Fully automatic cardiac T2∗ relaxation time estimation using marker-controlled watershed

• Main Authors: Kittichai Wantanajittikul, Suwit Saekho, Arintaya Phrommintikul, Nipon Theera-Umpon, Sansanee Auephanwiriyakul, Monruedee Tapanya, Busakol Ngammuang
• Format: Conference Proceeding
• Published: 2018
• Subjects: Computer Science; Engineering
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84978863567&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/55521
• Institution: Chiang Mai University

© 2015 IEEE. Heart failure due to iron-overload cardiomyopathy is one of the main causes of mortality. The cardiomyopathy is reversible if intensive iron chelation treatment is done in time. However, the diagnosis is often delayed because the cardiac iron deposition is unpredictable and the symptoms are lately detected. The widely used method in many countries is by calculating a parameter called the T2∗ (T2-star) from magnetic resonance (MR) image sequences. In order to compute the T2∗ value, the region of interest (ROI) is manually selected by an expert which may require considerable time and skills. The aim of this work is hence to develop the fully automatic cardiac T2∗ measurement by using marker-controlled watershed for segmenting the interventricular septum in cardiac MR images. Mathematical morphologies are also used to reduce some errors. Moreover, a new approach for T2∗ evaluation is suggested in this work. The interventricular septa from MR images in all echo times (TE's) are segmented to calculate the signal intensities (SI's) while the classical method segments the interventricular septum only in the first TE. Thirty cardiac MR images were used in this work. The segmentation performances are evaluated only from the MR images in the first TE. The precision and recall are 0.9 and 0.8 compared with the two experts' opinions. It shows that the segmentation results of our proposed technique are quite similar to the experts' opinions. The T2∗ values were carried out based on the automatically segmented ROI's. The mean difference of T2∗ values between the proposed method and the experts' opinions is about 1 ms. This demonstrates that the T2∗ values from the proposed technique are also close to T2∗ values calculated by the experts.