PMID- 26946139 OWN - NLM STAT- MEDLINE DCOM- 20161007 LR - 20240324 IS - 1471-2342 (Electronic) IS - 1471-2342 (Linking) VI - 16 DP - 2016 Mar 5 TI - Automatic segmentation of myocardium at risk from contrast enhanced SSFP CMR: validation against expert readers and SPECT. PG - 19 LID - 10.1186/s12880-016-0124-1 [doi] LID - 19 AB - BACKGROUND: Efficacy of reperfusion therapy can be assessed as myocardial salvage index (MSI) by determining the size of myocardium at risk (MaR) and myocardial infarction (MI), (MSI = 1-MI/MaR). Cardiovascular magnetic resonance (CMR) can be used to assess MI by late gadolinium enhancement (LGE) and MaR by either T2-weighted imaging or contrast enhanced SSFP (CE-SSFP). Automatic segmentation algorithms have been developed and validated for MI by LGE as well as for MaR by T2-weighted imaging. There are, however, no algorithms available for CE-SSFP. Therefore, the aim of this study was to develop and validate automatic segmentation of MaR in CE-SSFP. METHODS: The automatic algorithm applies surface coil intensity correction and classifies myocardial intensities by Expectation Maximization to define a MaR region based on a priori regional criteria, and infarct region from LGE. Automatic segmentation was validated against manual delineation by expert readers in 183 patients with reperfused acute MI from two multi-center randomized clinical trials (RCT) (CHILL-MI and MITOCARE) and against myocardial perfusion SPECT in an additional set (n = 16). Endocardial and epicardial borders were manually delineated at end-diastole and end-systole. Manual delineation of MaR was used as reference and inter-observer variability was assessed for both manual delineation and automatic segmentation of MaR in a subset of patients (n = 15). MaR was expressed as percent of left ventricular mass (%LVM) and analyzed by bias (mean +/- standard deviation). Regional agreement was analyzed by Dice Similarity Coefficient (DSC) (mean +/- standard deviation). RESULTS: MaR assessed by manual and automatic segmentation were 36 +/- 10% and 37 +/- 11%LVM respectively with bias 1 +/- 6%LVM and regional agreement DSC 0.85 +/- 0.08 (n = 183). MaR assessed by SPECT and CE-SSFP automatic segmentation were 27 +/- 10%LVM and 29 +/- 7%LVM respectively with bias 2 +/- 7%LVM. Inter-observer variability was 0 +/- 3%LVM for manual delineation and -1 +/- 2%LVM for automatic segmentation. CONCLUSIONS: Automatic segmentation of MaR in CE-SSFP was validated against manual delineation in multi-center, multi-vendor studies with low bias and high regional agreement. Bias and variability was similar to inter-observer variability of manual delineation and inter-observer variability was decreased by automatic segmentation. Thus, the proposed automatic segmentation can be used to reduce subjectivity in quantification of MaR in RCT. CLINICAL TRIAL REGISTRATION: NCT01379261. NCT01374321. FAU - Tufvesson, Jane AU - Tufvesson J AD - Department of Clinical Physiology, Skane University Hospital in Lund, Lund University, Lund, Sweden. jane.tufvesson@med.lu.se. AD - Department of Biomedical Engineering, Faculty of Engineering, Lund University, Lund, Sweden. jane.tufvesson@med.lu.se. FAU - Carlsson, Marcus AU - Carlsson M AD - Department of Clinical Physiology, Skane University Hospital in Lund, Lund University, Lund, Sweden. marcus.carlsson@med.lu.se. FAU - Aletras, Anthony H AU - Aletras AH AD - Department of Clinical Physiology, Skane University Hospital in Lund, Lund University, Lund, Sweden. aletras@hotmail.com. AD - Laboratory of Medical Informatics, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece. aletras@hotmail.com. FAU - Engblom, Henrik AU - Engblom H AD - Department of Clinical Physiology, Skane University Hospital in Lund, Lund University, Lund, Sweden. henrik.engblom@med.lu.se. FAU - Deux, Jean-Francois AU - Deux JF AD - Department of Cardiology, Henri Mondor Hospital, Creteil, France. jean-francois.deux@hmn.aphp.fr. FAU - Koul, Sasha AU - Koul S AD - Department of Cardiology, Lund University, Lund, Sweden. sasha.koul@med.lu.se. FAU - Sorensson, Peder AU - Sorensson P AD - Department of Medicine, Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden. peder.sorensson@karolinska.se. FAU - Pernow, John AU - Pernow J AD - Department of Medicine, Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden. john.pernow@ki.se. FAU - Atar, Dan AU - Atar D AD - Department of Cardiology B, Oslo, University Hospital Ulleval and Faculty of Medicine, University of Oslo, Oslo, Norway. dan.atar@online.no. FAU - Erlinge, David AU - Erlinge D AD - Department of Cardiology, Lund University, Lund, Sweden. david.erlinge@med.lu.se. FAU - Arheden, Hakan AU - Arheden H AD - Department of Clinical Physiology, Skane University Hospital in Lund, Lund University, Lund, Sweden. hakan.arheden@med.lu.se. FAU - Heiberg, Einar AU - Heiberg E AD - Department of Clinical Physiology, Skane University Hospital in Lund, Lund University, Lund, Sweden. einar.heiberg@med.lu.se. AD - Department of Biomedical Engineering, Faculty of Engineering, Lund University, Lund, Sweden. einar.heiberg@med.lu.se. LA - eng SI - ClinicalTrials.gov/NCT01374321 SI - ClinicalTrials.gov/NCT01379261 PT - Journal Article PT - Multicenter Study PT - Randomized Controlled Trial PT - Research Support, Non-U.S. Gov't DEP - 20160305 PL - England TA - BMC Med Imaging JT - BMC medical imaging JID - 100968553 SB - IM MH - Algorithms MH - Humans MH - Magnetic Resonance Angiography/*methods MH - Myocardial Infarction/*diagnosis MH - Myocardium/*pathology MH - Observer Variation MH - Validation Studies as Topic PMC - PMC4779553 EDAT- 2016/03/08 06:00 MHDA- 2016/10/08 06:00 PMCR- 2016/03/05 CRDT- 2016/03/07 06:00 PHST- 2015/09/01 00:00 [received] PHST- 2016/02/24 00:00 [accepted] PHST- 2016/03/07 06:00 [entrez] PHST- 2016/03/08 06:00 [pubmed] PHST- 2016/10/08 06:00 [medline] PHST- 2016/03/05 00:00 [pmc-release] AID - 10.1186/s12880-016-0124-1 [pii] AID - 124 [pii] AID - 10.1186/s12880-016-0124-1 [doi] PST - epublish SO - BMC Med Imaging. 2016 Mar 5;16:19. doi: 10.1186/s12880-016-0124-1.