Uli Niemann

Learning and inspecting classification rules from longitudinal epidemiological data to identify predictive features on hepatic steatosis

Abstract Personalized medicine requires the analysis of epidemiological data for the identification of subgroups sharing some risk factors and exhibiting dedicated outcome risks. We investigate the potential of data mining methods for the analysis of subgroups of cohort participants on hepatic steatosis. We propose a workflow for data preparation and mining on epidemiological data and we present InteractiveRuleMiner , an interactive tool for the inspection of rules in each subpopulation, including functionalities for the juxtaposition of labeled individuals and unlabeled ones. We report on our insights on specific subpopulations that have been discovered in a data-driven rather than hypothesis-driven way.

Can we distinguish between benign and malignant breast tumors in DCE-MRI by studying a tumor's most suspect region only?

We investigate the task of breast tumor classification based on dynamic contrast-enhanced magnetic resonance image data (DCE-MRI). Our objective is to study how the formation of regions of similar voxels contributes to distinguishing between benign and malignant tumors. First, we perform clustering on each tumor with different algorithms and parameter settings, and then combine the clustering results to identify the most suspect region of the tumor and derive features from it. With these features we train classifiers on a set of tumors that are difficult to classify, even for human experts. We show that the features of the most suspect region alone cannot distinguish between benign and malignant tumors, yet the properties of this region are indicative of tumor malignancy for the dataset we studied.

3D Regression Heat Map Analysis of Population Study Data

Epidemiological studies comprise heterogeneous data about a subject group to define disease-specific risk factors. These data contain information (features) about a subjects lifestyle, medical status as well as medical image data. Statistical regression analysis is used to evaluate these features and to identify feature combinations indicating a disease (the target feature). We propose an analysis approach of epidemiological data sets by incorporating all features in an exhaustive regression-based analysis. This approach combines all independent features w.r.t. a target feature. It provides a visualization that reveals insights into the data by highlighting relationships. The 3D Regression Heat Map, a novel 3D visual encoding, acts as an overview of the whole data set. It shows all combinations of two to three independent features with a specific target disease. Slicing through the 3D Regression Heat Map allows for the detailed analysis of the underlying relationships. Expert knowledge about disease-specific hypotheses can be included into the analysis by adjusting the regression model formulas. Furthermore, the influences of features can be assessed using a difference view comparing different calculation results. We applied our 3D Regression Heat Map method to a hepatic steatosis data set to reproduce results from a data mining-driven analysis. A qualitative analysis was conducted on a breast density data set. We were able to derive new hypotheses about relations between breast density and breast lesions with breast cancer. With the 3D Regression Heat Map, we present a visual overview of epidemiological data that allows for the first time an interactive regression-based analysis of large feature sets with respect to a disease.

Can We Classify the Participants of a Longitudinal Epidemiological Study from Their Previous Evolution

Medical research can greatly benefit from advances in data mining. We propose a mining approach for cohort analysis in a longitudinal population-based epidemiological study, and show that modelling and exploiting the evolution of cohort participants over time improves classification quality towards an outcome (a disease). Our mining workflow encompasses steps for tracing the evolution of the cohort participants and for using evolution features in classification. We show that our approach separates better between classes and that change in the values of variables is predictive. We report on results for the liver disorder hepatic steatosis (high fat accumulation in the liver), but our approach is appropriate for classification of longitudinal epidemiological data on further disorders.

Combining Subgroup Discovery and Clustering to Identify Diverse Subpopulations in Cohort Study Data

Subgroup discovery (SD) exploits its full value in applications where the goal is to generate understandable models. Epidemiologists search for statistically significant relationships between risk factors and outcome in large and heterogeneous datasets encompassing information about the participants health status gathered from questionnaires, medical examinations and image acquisition. SD algorithms can help epidemiologists by automatically detecting such relationships presented as comprehensible rules, aiming to ultimately improve prevention, diagnosis and treatment of diseases. However, SD algorithms often produce large and overlapping rule sets requiring the expert to conduct a manual post-filtering step that is time-consuming and tedious. In this work, we propose a clustering-based algorithm that hierarchically reorganizes rule sets and summarizes all important concepts while maintaining diversity between the rule clusters. For each cluster, a representative rule is selected and then displayed to the expert who in turn can drill-down to other cluster members. We evaluate our algorithm on two cohort study datasets where the diseases hepatic steatosis and goiter serve as target variable, respectively. We report on our findings with respect to effectiveness of our algorithm and we present selected subpopulations.

Classification of Benign and Malignant DCE-MRI Breast Tumors by Analyzing the Most Suspect Region

Classification of breast tumors solely based on dynamic contrast enhanced magnetic resonance data is a challenge in clinical research. In this paper, we analyze how the most suspect region as group of similarly perfused and spatially connected voxels of a breast tumor contributes to distinguishing between benign and malignant tumors. We use three density-based clustering algorithms to partition a tumor in regions and depict the most suspect one, as delivered by the most stable clustering algorithm. We use the properties of this region for each tumor as input to a classifier. Our preliminary results show that the classifier separates between benign and malignant tumors, and returns predictive attributes that are intuitive to the expert.

Patient Empowerment Through Summarization of Discussion Threads on Treatments in a Patient Self-help Forum

Self-help patient fora are widely used for information acquisition and exchange of experiences, e.g., on the effects of medical treatments for a disease. However, a new patient may have difficulties in getting a fast overview of the information inside a large forum. We propose TinnitusTreatmentMonitor, a prototype tool for the summarization and sentiment characterization of postings on medical treatments. We report on applying TinnitusTreatmentMonitor on the platform TinnitusTalk, a self-help platform for tinnitus patients.

ICE: Interactive Classification Rule Exploration on Epidemiological Data

Personalized medicine benefits from the identification of subpopulations that exhibit higher prevalence of a disease than the general population: such subpopulations can become the target of more intensive investigations to identify risk factors and to develop dedicated therapies. Classification rule discovery algorithms are an appropriate tool for discovering such subpopulations: they scale well, even for multi-dimensional data and deliver comprehensible patterns. However, they may generate hundreds of rules and thus call for exploration methods. In this study, we extend the tool Interactive Medical Miner for the discovery of classification rules, into the Interactive Classification rule Explorer ICE, which offers functionalities for rule exploration, grouping, rule visualization and statistics. We report on our first results for the classification of cohort data on goiter, a disorder of the thyroid gland.

Comparative Clustering of Plantar Pressure Distributions in Diabetics with Polyneuropathy May Be Applied to Reveal Inappropriate Biomechanical Stress

In diabetic patients, excessive peak plantar pressure has been identified as major risk factor for ulceration. Analyzing plantar pressure distributions potentially improves the identification of patients with a high risk for foot ulceration development. The goal of this study was to classify regional plantar pressure distributions. By means of a sensor-equipped insole, pressure recordings of healthy controls (n = 18) and diabetics with severe polyneuropathy (n = 25) were captured across eight foot regions. The study involved a controlled experimental protocol with multiple sessions, where a session contained several cycles of pressure exposure. Clustering was used to identify subgroups of study participants that are characterized by similar pressure distributions. For both analyzed groups, the number of clusters to best describe the pressure profiles was four. When both groups were combined, analysis again led to four distinct clusters. While three clusters did not separate between healthy and diabetic volunteers the fourth cluster was only represented by diabetics. Here the pressure distribution pattern is characterized by a focal point of pressure application on the forefoot and low pressure on the lateral region. Our data suggest that pressure clustering is a feasible means to identify inappropriate biomechanical plantar stress.

Interactive medical miner: interactively exploring subpopulations in epidemiological datasets

We present our Interactive Medical Miner, a tool for classification and model drill-down, designed to study epidemiological data. Our tool encompasses supervised learning (with decision trees and classification rules), utilities for data selection, and a rich panel with options for inspecting individual classification rules, and for studying the distribution of variables in each of the target classes. Since some of the epidemiological data available to the medical researcher may be still unlabeled (e.g. because the medical recordings for some part of the cohort are still in progress), our Interactive Medical Miner also supports the juxtaposition of labeled and unlabeled data. The set of methods and scientific workflow supported with our tool have been published in [1].