Objective: A stable and accurate automatic tumor delineation method has been developed to facilitate the intelligent design of lung cancer radiotherapy process. The purpose of this paper is to introduce an automatic tumor segmentation network for lung cancer on CT images based on deep learning. Methods: In this paper, a hybrid convolution neural network (CNN) combining 2D CNN and 3D CNN was implemented for the automatic lung tumor delineation using CT images. 3D CNN used V-Net model for the extraction of tumor context information from CT sequence images. 2D CNN used an encoder–decoder structure based on dense connection scheme, which could expand information flow and promote feature propagation. Next, 2D features and 3D features were fused through a hybrid module. Meanwhile, the hybrid CNN was compared with the individual 3D CNN and 2D CNN, and three evaluation metrics, Dice, Jaccard and Hausdorff distance (HD), were used for quantitative evaluation. The relationship between the segmentation performance of hybrid network and the GTV volume size was also explored. Results: The newly introduced hybrid CNN was trained and tested on a dataset of 260 cases, and could achieve a median value of 0.73, with mean and stand deviation of 0.72 ± 0.10 for the Dice metric, 0.58 ± 0.13 and 21.73 ± 13.30 mm for the Jaccard and HD metrics, respectively. The hybrid network significantly outperformed the individual 3D CNN and 2D CNN in the three examined evaluation metrics (p < 0.001). A larger GTV present a higher value for the Dice metric, but its delineation at the tumor boundary is unstable. Conclusions: The implemented hybrid CNN was able to achieve good lung tumor segmentation performance on CT images. Advances in knowledge: The hybrid CNN has valuable prospect with the ability to segment lung tumor.

This paper addresses the problem of the uncertainty description for multi-source information fusion. Based on the uncertainty description, an overall framework of data-driven multi-source information fusion is discussed. In this framework, the theoretical models existed in traditional algorithms are reasonably reserved or modified, experimental evidence and background theory are collaboratively worked in fusion process. Through the core data-driven engine, a loop between the traditional fusion process and performance evaluation is established, where the feedback information from evaluation can be used to adjust the fusion process.

Electronic support measure (ESM) can detect the bearings and Doppler frequencies simultaneously. A target tracking algorithm is proposed which uses ESM’s Doppler frequency and bearing measurements using extended Kalman filter (EKF). Compared with traditional bearings-only target tracking methods, our algorithm increases the Doppler frequency measurements and introduces the second-order EKF which can preferably handle the nonlinear tracking problem. Finally the simulation results show that the method improves the accuracy of target state estimation as well as the filtering stability. The maneuvers of sensor platform also can be avoided. Monte-Carlo runs with result analysis further illustrate the effectiveness of the algorithm.

A new algorithm for tracking a maneuvering target in presence of clutter or false measurements is addressed. Due to the availability of feature or attribute information in measurement vector, a joint probability density function description of the target state and target class is given. Using the joint state-class description the predictive measurement pdf can be proven to be a Gaussian mixture distribution. A Gaussian mixture Kalman filter is used for state estimation, where maneuver detection can also be avoided. In simulation the results with three tracking algorithms are compared, which have shown that proposed method here is more effective.

Traditional methods encountered two serious problems in tracking dim targets. One is the nonlinearity of the system model, and other is the low SNR of measurement signals. The two problems are hardly solved simultaneously in practical engineering applications. The particle filter is a recursive numerical technique which uses random sampling to approximate the optimal evaluation to target tracking problems. In this paper, we developed a method for tracking dim target using particle filter. Simulation results showed that the tracking performance of this method has greatly improved compared with classical extended Kalman filter and unscented Kalman filter.