Automatic Segmentation of the Left Ventricle From Cardiac MRI Using Deep Learning and Double Snake Model

Abstract

The left ventricle segmentation (LVS) is of great important for the evaluation of cardiac function. This study aimed to establish new segmentation algorithms that can enhance the accuracy and robustness of automatic LVS on magnetic resonance images. The datasets involved 45 subjects, including 12 heart failure patients with ischemia, 12 heart failure patients without ischemia, 12 hypertrophy patients and 9 normal individuals. The experiments consisted of three important steps. At first, deep learning was employed for the coarse LVS on myocardial images. Next, a double snake model was applied to assess the endo-and epi-cardial boundaries. Finally, the optimal epicardial boundary was obtained by adopting radial region growing method. Additionally, the performance of the developed LVS method was evaluated by the previously established software. Furthermore, the developed LVS method was validated by applying the datasets of 45 subjects. The results showed that the good contours, overlapping dice metric and average perpendicular distance of both epi-and endo-cardial contours were approximately 97%, 0.97 and 1.8 mm respectively. The regression coefficient and coefficient of determination between the proposed method and clinical experts were 0.96 and 1.039, respectively for ejection fraction, while 0.92 and 0.994 for left ventricle mass. These findings reveal that the developed method can enhance the accuracy and robustness of LVS. This novel LVS approach exhibits outstanding performance and possesses promising potential to increase the reliability of computer-aided imaging detection system for cardiovascular disease.INDEX TERMS Convolutional neural network, snake model, left ventricle segmentation, magnetic resonance imaging, cardiac functional analysis.

An Overview of Deep Learning Methods for Left Ventricle Segmentation

“…Another hybrid approach based on CNN and the double snake model was proposed to segment the LV from MRI images [ 107 ]. A SegNet architecture was used for the initial segmentation result.…”

“…[77] (Intel(R) CPU i7-7770K, 4.2 GHz, 16G RAM) with an NVIDIA GPU (GeForce, GTX TITAN X, NVIDIA Corp., Santa Clara, CA, USA) [78] Two Intel Xeon 8 core CPUs, 12 GB of RAM, and an NVIDIA Quadro P6000 GPU [88] Four NVIDIA GTX 2080Ti GPU cards, each with 11 GB RAM [101] 3 Nvidia GTX 1080 Ti GPU [80] NVIDIA Titan GPU [104] NVIDIA GeForce GTX 1080 Ti GPU [111] Intel Core i5-7400 CPU. Te graphics card is an NVIDIA GeForce GTX 1060 [115] GeForce GTX 1080 ti GPU [107] Pentium dual-core 2.60 GHz hardware [108] CPU of AMD Phenom II X6 1055T Processor 2.8 GHz, 8G RAM, and VGA card of NVIDIA GeForce GTX 960 (CUDA v6.5) [64] GeForce GTX 1050 (4 GB GDDR5 dedicated) on an Intel Core i7-7700HQ (2.8 GHz, 6 MB cache, 4 cores) computer with 16 GB DDR4-2400 SDRAM [105] GTX 1080Ti graphic processor [121] GTX 1080Ti graphic processor [117] GTX 1080Ti graphic processor [118] NVIDIA Titan X GPU on Dell T7920 (GPU is Core I7, and memory size is 24 GB) [113] DELL TOWER 7910 workstation with 2.40 GHz Xeon E5-2620 v3 CPU, 32 GB RAM, and an Nvidia TITAN X Pascal GPU of 12 GB of memory [114] Two Intel Xeon 2.10 GHz CPU and four 12 GB Nvidia Titan XP GPU [122] GTX 1080Ti [119] NVIDIA Tesla P100 TensorFlow [75] TensorFlow [63] Not reported [66] Keras [38] MATLAB R2015b [69] Café [82] Keras [83] TensorFlow [84] TensorFlow [85] Café [91] TensorFlow [81] Keras Table Continued.…”

Decision Support Systems in HF based on Deep Learning Technologies

“…With the goal of predicting HF onset, applications of ML and DL to the field of image processing have been also proposed. In particular, several papers have focused on MRI imaging: in [34], a DL algorithm for the automatic segmentation of the left ventricle as a prior to evaluate the cardiac function in HF patients was proposed, where a Dice similarity coefficient of 0.97 was achieved. A ML method based on k-nearest neighbors was used in [35] to perform texture analysis of myocardial maps and identify early symptoms of HF, achieving an AUC of 0.85.…”

Detection of Left Ventricular Cavity from Cardiac MRI Images Using Faster R-CNN

“…Thus, LV localization is extremely useful in reducing computational load, particularly for deep learning algorithms. The fundamental advantage of LV localization is that it improves the effectiveness of subsequent techniques like regression [4] and segmentation [5][6][7] by extracting only the LV area and ignoring the other organs, as illustrated in Fig. 1.…”