Inspired by complexity and diversity of biological neurons, our group proposed quadratic neurons by replacing the inner product in current artificial neurons with a quadratic operation on input data, thereby enhancing the capability of an individual neuron. Along this direction, we are motivated to evaluate the power of quadratic neurons in popular network architectures, simulating human-like learning in the form of “quadratic-neuron-based deep learning”. Our prior theoretical studies have shown important merits of quadratic neurons and networks in representation, efficiency, and interpretability. In this paper, we use quadratic neurons to construct an encoder-decoder structure, referred as the quadratic autoencoder, and apply it to low-dose CT denoising. The experimental results on the Mayo low-dose CT dataset demonstrate the utility and robustness of quadratic autoencoder in terms of image denoising and model efficiency. To our best knowledge, this is the first time that the deep learning approach is implemented with a new type of neurons and demonstrates a significant potential in the medical imaging field.

Lung segmentation constitutes a critical procedure for any clinical-decision supporting system aimed to improve the early diagnosis and treatment of lung diseases. Abnormal lungs mainly include lung parenchyma with commonalities on CT images across subjects, diseases and CT scanners, and lung lesions presenting various appearances. Segmentation of lung parenchyma can help locate and analyze the neighboring lesions, but is not well studied in the framework of machine learning. We proposed to segment lung parenchyma using a convolutional neural network (CNN) model. To reduce the workload of manually preparing the dataset for training the CNN, one clustering algorithm based method is proposed firstly. Specifically, after splitting CT slices into image patches, the k-means clustering algorithm with two categories is performed twice using the mean and minimum intensity of image patch, respectively. A cross-shaped verification, a volume intersection, a connected component analysis and a patch expansion are followed to generate final dataset. Secondly, we design a CNN architecture consisting of only one convolutional layer with six kernels, followed by one maximum pooling layer and two fully connected layers. Using the generated dataset, a variety of CNN models are trained and optimized, and their performances are evaluated by eightfold cross-validation. A separate validation experiment is further conducted using a dataset of 201 subjects (4.62 billion patches) with lung cancer or chronic obstructive pulmonary disease, scanned by CT or PET/CT. The segmentation results by our method are compared with those yielded by manual segmentation and some available methods. A total of 121,728 patches are generated to train and validate the CNN models. After the parameter optimization, our CNN model achieves an average F-score of 0.9917 and an area of curve up to 0.9991 for classification of lung parenchyma and non-lung-parenchyma. The obtain model can segment the lung parenchyma accurately for 201 subjects with heterogeneous lung diseases and CT scanners. The overlap ratio between the manual segmentation and the one by our method reaches 0.96. The results demonstrated that the proposed clustering algorithm based method can generate the training dataset for CNN models. The obtained CNN model can segment lung parenchyma with very satisfactory performance and have the potential to locate and analyze lung lesions.

Positron emission tomography (PET) is an advanced medical imaging technique widely used in various clinical applications, such as tumor detection and neurologic disorders. Reducing the radiotracer dose is desirable in PET imaging because it decreases the patient’s radiation exposure. However, reducing the dose can also increase noise, affecting the image quality. Therefore, an advanced image reconstruction algorithm based on low-dose PET data is needed to improve the quality of the reconstructed image. In this study, we propose the use of a direct PET image reconstruction network (DPIR-Net) using an improved Wasserstein generative adversarial network (WGAN) framework to enhance image quality. This study provides two main findings. First, our network uses sinogram data as input and outputs high-quality PET images direct, resulting in shorter reconstruction times compared with traditional model-based reconstruction networks. Second, we combine perceptual loss, mean square error, and the Wasserstein distance as the loss function, which effectively solves the problem of excessive smoothness and loss of detailed information in traditional network image reconstruction. We performed a comparative study using maximum-likelihood expectation-maximization (MLEM) with a post-Gaussian filter, a total variation (TV)-norm regularization, a nonlocal means (NLM) denoising method, a neural network denoising method, a traditional deep learning PET reconstruction network, and our proposed DPIR-Net method and evaluated the proposed method using both human and mouse data. The mouse data were obtained from a small-animal PET prototype system developed by our laboratory. The quantitative and qualitative results show that our proposed method outperformed the conventional methods.

Telemedicine Monitoring (TM) integrates mobile communication technology and Internet of Things (IoT) technology for health monitoring and data management. Amidst the escalating demand for telemedicine, traditional cloud computing struggles to guarantee real-time performance and data privacy. To address these challenges, we systematically survey the application of fog and edge computing technologies in TM systems. We focus on the following key aspects: (1) We delve into the theoretical foundations of fog and edge computing, underscoring their salient advantages including low latency, location awareness, high mobility, and more. (2) We elaborate on the architecture of a TM system hinged on fog and edge computing. (3) We outline key challenges facing fog/edge computing-based TM systems, including bandwidth limitations, low latency, data security, privacy, heterogeneity, and reliability. (4) We discuss the need for future advancements in the realms of security defense capability, system adaptability, and convergence of scheduling algorithms to refine the construction of the TM system and stimulate the development of telemedicine.

Due to the widespread of positron emission tomography (PET) in clinical practice, the potential risk of PET-associated radiation dose to patients needs to be minimized. However, with the reduction in the radiation dose, the resultant images may suffer from noise and artifacts that compromise diagnostic performance. In this article, we propose a parameter-transferred Wasserstein generative adversarial network (PT-WGAN) for low-dose PET image denoising. The contributions of this article are twofold: 1) a PT-WGAN framework is designed to denoise low-dose PET images without compromising structural details and 2) a task-specific initialization based on transfer learning is developed to train PT-WGAN using trainable parameters transferred from a pretrained model, which significantly improves the training efficiency of PT-WGAN. The experimental results on clinical data show that the proposed network can suppress image noise more effectively while preserving better image fidelity than recently published state-of-the-art methods. We make our code available at https://github.com/90n9-yu/PT-WGAN .

The main obstacle to image augmentation with Generative Adversarial Networks (GANs) is the need for a large amount of training data, but this is difficult for small datasets like Environmental Microorganisms (EMs). EM image analysis plays a vital role in environmental monitoring and protection, but it is often encountered with small datasets due to the difficulty of EM image collection. To this end, we propose an Enhanced Framework of GANs (EF-GANs) that combines geometric transformation methods and GANs for EM image augmentation. First of all, the color of an EM image has an insignificant impact on its class label, based on this fact, we perform color space augmentation to the original EM images. Secondly, we train EF-GANs with augmented EM images to generate utterly new EM images. Finally, we rotate the generated samples in various directions to obtain a more natural performance. In this study, we use VGG16 and ResNet50 networks to evaluate the proposed EF-GANs on 21 different types of EMs (420 EM images). It is observed that the average precision (AP) of VGG16 increases between 4.5% and 84.1% in 20 EM classes and one class remains unchanged. The AP of Resnet50 rises between 8.7% and 38.7% in 12 EM classes and reaches 100% in two EM classes. Furthermore, to reflect the generalization performance of EF-GANs, we employ an utterly new EM image dataset (630 EM images) to test the previous VGG16 networks. We select the VGG16 networks with original and optimal settings for all the EM classes, and for testing, optimal settings for a single EM class is considered. In the 20 of 21 one-vs-rest EM image classification tasks, the AP of VGG16 increases between 1.66% and 88.1%. The results demonstrate that the proposed EF-GANs can achieve outstanding performance in augmenting single EM images with high quality and resolution, thus, to improve the APs of EM image classification.