{M}odel-{B}ased {D}etection of {W}hite {M}atter in {O}ptical {C}oherence {T}omography {D}ata

Type of publication:  Artikel in einem Konferenzbericht
Buchtitel: Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Jahr: 2007
Monat: August 23rd-26th
Ort: Lyon, France
Notiz: OCT-based Neuro-Imaging
Abriss: A method for white matter detection in Optical Coherence Tomography A-Scans is presented. The Kalman filter is used to obtain a slope change estimate of the intensity signal. The estimate is subsequently analyzed by an spike detection algorithm and then evaluated by a neural network binary classifier (Perceptron). The capability of the proposed method is shown through the quantitative evaluation of simulated A-Scans, as well as applying the method on data obtained from a rat's brain in vitro. Results show that the developed algorithm identifies less false positives than other two spike detection methods, thus, enhancing the robustness and quality of detection.
Autoren: Gasca, Fernando
Ramrath, Lukas