Search

Study Design Comparative studyObjective To compare manual and deep learning-based automated measurement of Cobb angle in adolescent idiopathic scoliosis.Methods We proposed a fully automated framework to measure the Cobb angle of AIS patients. Whole-spine images of 500 AIS individuals were collected. 200 digital radiographic (DR) images were labeled manually as training set, and the remaining 300 images were used to validate by mean absolute error (MAE), Pearson or spearman correlation coefficients, and intra/interclass correlation coefficients (ICCs). The relationship between accuracy of vertebral boundary identification and the subjective image quality score was evaluated.Results The PT, MT, and TL/L Cobb angles were measured by the automated framework within 300 milliseconds. Remarkable 2.92° MAE, .967 ICC, and high correlation coefficient (r = .972) were obtained for the major curve. The MAEs of PT, MT, and TL/L were 3.04°, 2.72°, and 2.53°, respectively. The ICCs of these 3 curves were .936, .977, and .964, respectively. 88.7% (266/300) of cases had a difference range of ±5°, with 84.3% (253/300) for PT, 89.7% (269/300) for MT, and 93.0% (279/300) for TL/L. The decreased bone/soft tissue contrast (2.94 vs 3.26; P=.039) and bone sharpness (2.97 vs 3.35; P=.029) were identified in the images with MAE exceeding 5°.Conclusion The fully automated framework not only identifies the vertebral boundaries, vertebral sequences, the upper/lower end vertebras and apical vertebra, but also calculates the Cobb angle of PT, MT, and TL/L curves sequentially. The framework would shed new light on the assessment of AIS curvature.

AbstractDisplacement prediction is a direct and effective method for mitigating geohazards. Due to the influence of rainfall and reservoir water level variations, landslides often display step-like deformations with an increasing trend and periodic fluctuation, indicating long-term memory in displacement time series. Traditional data-driven methods are mostly suitable for short-term prediction, and extra data processing is applied to solve this problem. This paper proposes a novel deep learning-based displacement prediction method using long short-term memory (LSTM) networks. Based on open-source frameworks for deep learning, namely, Keras and TensorFlow, a detailed implementation of displacement prediction is proposed and illustrated. The Baishuihe landslide, a typical landslide with long-term monitoring, is taken as a case study, and both single-factor and multi-factor predictions are performed. The results indicate that multi-factor prediction can reduce overfitting and improve accuracy. Compared with the existing method, the multi-factor deep-learning model displays better performance. This study indicates that the LSTM-based deep-learning model is suitable and convenient for displacement prediction and has broad prospects in safety monitoring of water-induced landslides.

AbstractThe triaxial compression tests under different hydraulic pressures and low confining pressure were carried out to study the hydraulic pressure effect on the mechanical properties and permeabilities of layered rock mass. In general, the results show that the increase of hydraulic pressure reduces the peak strength of layered rock mass and the layer structure changes the mechanical properties of rock mass. With the increase of hydraulic pressure, the peak strength of layered rock mass decreases sharply and then slows down. The peak permeabilities of layered rock mass occurred later than the time when the strength come to the peak. The increase of hydraulic pressure will accelerate the propagation of cracks and significantly enhance the peak permeability of layered rock mass. With the increase of the layer number, the effect of hydraulic pressure becomes more distinct. The initial permeability and the peak permeability increase largely. The increase of hydraulic pressure promotes the expansion and penetration of cracks, so the rock mass develops from tensile failure to the composite failure forms including shear failure and interfacial shear slip failure.

Thyroid nodule is a common thyroid disease, but the study of its pathology and pathogenesis is still limited. As a non-invasive diagnostic method, medical image examination is of great value to study the pathological correlation of thyroid nodules. The purpose of this study was to investigate the expression of eosinophils in medical image examination and the pathological correlation between eosinophils and thyroid nodules. The study analyzed the pathological reports of a group of patients with thyroid nodules examined by medical images and performed corresponding imaging scans or examinations. The imaging data is processed, including image reconstruction, data transmission and other steps, to generate images that can be diagnosed by doctors. Thyroid function and parameters of leukocyte were collected and compared.The serum levels of TT4 and fT4 were observed lower in G2 group, while thyroid stimulating hormone (TSH) was higher compared to G1 group before surgery. Compared to G2 group, eosinophils count and percentage were lower in G1group (p< 0.05) post-surgery and lower ratio of eosinophils count with lymphocyte count (ELR) were observed in G1 group patients (p< 0.05).Elevated TSH is closely related to malignant TN per surgery, while lower ELR suggesting that TN removed thoroughly. Relevant cut-off values required further study to guide the diagnosis, treatment and follow-up of TN.

Strains from the Cryptococcus gattiispecies complex (CGSC) have caused the Pacific Northwest cryptococcosis outbreak, the largest cluster of life-threatening fungal infections in otherwise healthy human hosts known to date. In this study, we utilized a pan-phenome-based method to assess the fitness outcomes of CGSC strains under 31 stress conditions, providing a comprehensive overview of 2,821 phenotype-strain associations within this pathogenic clade. Phenotypic clustering analysis revealed a strong correlation between distinct types of stress phenotypes in a subset of CGSC strains, suggesting that shared determinants coordinate their adaptations to various stresses. Notably, a specific group of strains, including the outbreak isolates, exhibited a remarkable ability to adapt to all three of the most commonly used antifungal drugs for treating cryptococcosis (amphotericin B, 5-fluorocytosine, and fluconazole). By integrating pan-genomic and pan-transcriptomic analyses, we identified previously unrecognized genes that play crucial roles in conferring multidrug resistance in an outbreak strain with high multidrug adaptation. From these genes, we identified biomarkers that enable the accurate prediction of highly multidrug-adapted CGSC strains, achieving maximum accuracy and area under the curve (AUC) of 0.79 and 0.86, respectively, using machine learning algorithms. Overall, we developed a pan-omic approach to identify cryptococcal multidrug resistance determinants and predict highly multidrug-adapted CGSC strains that may pose significant clinical concern.

Temperature of a plant organ constitutes an integrative index to its eco-physiological properties and status. However, little attempt has been made to dissect the combined effect of ecological and physiological factors on the surface temperature of a plant organ such as the rice spikelet. In this study, using a deactivated plant as reference, we developed a novel comparison method to dissect the environmental and physiological effects on temperature of rice spikelet. Three japonicarice cultivars with contrasting canopy features were used as testing materials. Temperatures of flag leaf, superior and inferior spikelets and their diurnal rhythm during grain filling stage were precisely measured by a hand-held infrared thermometer. The results showed that the variation of environmental conditions within a panicle was relatively minor, posing a limited influence on temperature difference between the superior and inferior spikelet. On the other hand, it was the intrinsic physiological properties that considerably affected the spatial variations of spikelet temperature within a panicle. Chemical analysis of sucrose and starch in grains and bracts indicated that the superior spikelet is more physiologically active at photosynthetic assimilation and starch biosynthesis. Interestingly, sugar in bracts exhibited a pattern of diurnal changes similar to the source leaf but different from the sink grain, confirming that bracts are source organs for grain filling. Our findings yield penetrating insight into the eco-physiological foundation of spikelet temperature, thus being helpful for the application of physiological approaches in crop breeding for cooler canopy.

At the plasma membrane, in response to biotic and abiotic cues, specific ligands initiate the formation of receptor kinase heterodimers, which regulate activities of plasma membrane proteins and initiate signaling cascades to the nucleus. In this study, we utilized affinity enrichment mass spectrometry (AE-MS) to investigate the stimulus-dependent interactomes of LRR receptor kinases in response to their respective ligands, with an emphasis on exploring structural influences and potential cross-talk events at the plasma membrane. BRI1 and SIRK1 were chosen as receptor kinases with distinct coreceptor preference. By using interactome characteristic of domain-swap chimera following a gradient boosting learning algorithm trained on SIRK1 and BRI1 interactomes, we attribute contributions of extracellular domain, transmembrane domain, juxtamembrane domain and kinase domain of respective ligand-binding receptors to their interaction with their coreceptors and substrates. Our results revealed juxtamembrane domain as major structural element defining the specific substrate recruitment for BRI1 and extracellular domain for SIRK1. Furthermore, the learning algrorithm enabled us to predict the phenotypic outcomes of chimeric receptors based on different domain combinations, which was verified by dedicated experiments. As a result, our work reveals a tightly controlled balance of signaling cascade activation dependent on ligand-binding receptors domains and the internal ligand status of the plant. Moreover, our study shows the robust utility of machine learning classification as a quantitative metric for studying dynamic interactomes, dissecting the contribution of specific domains and predicting their phenotypic outcome.