Low-dose denoising is an effective method that utilizes the power of CT for screening while avoiding high radiation exposure. Several research work has reported the feasibility of deep learning based denoising, but none of them have explored the influence of different network designs on the denoising performance. In this work, we explored the impact of three commonly adapted network design concepts in denoising: (1) Network structure, (2) Residual learning, and (3) Training loss. We evaluated the network performance using the dataset containing 76 real patient scans from Mayo Clinic Low-dose CT dataset. Experimental results demonstrated that residual blocks and residual learning are recommended to be utilized in design, while pooling is not recommended. In addition, among the classical training losses, the mean absolute error (L1) loss outperforms the mean squared error (MSE) loss.
Machine learning, especially convolutional neural network (CNN) approach has been successfully applied in noise suppression in natural image. However, shifting from natural image to medical image filed remains challenging due to specific difficulties such as training samples limitation, clinically meaningful image quality requirement and so on. To address this challenge, one possible solution is to incorporate our human prior knowledge into the machine learning model to better benefit its power. Therefore, in this work, we propose one prior knowledge driven machine learning based approach for positron emission tomography (PET) sinogram data denoising. Two main properties of PET sinogram data were considered in CNN architecture design, which are the Poisson statistics of the data and different correlation strength in the detector and view directions. Specially, for the statistical property, the sparse non-local method was used. For the correlation property, separate convolution was applied in two directions respectively. Experimental results showed the proposed model outperform the CNN model without prior knowledge. Results also demonstrate our insight of applying human knowledge strength the power of machine learning in medical imaging field.
Deep learning such as Convolutional Neural Network (CNN) has demonstrated its superior in the field of image analysis. However, in the medical imaging field, deep learning faces more challenges for tumor classification in computer-aided diagnosis due to uncertainties of lesions including their size, scaling factor, rotation, shapes, etc. Thus, instead of feeding raw images, texture-based CNN model has been designed to classify the objects with their good attributes. For example, gray level co-occurrence matrix (GLCM) can be chosen as the descriptor of the texture pattern for many good properties such as uniform size, shape invariance, scaling invariance. However, there are many different texture metrics to measure the different texture patterns. Thus, an effective and efficient integration model is essential to further improve the classification performance from different texture patterns. In this paper, we proposed a multi-channel texture-based CNN model to effectively integrate intensity, gradient and curvature texture patterns together for differentiating the malignant from benign polyps. Performance was evaluated by the merit of area under the curve of receiver operating characteristics (AUC). Around 0.3~4.8% improvement has been observed by combining different texture patterns together. Finally, classification performance of AUC=86.7% has been achieved for a polyp mass dataset of 87 samples, which obtains 1.8% improvement compared with a state-of-the-art method. The results indicate that texture information from different metrics could be fused and classified with a better classification performance. It also sheds lights that data integration is important and indispensable to pursuit improvement in classification task.
Computer aided diagnosis (CADx) of polyps has shown great potential to advance the computed tomography colonography (CTC) technique with diagnostic capability. Facing the problem of numerous uncertainties such as polyp size, shape, and orientation in CTC, GLCM-CNN has been proved to be an effective deep learning based tumor classification method, where convolution neural network (CNN) makes decision based on the texture pattern encoded in gray level co-occurrence matrix (GLCM) containing 13 directions. The 13 directional GLCM, by sampling displacement, can be classified into 3 subgroups. Based on our evaluation on the information encoded in the three subgroups, we propose a multi-stage fusion CNN model, which makes the final decision based on two types of features, i.e. (1) a gate module selected group-specific features and (2) fused features learnt from all the features from three groups. On our polyp dataset, which contains 87 polyp masses, our proposed method outperforms both single sub-group based and 13 directional GLCM based CNN model by at least 1.3% in AUC by the average of 20 times 2 fold cross validation experiment results.
Purpose: Bayesian theory provides a sound framework for ultralow-dose computed tomography (ULdCT) image reconstruction with two terms for modeling the data statistical property and incorporating a priori knowledge for the image that is to be reconstructed. We investigate the feasibility of using a machine learning (ML) strategy, particularly the convolutional neural network (CNN), to construct a tissue-specific texture prior from previous full-dose computed tomography.
Approach: Our study constructs four tissue-specific texture priors, corresponding with lung, bone, fat, and muscle, and integrates the prior with the prelog shift Poisson (SP) data property for Bayesian reconstruction of ULdCT images. The Bayesian reconstruction was implemented by an algorithm called SP-CNN-T and compared with our previous Markov random field (MRF)-based tissue-specific texture prior algorithm called SP-MRF-T.
Results: In addition to conventional quantitative measures, mean squared error and peak signal-to-noise ratio, structure similarity index, feature similarity, and texture Haralick features were used to measure the performance difference between SP-CNN-T and SP-MRF-T algorithms in terms of the structure and tissue texture preservation, demonstrating the feasibility and the potential of the investigated ML approach.
Conclusions: Both training performance and image reconstruction results showed the feasibility of constructing CNN texture prior model and the potential of improving the structure preservation of the nodule comparing to our previous regional tissue-specific MRF texture prior model.
Bayesian theory lies down a sound framework for ultralow-dose computed tomography (ULdCT) image reconstruction with two terms for modeling the data statistical property and incorporating a priori knowledge for the tobe- reconstructed image. This study investigates the feasibility of using machine learning strategy, particularly the convolutional neural network (CNN), to construct a tissue-specific texture prior from previous full-dose CT (FdCT) and integrates the prior with the pre-log shift Poisson (SP) data property for Bayesian reconstruction of ULdCT images. The Bayesian reconstruction was implemented by an algorithm, called SP-CNN-T, and compared with our previous Markov random field (MRF) based tissue-specific texture prior algorithm, called SP-MRF-T. Both training performance and image reconstruction results showed the feasibility of constructing CNN texture prior model and the potential of improving the structure preservation of the nodule comparing to our previous regional tissue-specific MRF texture prior model. Quantitative structure similarity index (SSIM) and texture Haralick features (HF) were used to measure the performance difference between SP-CNN-T and SP-MRF-T algorithms, demonstrating the feasibility and the potential of the investigated machine learning approach.
Sparse view computed tomography (CT) is an effective way to lower the radiation exposure, but results in streaking artifacts in the constructed CT image due to insufficient projection views. Several approaches have been reported for full view sinogram synthesis by interpolating the missing data into the sparse-view sinogram. However, current interpolation methods tend to generate over-smoothed sinogram, which could not preserve the sharpness of the image. Such sharpness is often referred to the region boundaries or tissue texture and of high importance as clinical indicators. To address this issue, this paper aims to propose an efficient sharpness-preserve spare-view CT sinogram synthesis method based on convolutional neural network (CNN). The sharpness preserving is stressed by the zero-order and first-order difference based loss function in the model. This study takes advantage of the residual design to overcome the problem of degradation for our deep network (20 layers), which is capable of extracting high level information and dealing with large sample dimensions (672 x 672). The proposed model design and loss function achieved a better performance in both quantitative and qualitative evaluation comparing to current state-of-the-art works. This study also performs ablation test on the effect of different designs and researches on hyper-parameter settings in the loss function.
Markov random field (MRF) has been widely used to incorporate a priori knowledge as a penalty for regional smoothing in ultralow-dose computed tomography (ULdCT) image reconstruction, while the regional smoothing does not explicitly consider the tissue-specific textures. Our previous work showed the tissue-specific textures can be enhanced by extracting the tissue-specific MRF from the to-be-reconstructed subject’s previous full-dose CT (FdCT) scans. However, the same subject’s FdCT scans might not be available in some applications. To address this limitation, we have also investigated the feasibility of extracting the tissue-specific textures from an existing FdCT database instead of the to-be-reconstructed subject. This study aims to implement a machine learning strategy to realize the feasibility. Specifically, we trained a Random Forest (RF) model to learn the intrinsic relationship between the tissue textures and subjects’ physiological features. By learning this intrinsic correlation, this model can be used to identify one MRF candidate from the database as the prior knowledge for any subject’s current ULdCT image reconstruction. Besides the conventional physiological factors (like body mass index: BMI, gender, age), we further introduced another two features LungMark and BodyAngle to address the scanning position and angle. The experimental results showed that the BMI and LungMark are two features of the most importance for the classification. Our trained model can predict 0.99 precision at the recall rate of 2%, which means that for each subject, there will be 3390*0.02 = 67.8 valid MRF candidates in the database, where 3,390 is the total number of candidates in the database. Moreover, it showed that introducing the ULdCT texture prior into the RF model can increase the recall rate by 3% while the precision remaining 0.99.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.