Biomedical subjects

Christian Böhm

Publications and source records attributed to Christian Böhm.

4 recordsLinked to original sources

PubMed2006

Enhancing instance-based classification with local density: a new algorithm for classifying unbalanced biomedical data.

MOTIVATION: Classification is an important data mining task in biomedicine. In particular, classification on biomedical data often claims the separation of pathological and healthy samples with highest discriminatory performance for diagnostic issues. Even more important than the overall accuracy is the balance of a classifier, particularly if datasets of unbalanced class size are examined. RESULTS: We present a novel instance-based classification technique which takes both information of different local density of data objects and local cluster structures into account. Our method, which adopts the basic ideas of density-based outlier detection, determines the local point density in the neighborhood of an object to be classified and of all clusters in the corresponding region. A data object is assigned to that class where it fits best into the local cluster structure. The experimental evaluation on biomedical data demonstrates that our approach outperforms most popular classification methods. AVAILABILITY: The algorithm LCF is available for testing under http://biomed.umit.at/upload/lcfx.zip.

Algorithms↗

PubMed2005

Functional cluster analysis of CT perfusion maps: a new tool for diagnosis of acute stroke?

CT perfusion imaging constitutes an important contribution to the early diagnosis of acute stroke. Cerebral blood flow (CBF), cerebral blood volume (CBV) and time-to-peak (TTP) maps are used to estimate the severity of cerebral damage after acute ischemia. We introduce functional cluster analysis as a new tool to evaluate CT perfusion in order to identify normal brain, ischemic tissue and large vessels. CBF, CBV and TTP maps represent the basis for cluster analysis applying a partitioning (k-means) and density-based (density-based spatial clustering of applications with noise, DBSCAN) paradigm. In patients with transient ischemic attack and stroke, cluster analysis identified brain areas with distinct hemodynamic properties (gray and white matter) and segmented territorial ischemia. CBF, CBV and TTP values of each detected cluster were displayed. Our preliminary results indicate that functional cluster analysis of CT perfusion maps may become a helpful tool for the interpretation of perfusion maps and provide a rapid means for the segmentation of ischemic tissue.

Adult↗

PubMed2005

Modelling of classification rules on metabolic patterns including machine learning and expert knowledge.

Machine learning has a great potential to mine potential markers from high-dimensional metabolic data without any a priori knowledge. Exemplarily, we investigated metabolic patterns of three severe metabolic disorders, PAHD, MCADD, and 3-MCCD, on which we constructed classification models for disease screening and diagnosis using a decision tree paradigm and logistic regression analysis (LRA). For the LRA model-building process we assessed the relevance of established diagnostic flags, which have been developed from the biochemical knowledge of newborn metabolism, and compared the models' error rates with those of the decision tree classifier. Both approaches yielded comparable classification accuracy in terms of sensitivity (>95.2%), while the LRA models built on flags showed significantly enhanced specificity. The number of false positive cases did not exceed 0.001%.

Algorithms↗

PubMed2003

Murine pancreatic tumor cell line TD2 bears the characteristic pattern of genetic changes with two independently amplified gene loci.

TGFalpha/p53(+/-) transgenic mice represent a genetically engineered mouse model for pancreatic adenocarcinoma. The tumors develop a characteristic pattern of secondary genetic changes. From one of these tumors, the permanent cell line TD2 was established. Here, we describe in detail the genetic changes by molecular-cytogenetic techniques. The original tumor-specific CGH profile has been retained unchanged. The most characteristic aberration pattern bears chromosome 11. Egfr, localized on proximal chromosome 11, is amplified two to three times and leads to an easily identifiable, stable marker chromosome with a large amplification unit, which is present in each metaphase. The wild-type p53 gene on distal chromosome 11 is lost. The p16Ink4a locus on chromosome 4 is hypermethylated. For c-Myc a 15-fold amplification, present in a 1.65 Mb amplification unit, is detected on chromosome 15. Transition between presence in the form of several double minutes, DMs, or a single homogeneously staining region, HSR, was observed for c-Myc. Molecular-cytogenetic analysis of both amplification units show that Egfr amplification and c-Myc amplification represent two alternative modes by which genes get amplified in tumor cells. The expression level of the respective genes was proven by Northern blot analysis. The cell line TD2 represents a valuable in vitro model for pancreatic adenocarcinoma.

Animals↗