Nuances of Interpreting X-ray Analysis by Deep Learning and Lessons for Reporting Experimental Findings

With the increase in the availability of annotated X-ray image data, there has been an accompanying and consequent increase in research on machine-learning-based, and ion particular deep-learning-based, X-ray image analysis. A major problem with this body of work lies in how newly proposed algorithms are evaluated. Usually, comparative analysis is reduced to the presentation of a single metric, often the area under the receiver operating characteristic curve (AUROC), which does not provide much clinical value or insight and thus fails to communicate the applicability of proposed models. In the present paper, we address this limitation of previous work by presenting a thorough analysis of a state-of-the-art learning approach and hence illuminate various weaknesses of similar algorithms in the literature, which have not yet been fully acknowledged and appreciated. Our analysis was performed on the ChestX-ray14 dataset, which has 14 lung disease labels and metainfo such as patient age, gender, and the relative X-ray direction. We examined the diagnostic significance of different metrics used in the literature including those proposed by the International Medical Device Regulators Forum, and present the qualitative assessment of the spatial information learned by the model. We show that models that have very similar AUROCs can exhibit widely differing clinical applicability. As a result, our work demonstrates the importance of detailed reporting and analysis of the performance of machine-learning approaches in this field, which is crucial both for progress in the field and the adoption of such models in practice.

Fully distributed rainfall-runoff modeling using spatial-temporal graph neural network

Recent studies using latest deep learning algorithms such as LSTM (Long Short-Term Memory) have shown great promise in time-series modeling. There are many studies focusing on the watershed-scale rainfall-runoff modeling or streamflow forecasting, often considering a single watershed with limited generalization capabilities. To improve the model performance, several studies explored an integrated approach by decomposing a large watershed into multiple sub-watersheds with semi-distributed structure. In this study, we propose an innovative physics-informed fully-distributed rainfall-runoff model, NRM-Graph (Neural Runoff Model-Graph), using Graph Neural Networks (GNN) to make full use of spatial information including the flow direction and geographic data. Specifically, we applied a time-series model on each grid cell for its runoff production. The output of each grid cell is then aggregated by a GNN as the final runoff at the watershed outlet. The case study shows that our GNN based model successfully represents the spatial information in predictions. NRM-Graph network has shown less over-fitting and a significant improvement on the model performance compared to the baselines with spatial information. Our research further confirms the importance of spatially distributed hydrological information in rainfall-runoff modeling using deep learning, and we encourage researchers to incorporate more domain knowledge in modeling.

Concatenated Convolutional Neural Network Model for Cuffless Blood Pressure Estimation Using Fuzzy Recurrence Properties of PPG Signals

Due to the importance of continuous monitoring of blood pressure (BP) in controlling hypertension, the topic of cuffless blood pressure (BP) estimation has been widely studied in recent years. A most important approach is to explore the nonlinear mapping between the recorded peripheral signals and the BP values which is usually conducted by deep neural networks. Because of the sequence-based pseudo periodic nature of peripheral signals such as photoplethysmogram (PPG), a proper estimation model needed to be equipped with the 1-dimensional (1-D) and recurrent layers. This, in turn, limits the usage of 2-dimensional (2-D) layers adopted in convolutional neural networks (CNN) for embedding spatial information in the model. In this study, considering the advantage of chaotic approaches, the recurrence characterization of peripheral signals was taken into account by a visual 2-D representation of PPG in phase space through fuzzy recurrence plot (FRP). FRP not only provides a beneficial framework for capturing the spatial properties of input signals but also creates a reliable approach for embedding the pseudo periodic properties to the neural models without using recurrent layers. Moreover, this study proposes a novel deep neural network architecture that combines the morphological features extracted simultaneously from two upgraded 1-D and 2-D CNNs capturing the temporal and spatial dependencies of PPGs in systolic and diastolic BP estimation. The model has been fed with the 1-D PPG sequences and the corresponding 2-D FRPs from two separate routes. The performance of the proposed framework was examined on the well-known public dataset, namely, Multi-Parameter Intelligent in Intensive Care II. Our scheme is analyzed and compared with the literature in terms of the requirements of the standards set by the British Hypertension Society (BHS) and the Association for the Advancement of Medical Instrumentation (AAMI). The proposed model met the AAMI requirements, and it achieved a grade of A as stated by the BHS standard. In addition, its mean absolute errors (MAE) and standard deviation for both systolic and diastolic blood pressure estimations were considerably low, 3.05±5.26 mmHg and 1.58±2.6 mmHg, in turn.

Flexible cue anchoring strategies enable stable head direction coding in both sighted and blind animals

Vision plays a crucial role in instructing the brain's spatial navigation systems. However, little is known about how vision loss affects the neuronal encoding of spatial information. Here, recording from head direction (HD) cells in the anterior dorsal nucleus of the thalamus in mice, we find stable and robust HD tuning in blind animals. In contrast, placing sighted animals in darkness significantly impairs HD cell tuning. We find that blind mice use olfactory cues to maintain stable HD tuning and that prior visual experience leads to refined HD cell tuning in blind adult mice compared to congenitally blind animals. Finally, in the absence of both visual and olfactory cues, the HD attractor network remains intact but the preferred firing direction of HD cells continuously drifts over time. We thus demonstrate remarkable flexibility in how the brain uses diverse sensory information to generate a stable directional representation of space.

Arabidopsis cyclophilins direct plasmodesmata-targeting of mobile mRNA via organelle hitchhiking

Plants selectively transport mobile mRNAs through intercellular pores, plasmodesmata (PD), to distribute spatial information for synchronizing meristematic differentiation with environmental dynamics. However, how plants recognize and deliver mobile mRNAs to PD remains unknown. Here, by using RNA-live cell imaging, we show that mobile mRNAs hitchhike on organelle trafficking to transport to PD. Perturbed cytoskeleton organization or organelle trafficking severely disrupts the subcellular distribution of mobile mRNAs. We further show that Arabidopsis rotamase cyclophilins (ROCs), which are organelle-localized RNA-binding proteins (RBPs), specifically bind mobile mRNAs on the surface of organelles to direct PD-targeting. Arabidopsis roc quadruple mutants showed reduced in PD-targeting of mobile mRNAs, along with phenotype alterations. ROCs can move intercellularly and form RNA-protein complexes in phloem, suggesting the roles of ROCs in delivery of mobile mRNAs through PD. Our results highlight that an RBP-mediated hitchhiking system is purposely recruited to orient plant-mobile mRNAs to PD for intercellular transport.

BSIRNet: A Road Extraction Network with Bidirectional Spatial Information Reasoning

Automatic extraction of road information from remote sensing images is widely used in many fields, such as urban planning and automatic navigation. However, due to interference from noise and occlusion, the existing road extraction methods can easily lead to road discontinuity. To solve this problem, a road extraction network with bidirectional spatial information reasoning (BSIRNet) is proposed, in which neighbourhood feature fusion is used to capture spatial context dependencies and expand the receptive field, and an information processing unit with a recurrent neural network structure is used to capture channel dependencies. BSIRNet enhances the connectivity of road information through spatial information reasoning. Using the public Massachusetts road dataset and Wuhan University road dataset, the superiority of the proposed method is verified by comparing its results with those of other models.

Remote Sensing Image Fusion Algorithm Based on Two-Stream Fusion Network and Residual Channel Attention Mechanism

A two-stream remote sensing image fusion network (RCAMTFNet) based on the residual channel attention mechanism is proposed by introducing the residual channel attention mechanism (RCAM) in this paper. In the RCAMTFNet, the spatial features of PAN and the spectral features of MS are extracted, respectively, by a two-channel feature extraction layer. Multiresidual connections allow the network to adapt to a deeper network structure without the degradation. The residual channel attention mechanism is introduced to learn the interdependence between channels, and then the correlation features among channels are adapted on the basis of the dependency. In this way, image spatial information and spectral information are extracted exclusively. What is more, pansharpening images are reconstructed across the board. Experiments are conducted on two satellite datasets, GaoFen-2 and WorldView-2. The experimental results show that the proposed algorithm is superior to the algorithms to some existing literature in the comparison of the values of reference evaluation indicators and nonreference evaluation indicators.

effectiveness of learning materials implementation for spatial thinking ability in geographic information systems (GIS) course

Geography education is realized in learning that combines the study of physical and human geography in a spatial context. GIS Learning in universities is directed to be able to equip students in the use of spatial information which must be accompanied by the ability to manage it cognitively. This study aimed to assess the effectiveness of Spatial Thinking Ability learning materials development. The model used in this research and development study was the Dick & Carey model. Field trial was carried out by experimental research using a Quasi Experiment model. The trial design was Post-test Only, Non-Equivalent Control Group Design. The trial was carried out on sixth-semester undergraduate students at Muhammadiyah Surakarta who had taken a GIS course with as many as 41 students. The activity took place in March 2018. In this study, some students were given treatment in spatial thinking ability learning. The result shows a U value of 56 and a W value of 209. When converted to a Z value, the value is -3.943. Sig value or P-Value of 0.000 <0.05.

Spatially resolved transcriptomics in immersive environments

Spatially resolved transcriptomics is an emerging class of high-throughput technologies that enable biologists to systematically investigate the expression of genes along with spatial information. Upon data acquisition, one major hurdle is the subsequent interpretation and visualization of the datasets acquired. To address this challenge, VR-Cardiomicsis presented, which is a novel data visualization system with interactive functionalities designed to help biologists interpret spatially resolved transcriptomic datasets. By implementing the system in two separate immersive environments, fish tank virtual reality (FTVR) and head-mounted display virtual reality (HMD-VR), biologists can interact with the data in novel ways not previously possible, such as visually exploring the gene expression patterns of an organ, and comparing genes based on their 3D expression profiles. Further, a biologist-driven use-case is presented, in which immersive environments facilitate biologists to explore and compare the heart expression profiles of different genes.

Point Cloud Semantic Segmentation with Cross-Correction Features

Real time large scale point cloud segmentation is an important but challenging task for practical application like autonomous driving. Existing real time methods have achieved acceptance performance by aggregating local information. However, most of them only exploit local spatial information or local semantic information dependently, few considering the complementarity of both. In this paper, we propose a model named Spatial-Semantic Incorporation Network (SSI-Net) for real time large scale point cloud segmentation. A Spatial-Semantic Cross-correction (SSC) module is introduced in SSI-Net as a basic unit. High quality contextual features can be learned through SSC by correct and update semantic features using spatial cues, and vice verse. Adopting the plug-and-play SSC module, we design SSI-Net as an encoder-decoder architecture. To ensure efficiency, it also adopts a random sample based hierarchical network structure. Extensive experiments on several prevalent datasets demonstrate that our method can achieve state-of-the-art performance.