关于我国设立刑事证据开示制度的思考

1996年《刑事诉讼法》的修改，产生了律师阅卷难问题。为解决这一问题，真正实现控辩双方公平对抗，我国应设立刑事证据开示制度。文章对刑事证据开示制度的内涵、设立的必要性、设立的原则以及具体建议进行法律思考。     

Transplantation of human amnion epithelial cells improves learning and memory function in Alzheimer's disease-like pathology rat model

AIM: To observe the treatment effect and its immune regulation of human amnion epithelial cells (hAECs) on Alzheimer's disease (AD)-like pathology rat model. METHODS: The hAECs were isolated from amnion with trypsin digestion, and the phenotype of hAECs was analyzed by flow cytometry. SD rats (n=48) were randomly divided into sham control group, model group, medium group and hAECs group. AD-like pathology rat model was induced by bilateral intraventricular injection of lipopolysaccharide (LPS). hAECs (5×10⁵) were injected into the hippocampus of the AD-like pathology rats. At 2 weeks after transplantation, the animals were tested by Morris water maze to observe the function of learning and memory. The pathological change of the brain was observed by HE staining. The expression of amyloid β-protein 42(Aβ42) and Tau protein and the level of acetylcholine (ACh) in the injury brain were determined by immunohistochemistry. The survival and differentiation of hAECs in the hippocampus were measured by immunofluorescent technique. The percentages of lymphocyte subsets in the peripheral blood mononuclear cells were analyzed by flow cytometry. The contents of serum cytokines were detected by cytometric bead array. RESULTS: Compared with model group and medium group, hAECs group showed shortened escape latency (P<0.01), increased frequency of going through the platform (P<0.05), reduced loss of hippocampal neurons, decreased expression of Tau protein and Aβ42 in the hippocampus (P<0.05), increased ACh level in the hippocampus (P<0.05), decreased percentages of Th1 and Th17 subsets, increased percentages of Th2 and Treg cells (P<0.05), decreased concentrations of IFN-γ and IL-2 in the serum, and increased concentration of IL-4(P<0.05). CONCLUSION: hAECs improve the cognitive learning and memory function and alleviate pathologic damage of hippocampus through immune regulation in AD-like pathology rats.     

Soil aggregate formation and stability induced by starch and cellulose

Soil aggregate (SA) can be formed and stabilized when soil organic matter (SOM) is decomposed in the soil. However, the relationships between the SA dynamics and SOM with different decomposition rates have not been clarified. Therefore, this study examined the effects of the addition of polysaccharides to soil on SA formation and stability. A Japanese tropical soil was incubated for 99 d at 30 °C in a dark environment following the addition of 0.5% (w/w) starch or cellulose. The decomposition rates of the amendments, and SA formation and stability were evaluated by measuring soil respiration rates, and distribution fractions of soil aggregate sizes and mean weight diameter (MWD) of SA, respectively. The cumulative soil respirations with all treatments rapidly increased until Day 12 of the incubation. The initial slope of the cumulative soil respiration in the soil with starch was significantly higher than that in the soil with cellulose. In either soil with starch or cellulose, the fractions of macro-aggregates (>1000 μm in diameter) significantly increased, respectively, compared with control soil. However, the fractions of meso-aggregates (250–1000 μm) and nano-aggregate (<20 μm) in the soil with starch significantly decreased, while those fractions in the soil with cellulose fluctuated until Day 6. The MWDs reached the maximum on Day 6, indicating the SA formation in the soils with starch or cellulose. The increasing rate of the SA formation in the starch-amended soil was greatly higher than that in the cellulose-amended soil. After Day 6, the MWDs in the soils with either polysaccharide decreased with similar trends with no significant differences between treatments, indicating similar stability of the SA in both treatments. This study showed that the different decomposability of the organic amendments might influence the SA formation differently, but not the SA stability.     

Antioxidant,anti-amylase and anti-glycation potential of brans of some Sri Lankan traditional and improved rice (Oryza sativa L.) varieties

Brans of 23 traditional and 12 improved (both red and white) rice varieties in Sri Lanka were screened for anti-amylase and anti-glycation activities in vitro. Varieties which showed the highest inhibitory activities at screening were further investigated for anti-glucosidase and glycation reversing as anti-diabetic properties. The same varieties were studied for selected antioxidant properties. Significantly high anti-amylase and anti-glycation activities were observed for bran extracts of red varieties compared to white varieties at screening. Traditional red rice varieties, Masuran, Sudu Heeneti, Dik Wee and Goda Heeneti, exhibited significant and dose dependent anti-amylase, anti-glycation and glycation reversing activities. These varieties also showed marked antioxidant properties. It is concluded that brans of Sri Lankan traditional red rice varieties Masuran, Sudu Heeneti, Dik Wee and Goda Heeneti may be potential food supplements for diabetes.     

基于CNN-BiLSTM的棉花产量预测方法

针对传统的作物产量估算方法在模型泛化方面的不足及缺少时序特征和空间特征的问题,该研究以机采棉为研究对象,结合无人机遥感平台与深度学习技术对棉花进行多期遥感观测与估产。以棉花苗期、蕾期和花期的影像为时间序列数据集,构建基于卷积神经网络（Convolutional Neural Network,CNN）和双向长短期记忆（Bidirectional Long Short-Term Memory,BiLSTM）的混合模型（CNN-BiLSTM）进行棉花产量预估,提高时间维度和空间维度方面的特征提取能力,并分别验证CNN和BiLSTM的性能以及不同网络深度对估产的影响。研究结果表明,CNN和BiLSTM深度分别为14和1的CNN14-BiLSTM模型准确率最高,对比单一结构的BiLSTM该模型决定系数从0.851提升至0.885,其均方根误差和平均绝对百分比误差也均明显下降,在2.3 m×2.3 m的样方内,结果分别为147.167 g和6.711%。该研究实现了基于时间序列的棉花产量估算,可为类似的作物产量估算提供参考。     

基于MFO-LSTM的母猪发情行为识别

及时准确识别母猪的发情行为可以有效增加受胎率和产仔量,对提高养殖企业的繁育水平和经济效益具有重要意义。该研究针对生猪养殖过程中母猪发情行为识别存在主观性强、智能化水平低、假警报和错误率高、识别不及时等问题,提出了一种基于飞蛾扑火算法(Moth-Flame Optimization,MFO)优化长短时记忆网络(Long Short Term Memory,LSTM)的母猪发情行为识别方法。利用安装在母猪颈部的姿态传感器获得母猪姿态数据,然后使用姿态数据训练MFO-LSTM姿态分类模型,将母猪姿态分为立姿、卧姿和爬跨3类。通过对姿态分类结果进行分析,确定以爬跨行为和活动量2个特征作为发情行为识别依据,使用MFO-LSTM分类算法判断母猪是否发情。以山西省太原市杏花岭区五丰养殖场的试验数据对该方法进行验证,结果表明,该方法在以30 min为发情行为识别时间时的识别效果最好,发情行为识别的错误率为13.43%,召回率为90.63%,特效性为81.63%,与已有的母猪发情行为识别方法相比错误率降低了80%以上。该方法在保证识别准确率的情况下有效降低了错误率,可满足母猪养殖生产过程中发情行为自动识别要求。     

智能气动肌肉的静态驱动特性研

研究了形状记忆合金丝(SMA)编织网的主动变形及对气动肌肉静态驱动特性的影响.建立了SMA丝内应力与静态驱动力的平衡方程.针对升降温中SMA丝的相变,分析了马氏体和奥氏体体积比变化,建立了温度SMA收缩率模型,并应用到气动肌肉的收缩率、收缩力和刚密度特性计算中.Matlab仿真得到了SMA变形曲线和气动肌肉特性变化曲面.分析结果表明SMA变形存在迟滞,SMA主动伸缩使得编织角变化范围更广,刚密度变化更突出.SMA收缩率变大,气动肌肉收缩力增强.     

基于SSA-LSTM的玉米土壤含氧量预测模型

土壤含氧量（Soil oxygen content,SOC）是影响作物生长的重要土壤环境因素之一,具有时序性、不稳定性和非线性等特点,精确预测土壤环境中含氧量的变化趋势,有助于制定更加合理的土壤通气增氧方案。本研究提出基于麻雀搜索算法（Sparrow search algorithm,SSA）和长短时记忆（Long and short-term memory,LSTM）神经网络预测模型,利用国家土壤质量湛江观测实验站记录玉米种植期间的气象环境和土壤环境数据,基于SSA-LSTM模型对SOC变化进行预测及相关性分析,并与传统的BP预测模型、LSTM预测模型、GA-LSTM预测模型及PSO-LSTM预测模型进行对比。试验结果表明,SOC与降雨量、土壤含水率、土壤温度、土壤充气孔隙度相关性极显著,相关系数高于0.8,与大气温度和风速相关性显著,与大气湿度和土壤呼吸速率相关性较弱。SSA-LSTM模型预测精度明显高于其他4组对照预测模型,R2达到0.95979,RMSE仅为0.4917%,MAPE为3.7331%,MAE为 0.3620%,预测值与试验值之间的拟合程度高。本研究可为土壤含氧量变化的精准预测及土壤通气增氧技术的应用推广提供理论支撑与科学依据。     

家禽诊疗文本多实体关系联合抽取模型研究

针对传统实体关系抽取方法中主体特征与句向量难以有效融合、现有BIO标注策略难以有效处理重叠关系的问题,提出一种基于BERT和双重指针标注的家禽疾病诊疗文本实体关系联合抽取模型(Joint extraction of entity relationship of poultry disease diagnosis and treatment text,JEER_PD)。JEER_PD使用双重指针标注(Dual-pointer labeling, DPL)策略,建立头、尾2个指针标注器,一次性标注出所有实体的开始和结束位置;引入CLN(Conditional layer normalization)网络层,强化主体抽取任务与客体关系联合抽取任务之间的联系;利用概率平衡策略PBS对抗正负类标签类别失衡,以加速模型收敛。实验表明,JEER_PD准确率、召回率和F1分别为97.69%、97.59%和97.64%,3项指标较现有方法均有显著提升,说明JEER_PD能够快速、准确地抽取家禽疾病诊疗复杂知识文本中的实体关系三元组。     

A rapid extensigraph protocol for measuring dough viscoelasticity and mixing requirement

The extensigraph is particularly useful in characterizing dough viscoelastic properties; however, testing throughput for standard method is low due to the prerequisite for farinograph water absorption, long dough resting and milling to prepare large amounts of flour. Therefore, a rapid extensigraph method was developed that reduced sample size (165 g wheat) for milling and more than tripled throughput. Wheat is milled in Quadrumat Junior mill with a modified sieving system. The resulting flour (100 g) was mixed with a pin mixer at constant water absorption to allow the evaluation of wheat genotypes at the absorption level they are expected to perform. Dough was subsequently stretched by an extensigraph after 15 min of floor time and 30 min resting. Strong correlations for extensigram R_max (r > 0.93), extensibility (r > 0.64) and area (r > 0.88) were found for the proposed method compared to the standard method. Mixing parameters (time and energy) obtained during dough preparation provided further information about dough strength and mixing requirement. By significantly reducing sample size requirement and increasing testing throughput, this rapid extensigraph method can be widely adopted in milling and baking industry and meets the need for a fast evaluation of dough strength in breeding trials.