen Daten zweier Datenbanken septischer Schock-Patienten angewendet. Es wurden jeweils Generalisierungsregeln generiert, die neben einer robusten Klassifikation der Patienten in die Klassen "überlebt" und "verstorben" auch eine Interpretation der Daten ermöglichten. Ein Vergleich mit den aktuellen Verfahren A-priori und FP-Baum haben die gute Verwendbarkeit des Algorithmus belegt. Die Heuristiken führten zu Laufzeitverbesserungen. Insbesondere die Möglichkeit, die Wichtigkeit von Variablen pro Klasse zu berechnen, führte zu einer Variablenreduktion im Eingaberaum und zu der Identifikation wichtiger Items. Einige Regelbeispiele wurden für jeden Datensatz genannt. Die Frühzeitigkeit von Regeln lieferte für die beiden Datenbanken ein unterschiedliches Ergebnis: Bei den ASK-Daten treten die Regeln für die Klasse "verstorben" früher als die der Klasse "überlebt" auf; bei den MEDAN-Klinikdaten ist es umgekehrt. Eine Erklärung hierfür könnte sein, dass es sich im Vergleich zu den MEDAN-Klinikdaten bei den ASK-Daten um ein Patientenkollektiv mit einer anderen, speziellen Patientencharakteristik handelt. Anhand der Ähnlichkeit der Regeln konnten für den Anwender eine überschaubare Anzahl zuverlässiger Regeln ausgegeben werden, die möglichst unähnlich zueinander sind und somit für einen Arzt in ihrer Gesamtheit interessant sind. Assoziationsregeln und FP-Baum-Regeln erzeugen zwar kürzere Regeln, die aber zu zahlreich und nicht hinreichend sind (vgl. [Pae02, Abschnitt 4]). Zusätzlich zu der Analyse der symbolischen Daten ist auch die Analyse der metrischen MEDAN-Klinikdaten der septischen Schock-Patienten interessant. Ebenfalls ist eine Kombination der Analysen der metrischen und symbolischen Daten sinnvoll. Solche Analysen wurden ebenfalls durchgeführt; die Ergebnisse dieser Analysen werden an anderer Stelle präsentiert werden. Weitere Anwendungen der Generalisierungsregeln sind denkbar. Auch eine Verbesserung des theoretischen Fundaments (vgl. [Pae02]) erscheint sinnvoll, da erst das Zusammenspiel theoretischer und praktischer Anstrengungen zum Ziel führt.
The early prediction of mortality is one of the unresolved tasks in intensive care medicine. This contribution models medical symptoms as observations cased by transitions between hidden markov states. Learning the underlying state transition probabilities results in a prediction probability success of about 91%. The results are discussed and put in relation to the model used. Finally, the rationales for using the model are reflected: Are there states in the septic shock data?
In intensive care units physicians are aware of a high lethality rate of septic shock patients. In this contribution we present typical problems and results of a retrospective, data driven analysis based on two neural network methods applied on the data of two clinical studies. Our approach includes necessary steps of data mining, i.e. building up a data base, cleaning and preprocessing the data and finally choosing an adequate analysis for the medical patient data. We chose two architectures based on supervised neural networks. The patient data is classified into two classes (survived and deceased) by a diagnosis based either on the black-box approach of a growing RBF network and otherwise on a second network which can be used to explain its diagnosis by human-understandable diagnostic rules. The advantages and drawbacks of these classification methods for an early warning system are discussed.
The Internet as the biggest human library ever assembled keeps on growing. Although all kinds of information carriers (e.g. audio/video/hybrid file formats) are available, text based documents dominate. It is estimated that about 80% of all information worldwide stored electronically exists in (or can be converted into) text form. More and more, all kinds of documents are generated by means of a text processing system and are therefore available electronically. Nowadays, many printed journals are also published online and may even discontinue to appear in print form tomorrow. This development has many convincing advantages: the documents are both available faster (cf. prepress services) and cheaper, they can be searched more easily, the physical storage only needs a fraction of the space previously necessary and the medium will not age. For most people, fast and easy access is the most interesting feature of the new age; computer-aided search for specific documents or Web pages becomes the basic tool for information-oriented work. But this tool has problems. The current keyword based search machines available on the Internet are not really appropriate for such a task; either there are (way) too many documents matching the specified keywords are presented or none at all. The problem lies in the fact that it is often very difficult to choose appropriate terms describing the desired topic in the first place. This contribution discusses the current state-of-the-art techniques in content-based searching (along with common visualization/browsing approaches) and proposes a particular adaptive solution for intuitive Internet document navigation, which not only enables the user to provide full texts instead of manually selected keywords (if available), but also allows him/her to explore the whole database.
In bioinformatics, biochemical pathways can be modeled by many differential equations. It is still an open problem how to fit the huge amount of parameters of the equations to the available data. Here, the approach of systematically learning the parameters is necessary. In this paper, for the small, important example of inflammation modeling a network is constructed and different learning algorithms are proposed. It turned out that due to the nonlinear dynamics evolutionary approaches are necessary to fit the parameters for sparse, given data. Proceedings of the 15th IEEE International Conference on Tools with Artificial Intelligence - ICTAI 2003
In bioinformatics, biochemical signal pathways can be modeled by many differential equations. It is still an open problem how to fit the huge amount of parameters of the equations to the available data. Here, the approach of systematically obtaining the most appropriate model and learning its parameters is extremely interesting. One of the most often used approaches for model selection is to choose the least complex model which “fits the needs”. For noisy measurements, the model which has the smallest mean squared error of the observed data results in a model which fits too accurately to the data – it is overfitting. Such a model will perform good on the training data, but worse on unknown data. This paper propose as model selection criterion the least complex description of the observed data by the model, the minimum description length. For the small, but important example of inflammation modeling the performance of the approach is evaluated. Keywords: biochemical pathways, differential equations, septic shock, parameter estimation, overfitting, minimum description length.
Data driven automatic model selection and parameter adaptation – a case study for septic shock
(2004)
In bioinformatics, biochemical pathways can be modeled by many differential equations. It is still an open problem how to fit the huge amount of parameters of the equations to the available data. Here, the approach of systematically learning the parameters is necessary. This paper propose as model selection criterion the least complex description of the observed data by the model, the minimum description length. For the small, but important example of inflammation modeling the performance of the approach is evaluated.
In bioinformatics, biochemical pathways can be modeled by many differential equations. It is still an open problem how to fit the huge amount of parameters of the equations to the available data. Here, the approach of systematically learning the parameters is necessary. In this paper, for the small, important example of inflammation modeling a network is constructed and different learning algorithms are proposed. It turned out that due to the nonlinear dynamics evolutionary approaches are necessary to fit the parameters for sparse, given data. Keywords: model parameter adaption, septic shock. coupled differential equations, genetic algorithm.
Since the description of sepsis by Schottmüller in 1914, the amount on knowledge available on sepsis and its underlying pathophysiology has substantially increased. Epidemiologic examinations of abdominal septic shock patients show the potential for high risk posed by and the extensive therapy situation in the intensive care unit (ICU) (5). Unfortunately, until now it has not been possible to significantly reduce the mortality rate of septic shock, which is as high as 50-60% worldwide, although PROWESS' results (1) are encouraging. This paper summarizes the main results of the MEDAN project and their medical impacts. Several aspects are already published, see the references. The heterogeneity of patient groups and the variations in therapy strategies is seen as one of the main problems for sepsis trials. In the MEDAN multi-center study of 71 intensive care units in Germany, a group of 382 patients made up exclusively of abdominal septic shock patients who met the consensus criteria for septic shock (3) was analysed. For use within scores or stand-alone experiments variables are often studied as isolated variables, not as a multidimensional whole, e.g. a recent study takes a look at the role thrombocytes play (15). To avoid this limitation, our study compares several established scores (SOFA, APACHE II, SAPS II, MODS) by a multi-dimensional neuronal network analysis. For outcome prediction the data of 382 patients was analysed by using most of the commonly documented vital parameters and doses of medicine (metric variables). Data was collected in German hospitals from 1998 to 2001. The 382 handwritten patient records were transferred to an electronic database giving the amount of 2.5 million data entries. The metric data contained in the database is composed of daily measurements and doses of medicine. We used range and plausibility checks to allow no faulty data in the electronic database. 187 of the 382 patients are deceased (49 %).
At present, there are no quantitative, objective methods for diagnosing the Parkinson disease. Existing methods of quantitative analysis by myograms suffer by inaccuracy and patient strain; electronic tablet analysis is limited to the visible drawing, not including the writing forces and hand movements. In our paper we show how handwriting analysis can be obtained by a new electronic pen and new features of the recorded signals. This gives good results for diagnostics. Keywords: Parkinson diagnosis, electronic pen, automatic handwriting analysis
Attraction and commercial success of web sites depend heavily on the additional values visitors may find. Here, individual, automatically obtained and maintained user profiles are the key for user satisfaction. This contribution shows for the example of a cooking information site how user profiles might be obtained using category information provided by cooking recipes. It is shown that metrical distance functions and standard clustering procedures lead to erroneous results. Instead, we propose a new mutual information based clustering approach and outline its implications for the example of user profiling.
The dynamics of many systems are described by ordinary differential equations (ODE). Solving ODEs with standard methods (i.e. numerical integration) needs a high amount of computing time but only a small amount of storage memory. For some applications, e.g. short time weather forecast or real time robot control, long computation times are prohibitive. Is there a method which uses less computing time (but has drawbacks in other aspects, e.g. memory), so that the computation of ODEs gets faster? We will try to discuss this question for the assumption that the alternative computation method is a neural network which was trained on ODE dynamics and compare both methods using the same approximation error. This comparison is done with two different errors. First, we use the standard error that measures the difference between the approximation and the solution of the ODE which is hard to characterize. But in many cases, as for physics engines used in computer games, the shape of the approximation curve is important and not the exact values of the approximation. Therefore, we introduce a subjective error based on the Total Least Square Error (TLSE) which gives more consistent results. For the final performance comparison, we calculate the optimal resource usage for the neural network and evaluate it depending on the resolution of the interpolation points and the inter-point distance. Our conclusion gives a method to evaluate where neural nets are advantageous over numerical ODE integration and where this is not the case. Index Terms—ODE, neural nets, Euler method, approximation complexity, storage optimization.