Pneumonia is among the top diseases which cause most of the deaths all over the world. Virus, bacteria and fungi can all cause pneumonia. However, it is difficult to judge the pneumonia just by looking at chest X-rays. The aim of this study is to simplify the pneumonia detection process for experts as well as for novices. We suggest a novel deep learning framework for the detection of pneumonia using the concept of transfer learning. In this approach, features from images are extracted using different neural network models pretrained on ImageNet, which then are fed into a classifier for prediction. We prepared five different models and analyzed their performance. Thereafter, we proposed an ensemble model that combines outputs from all pretrained models, which outperformed individual models, reaching the state-of-the-art performance in pneumonia recognition. Our ensemble model reached an accuracy of 96.4% with a recall of 99.62% on unseen data from the Guangzhou Women and Children’s Medical Center dataset.

A Novel Transfer Learning Based Approach for Pneumonia Detection in Chest X-ray Images

Lung cancer is the leading cause of cancer deaths worldwide and early diagnosis of lung nodule is of great importance for therapeutic treatment and saving lives. Automated lung nodule analysis requires both accurate lung nodule benign-malignant classification and attribute score regression. However, this is quite challenging due to the considerable difficulty of lung nodule heterogeneity modeling and the limited discrimination capability on ambiguous cases. To solve these challenges, we propose a Multi-Task deep model with Margin Ranking loss (referred as MTMR-Net) for automated lung nodule analysis. Compared to existing methods which consider these two tasks separately, the relatedness between lung nodule classification and attribute score regression is explicitly explored in a cause-and-effect manner within our multi-task deep model, which can contribute to the performance gains of both tasks. The results of different tasks can be yielded simultaneously for assisting the radiologists in diagnosis interpretation. Furthermore, a Siamese network with a margin ranking loss is elaborately designed to enhance the discrimination capability on ambiguous nodule cases. To further explore the internal relationship between two tasks and validate the effectiveness of the proposed model, we use the recursive feature elimination method to iteratively rank the most malignancy-related features. We validate the efficacy of our method MTMR-Net on the public benchmark LIDC-IDRI dataset. Extensive experiments show that the diagnosis results with internal relationship explicitly explored in our model has met some similar patterns in clinical usage and also demonstrate that our approach can achieve competitive classification performance and more accurate scoring on attributes over the state-of-the-arts. Codes are publicly available at: https://github.com/CaptainWilliam/MTMR-NET .

Multi-Task Deep Model With Margin Ranking Loss for Lung Nodule Analysis

A survey of computer-aided diagnosis of lung nodules from CT scans using deep learning.

Earth system science has changed rapidly due to global environmental changes and the advent of Earth observation technology. Therefore, new tools are required to monitor, measure, analyze, evaluate, and model Earth observation data. Google Earth (GE) was officially launched by Google in 2005 as a ”geobrowser”, and Google Earth Engine (GEE) was released in 2010 as a cloud computing platform with substantial computational capabilities. The use of these two tools or platforms in various applications, particularly as used by the remote sensing community, has developed rapidly. In this paper, we reviewed the applications and trends in the use of GE and GEE by analyzing peer-reviewed articles, dating up to January 2021, in the Web of Science (WoS) core collection using scientometric analysis (i.e., by using CiteSpace) and meta-analysis. We found the following: (1) the number of articles describing the use of GE or GEE increased substantially from two in 2006 to 530 in 2020. The number of GEE articles increased much faster than those concerned with the use of GE. (2) Both GE and GEE were extensively used by the remote sensing community as multidisciplinary tools. GE articles covered a broader range of research areas (e.g., biology, education, disease and health, economic, and information science) and appeared in a broader range of journals than those concerned with the use of GEE. (3) GE and GEE shared similar keywords (e.g., “land cover”, “water”, “model”, “vegetation”, and “forest”), which indicates that their application is of great importance in certain research areas. The main difference was that articles describing the use of GE emphasized its use as a visual display platform, while those concerned with GEE placed more emphasis on big data and time-series analysis. (4) Most applications of GE and GEE were undertaken in countries, such as the United States, China, and the United Kingdom. (5) GEE is an important tool for analysis, whereas GE is used as an auxiliary tool for visualization. Finally, in this paper, the merits and limitations of GE and GEE, and recommendations for further improvements, are summarized from an Earth system science perspective.

Progress and Trends in the Application of Google Earth and Google Earth Engine

Medical images are a rich source of invaluable necessary information used by clinicians. Recent technologies have introduced many advancements for exploiting the most of this information and use it to generate better analysis. Deep learning (DL) techniques have been empowered in medical images analysis using computer-assisted imaging contexts and presenting a lot of solutions and improvements while analyzing these images by radiologists and other specialists. In this paper, we present a survey of DL techniques used for variety of tasks along with the different medical image's modalities to provide critical review of the recent developments in this direction. We have organized our paper to provide significant contribution of deep leaning traits and learn its concepts, which is in turn helpful for non-expert in medical society. Then, we present several applications of deep learning (e.g., segmentation, classification, detection, etc.) which are commonly used for clinical purposes for different anatomical site, and we also present the main key terms for DL attributes like basic architecture, data augmentation, transfer learning, and feature selection methods. Medical images as inputs to deep learning architectures will be the mainstream in the coming years, and novel DL techniques are predicted to be the core of medical images analysis. We conclude our paper by addressing some research challenges and the suggested solutions for them found in literature, and also future promises and directions for further developments. 

https://link.springer.com/content/pdf/10.1007/s00530-021-00884-5.pdf

A holistic overview of deep learning approach in medical imaging

Remarkable progress has been made in image classification and segmentation, due to the recent study of deep convolutional neural networks (CNNs). To solve the similar problem of diagnostic lung nodule detection in low-dose computed tomography (CT) scans, we propose a new Computer-Aided Detection (CAD) system using CNNs and CT image segmentation techniques. Unlike former studies focusing on the classification of malignant nodule types or relying on prior image processing, in this work, we put raw CT image patches directly in CNNs to reduce the complexity of the system. Specifically, we split each CT image into several patches, which are divided into 6 types consisting of 3 nodule types and 3 non-nodule types. We compare the performance of ResNet with different CNNs architectures on CT images from a publicly available dataset named the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI). Results show that our best model reaches a high detection sensitivity of 92.8% with 8 false positives per scan (FPs/scan). Compared with related work, our work obtains a state-of-the-art effect.

Lung Nodule Detection in CT Images Using a Raw Patch-Based Convolutional Neural Network

Automatic ship detection in optical remote-sensing (ORS) images has wide applications in civil and military fields. Research on ship detection in ORS images started late compared to synthetic apert...

Ship detection based on fused features and rebuilt YOLOv3 networks in optical remote-sensing images

Optoelectronic devices are key building blocks for sustainable energy, imaging applications, and optical communications in modern society. Two-dimensional materials and perovskites have been considered promising candidates in this research area due to their fascinating material properties. Despite the significant progress achieved in the past decades, challenges still remain to further improve the performance of devices based on 2D materials or perovskites and to solve stability issues for their reliability. Recently, a novel concept of 2D material/perovskite heterostructure has demonstrated remarkable achievements by taking advantage of both materials. The diverse fabrication techniques and large families of 2D materials and perovskites open up great opportunities for structure modification, interface engineering, and composition tuning in state-of-the-art optoelectronics. In this review, we present comprehensive information on the synthesis methods, material properties of 2D materials and perovskites, and the research progress of optoelectronic devices, particularly solar cells and photodetectors which are based on 2D materials, perovskites, and 2D material/perovskite heterostructures with future perspectives.

/pdf/2d-material-and-perovskite-heterostructure-for-sqmtp57d.pdf

2D Material and Perovskite Heterostructure for Optoelectronic Applications

Domain adaptive semantic segmentation methods commonly utilize stage-wise training, consisting of a warm-up and a self-training stage. However, this popular approach still faces several challenges in each stage: for warm-up, the widely adopted adversarial training often results in limited performance gain, due to blind feature alignment; for self-training, finding proper categorical thresholds is very tricky. To alleviate these issues, we first propose to replace the adversarial training in the warm-up stage by a novel symmetric knowledge distillation module that only accesses the source domain data and makes the model domain generalizable. Surprisingly, this domain generalizable warm-up model brings substantial performance improvement, which can be further amplified via our proposed cross-domain mixture data augmentation technique. Then, for the self-training stage, we propose a threshold-free dynamic pseudo-label selection mechanism to ease the aforementioned threshold problem and make the model better adapted to the target domain. Extensive experiments demonstrate that our framework achieves remarkable and consistent improvements compared to the prior arts on popular benchmarks. Codes and models are available at https://github.com/fy-vision/DiGA

DiGA: Distil to Generalize and then Adapt for Domain Adaptive Semantic Segmentation

The present invention provides a pulmonary CT preprocessing method and system based on a convolutional neural network. The method comprises: slicing and numbering each original lung CT image in same order and size to obtain a plurality of image blocks; and dividing the image blocks into diseased nodules, no tissues or luminal walls, vascular tissues or other lung tissues, and pulmonary walls, andinputting the divided image blocks to the convolutional neural network for training. According to the technical scheme of the present invention, the image slice classification processing can reduce the system complexity, can ensure the integrity of the pulmonary nodule characteristics, can be well used for the training of the convolutional neural network, and has high sensitivity and a low misdiagnosis rate, so that the convolutional neural network can quickly identify and locate lung nodules in applications.

Fengyi Shen

Papers

Lung Nodule Detection in CT Images Using a Raw Patch-Based Convolutional Neural Network

Ship detection based on fused features and rebuilt YOLOv3 networks in optical remote-sensing images

2D Material and Perovskite Heterostructure for Optoelectronic Applications

DiGA: Distil to Generalize and then Adapt for Domain Adaptive Semantic Segmentation

Pulmonary CT preprocessing method and system based on convolutional neural network