Ren Togo1, Kenji Hirata2, Osamu Manabe3, Hiroshi Ohira4, Ichizo Tsujino4, Keiichi Magota5, Takahiro Ogawa6, Miki Haseyama6, Tohru Shiga2. 1. Graduate School of Information Science and Technology, Hokkaido University, Hokkaido, 060-0814, Japan. Electronic address: togo@lmd.ist.hokudai.ac.jp. 2. Department of Nuclear Medicine, Hokkaido University Graduate School of Medicine, Hokkaido, 060-8638, Japan. 3. Department of Nuclear Medicine, Hokkaido University Graduate School of Medicine, Hokkaido, 060-8638, Japan. Electronic address: osamumanabe817@med.hokudai.ac.jp. 4. First Department of Medicine, Hokkaido University Hospital, Hokkaido, 060-8638, Japan. 5. Division of Medical Imaging and Technology, Hokkaido University Hospital, Hokkaido, 060-8638, Japan. 6. Graduate School of Information Science and Technology, Hokkaido University, Hokkaido, 060-0814, Japan.
Abstract
AIMS: The aim of this study was to determine whether deep convolutional neural network (DCNN)-based features can represent the difference between cardiac sarcoidosis (CS) and non-CS using polar maps. METHODS: A total of 85 patients (33 CS patients and 52 non-CS patients) were analyzed as our study subjects. One radiologist reviewed PET/CT images and defined the left ventricle region for the construction of polar maps. We extracted high-level features from the polar maps through the Inception-v3 network and evaluated their effectiveness by applying them to a CS classification task. Then we introduced the ReliefF algorithm in our method. The standardized uptake value (SUV)-based classification method and the coefficient of variance (CoV)-based classification method were used as comparative methods. RESULTS: Sensitivity, specificity and the harmonic mean of sensitivity and specificity of our method with the ReliefF algorithm were 0.839, 0.870 and 0.854, respectively. Those of the SUVmax-based classification method were 0.468, 0.710 and 0.564, respectively, and those of the CoV-based classification method were 0.655, 0.750 and 0.699, respectively. CONCLUSION: The DCNN-based high-level features may be more effective than low-level features used in conventional quantitative analysis methods for CS classification.
AIMS: The aim of this study was to determine whether deep convolutional neural network (DCNN)-based features can represent the difference between cardiac sarcoidosis (CS) and non-CS using polar maps. METHODS: A total of 85 patients (33 CSpatients and 52 non-CSpatients) were analyzed as our study subjects. One radiologist reviewed PET/CT images and defined the left ventricle region for the construction of polar maps. We extracted high-level features from the polar maps through the Inception-v3 network and evaluated their effectiveness by applying them to a CS classification task. Then we introduced the ReliefF algorithm in our method. The standardized uptake value (SUV)-based classification method and the coefficient of variance (CoV)-based classification method were used as comparative methods. RESULTS: Sensitivity, specificity and the harmonic mean of sensitivity and specificity of our method with the ReliefF algorithm were 0.839, 0.870 and 0.854, respectively. Those of the SUVmax-based classification method were 0.468, 0.710 and 0.564, respectively, and those of the CoV-based classification method were 0.655, 0.750 and 0.699, respectively. CONCLUSION: The DCNN-based high-level features may be more effective than low-level features used in conventional quantitative analysis methods for CS classification.
Authors: Erito Marques de Souza Filho; Fernando de Amorim Fernandes; Christiane Wiefels; Lucas Nunes Dalbonio de Carvalho; Tadeu Francisco Dos Santos; Alair Augusto Sarmet M D Dos Santos; Evandro Tinoco Mesquita; Flávio Luiz Seixas; Benjamin J W Chow; Claudio Tinoco Mesquita; Ronaldo Altenburg Gismondi Journal: Front Cardiovasc Med Date: 2021-11-11
Authors: Cheuk To Chung; George Bazoukis; Sharen Lee; Ying Liu; Tong Liu; Konstantinos P Letsas; Antonis A Armoundas; Gary Tse Journal: Int J Arrhythmia Date: 2022-04-01