Robust-Deep: A Method for Increasing Brain Imaging Datasets to Improve Deep Learning Models’ Performance and Robustness

A small dataset commonly affects generalization, robustness, and overall performance of deep neural networks (DNNs) in medical imaging research. Since gathering large clinical databases is always difficult, we proposed an analytical method for producing a large realistic/diverse dataset. Clinical brain PET/CT/MR images including full-dose (FD), low-dose (LD) corresponding to only 5 % of events acquired in the FD scan, non-attenuated correction (NAC) and CT-based measured attenuation correction (MAC) PET images, CT images and T1 and T2 MR sequences of 35 patients were included. All images were registered to the Montreal Neurological Institute (MNI) template. Laplacian blending was used to make a natural presentation using information in the frequency domain of images from two separate patients, as well as the blending mask. This classical technique from the computer vision and image processing communities is still widely used and unlike modern DNNs, does not require the availability of training data. A modified ResNet DNN was implemented to evaluate four image-to-image translation tasks, including LD to FD, LD+MR to FD, NAC to MAC, and MRI to CT, with and without using the synthesized images. Quantitative analysis using established metrics, including the peak signal-to-noise ratio (PSNR), structural similarity index metric (SSIM), and joint histogram analysis was performed for quantitative evaluation. The quantitative comparison between the registered small dataset containing 35 patients and the large dataset containing 350 synthesized plus 35 real dataset demonstrated improvement of the RMSE and SSIM by 29% and 8% for LD to FD, 40% and 7% for LD+MRI to FD, 16% and 8% for NAC to MAC, and 24% and 11% for MRI to CT mapping task, respectively. The qualitative/quantitative analysis demonstrated that the proposed model improved the performance of all four DNN models through producing images of higher quality and lower quantitative bias and variance compared to reference images.     

Electrochemical Corrosion Behaviour of X70 Steel under the Action of Capillary Water in Saline Soils

In this paper, the electrochemical corrosion behavior of X70 steel in saline soil under capillary water was simulated by a Geo-experts one-dimensional soil column instrument. A volumetric water content sensor and conductivity test were used to study the migration mechanism of water and salt (sodium chloride) under the capillary water. The electrochemical corrosion behavior of the X70 steel in the corrosion system was analyzed by electrochemical testing as well as the macroscopic and microscopic corrosion morphology of the steel. The test results showed that the corrosion behavior of X70 steel was significantly influenced by the rise of capillary water. In particular, the wetting front during the capillary water rise meant that the X70 steel was located at the three-phase solid/liquid/gas interface at a certain location, which worsened its corrosion behavior. In addition, after the capillary water was stabilized, the salts were transported with the capillary water to the top of the soil column. This resulted in the highest salt content in the soil environment and the most severe corrosion of the X70 steel at this location.     

Digital eye strain and its associated factors in children during the COVID-19 pandemic

Purpose:This study was undertaken to identify the prevalence of symptoms related to the use of display devices and contributing factors in children engaged in distance learning during the COVID-19 pandemic.Methods:An online electronic survey form was prepared using Google Forms (Alphabet Co., Mountain View, CA) and sent to parents of children under the age of 18 years engaged in distance learning during the COVID-19 pandemic. The types of display devices children use, how often such devices are used, the symptoms of digital eye strain, and the severity and frequency of the symptoms were recorded, and the associations between the factors were analyzed.Results:A total of 692 participants were included. The mean age of the children was 9.72 ± 3.02 years. The most common display devices used were personal computers (n = 435, 61.7%) for online classes and smartphones (n = 400, 57.8%) for nonacademic purposes. The mean duration of display device use was 71.1 ± 36.02 min without a break and 7.02 ± 4.55 h per day. The most common reported symptom was headache (n = 361, 52.2%). Of the participants, 48.2% (n = 332) reported experiencing 3 or more symptoms. The multivariate analysis detected that being male (P = 0.005) and older age (P = 0.001) were independent risk factors for experiencing 3 or more symptoms.Conclusion:The increasing use of digital devices by children is exacerbating the problem of digital eye strain in children as a side effect of online learning. Public awareness should be improved.     

Activity Subspaces in Medial Prefrontal Cortex Distinguish States of the World

Medial prefrontal cortex (mPfC) activity represents information about the state of the world, including present behavior, such as decisions, and the immediate past, such as short-term memory. Unknown is whether information about different states of the world are represented in the same mPfC neural population and, if so, how they are kept distinct. To address this, we analyze here mPfC population activity of male rats learning rules in a Y-maze, with self-initiated choice trials to an arm end followed by a self-paced return during the intertrial interval (ITI). We find that trial and ITI population activity from the same population fall into different low-dimensional subspaces. These subspaces encode different states of the world: multiple features of the task can be decoded from both trial and ITI activity, but the decoding axes for the same feature are roughly orthogonal between the two task phases, and the decodings are predominantly of features of the present during the trial but features of the preceding trial during the ITI. These subspace distinctions are carried forward into sleep, where population activity is preferentially reactivated in post-training sleep but differently for activity from the trial and ITI subspaces. Our results suggest that the problem of interference when representing different states of the world is solved in mPfC by population activity occupying different subspaces for the world states, which can be independently decoded by downstream targets and independently addressed by upstream inputs.SIGNIFICANCE STATEMENT Activity in the medial prefrontal cortex plays a role in representing the current and past states of the world. We show that during a maze task, the activity of a single population in medial prefrontal cortex represents at least two different states of the world. These representations were sequential and sufficiently distinct that a downstream population could separately read out either state from that activity. Moreover, the activity representing different states is differently reactivated in sleep. Different world states can thus be represented in the same medial prefrontal cortex population but in such a way that prevents potentially catastrophic interference between them.     

Study on Hot Deformation Behavior of an Antibacterial 50Cr15MoVCu Tool Steel

Hot deformation behaviors of an antibacterial 50Cr15MoVCu tool steel were studied. The flow stress curves presented three typical characteristics: (i) a single peak dynamic recrystallization curve, (ii) a monotone incremental work-hardening curve, and (iii) the equilibrium dynamic recovery curve. The flow stress increased with the increase of the deformation rate at each deformation temperature and decreased with the increase of the deformation temperature at the same deformation rate. The thermal activation energy and material constants were Q of 461.6574 kJ/mol, A of 3.42 × 10¹⁷, and α of 0.00681 MPa⁻¹, respectively. The high temperature constitutive equation was: Z=ε˙expQ/RT=3.42×1017sinh0.0068×σ5.6807. Based on the processing maps and microstructure evolution, the best hot working process was a deformation temperature of 1050 °C and deformation rate of 0.001 s⁻¹.     

Recommendations for Better Civic Learning: Building and Rebuilding Democracy

This is the concluding essay for a special report from The Hastings Center entitled Democracy in Crisis: Civic Learning and the Reconstruction of Common Purpose, which grew out of a project supported by the John S. and James L. Knight Foundation. This essay provides an integrative discussion of various theoretical and practical reform perspectives offered by other essays in the report. It also offers a number of recommendations. It notes that the aim of the special report is not to propose specific reform measures but, rather, to consider larger, more theoretic concerns related to political and economic questions, which are personal and structural—psychological, cultural, and institutional—at the same time. In response, this essay argues that the best relationship between the citizenry and government in a democracy is not one of deference, nor one of contestation, but one that is critically constructive, which in turn is linked to practices of civic learning. To be constructive, citizens need scientific literacy, an understanding of how government and other institutions work, critical thinking abilities, and many open and diverse forums for civic learning to offset the increasingly isolating media “bubbles” that are the only source of information for many. The essay then formulates five recommendations designed to facilitate critically constructive citizenship and civic learning. These are creating a basis for civic participation, acquiring information, talking to each other, designing institutional change, and achieving deliberation.     

洋金花最佳采收期评价

目的:探索中药材洋金花最佳采收期评价新方法的可行性。方法:将洋金花的生长过程进行Logistic生长曲线方程拟合,建立数学模型,运用有序聚类分析的方法将其生长过程划分并将各划分期拟定为采收期,以生物活性成分为指标,结合产量对洋金花最佳采收期进行评价。结果:洋金花的生长过程与Logistic生长曲线方程拟合度良好,有序聚类分析后将洋金花采收期拟定为五个时期,评价后发现与以往采收期试验研究结果有差别。结论:洋金花药材最佳采收期评价新方法可行。     

异氟烷对幼小鼠自主活动和学习记忆行为的影响

目的观察异氟烷(Iso)对幼小鼠自主活动和学习记忆行为的影响,探讨其与γ-氨基丁酸A受体(GABAA)的关系。方法 360只小鼠分为3大组,分别进行自主活动实验、避暗实验和跳台实验,每大组按分层随机区组设计分为正常对照组,Iso 0.05,0.1和0.2 ml·kg-1组、GABAA受体特异性阻断剂一叶萩碱(Sec)2,4和8 mg.kg-1组和Iso 0.2 m.lkg-1+Sec 2,4和8 mg.kg-1组。小鼠sc给予Sec 10 min后,再ip给予Iso,测试给药后15,30,45和60 min小鼠5 min内活动次数;跳台仪和避暗仪记录小鼠步入、跳下的潜伏期和错误次数。结果与同一时间点的正常对照组相比,Iso可减少小鼠自主活动次数,给予Iso后1 d小鼠避暗实验和跳台实验的步入和跳下潜伏期缩短以及错误次数增加(P<0.05)。正常小鼠单独给Sec(除Sec 8 mg·kg-1组15 min外)对小鼠自主活动、步入和跳下潜伏期和错误次数无明显影响,但Sec可改善Iso导致的小鼠自主活动次数,拮抗Iso对小鼠自主活动的影响,但随着时间延长拮抗作用逐渐减弱,明显低于正常对照组(P<0.05)。Sec延长Iso小鼠的跳下和步入潜伏期,减少错误次数,基本恢复至正常对照组水平,明显改善Iso对幼小鼠学习记忆的影响(P<0.05)。结论 Iso可降低幼小鼠自主活动次数,损害学习记忆能力,GABAA受体可能部分参与了以上作用。