Visual function and biofeedback training of patients with central vision loss:a review

Older individuals with macular diseases,such as age-related macular degeneration,experience central vision loss(CVL)due to degeneration of their photoreceptors and retinal cells.Patients with CVL may experience various vision impairments,including of visual acuity,fixation stability,contrast sensitivity,and stereoacuity.After CVL,most patients develop a preferred retinal locus outside the affected macular region,which serves as a new visual reference.In this review,we provide an overview of the visual function and impairment in individuals with CVL.In addition,the important role of biofeedback training on the visual function and activity of individuals with CVL is also reviewed.Accordingly,the location and development of the preferred retinal loci are discussed.Finally,this review discusses how to conduct biofeedback training to treat individuals with CVL.

Visual impairment and depression: Age-specific prevalence, associations with vision loss, and relation to life satisfaction

BACKGROUND To our knowledge,no study has obtained specific estimates of depression for young and middle-aged adults with visual impairment(VI).As estimates of depression varies across age groups in the general population,it is of interest to examine whether the same applies to adults with low vision or blindness.AIM To estimate depression prevalence and its association with VI-related characteristics and life satisfaction in adults with VI.METHODS A telephone-based cross-sectional survey was conducted between January and May 2017 in an age-stratified sample of adults who were members of the Norwegian Association of the Blind and Partially Sighted.Participants were asked questions about their sociodemographic characteristics,VI characteristics,and life satisfaction.Depression was measured with the Patient Health Questionnaire.The diagnostic scoring algorithm was used to calculate the point prevalence of depression(i.e.,major depression and other depressive disorders)across categories of gender and age(years:18-35,36-50,51-65,≥66).The associations were estimated using regression models.RESULTS Overall,736 adults participated in the study(response rate:61%).The prevalence estimates of depression varied across different age groups,ranging from 11.1%-22.8%in women and 9.4%-16.5%in men,with the highest rates for the two youngest age groups.Results from the multivariable models including sociodemographic and VI-related variables showed that losing vision late in life[Prevalence ratio(PR),1.76,95%CI:1.11,2.79]and having other impairments(PR:1.88,95%CI:1.32,2.67)were associated with higher rates of depression,whereas older age was associated with lower rates(PR:0.83,95%CI:0.74,0.93).Additionally,participants who were depressed had lower life satisfaction than those who were not depressed(adjustedβ:-2.36,95%CI:-2.75,-1.98).CONCLUSION Our findings suggest that depression in adults with VI,and especially among young and middle-aged adults,warrants greater attention by user organisations,clinicians,and healthcare authorities.

AB031.Exploring the facilitators and barriers encountered by adults and seniors with acquired vision loss who pursue braille training

Background:As the prevalence of age-related vision loss continues to increase,vision rehabilitation practitioners will encounter a growing number of older adults who experience reading-related difficulties.Braille may be considered as an alternative to sight enhancement options for clients with progressive or significant visual pathologies.However,little is known about the impact of aging on braille reading performance.The goal of this research was to explore the experiences of older adults with acquired vision loss who have learned braille later in life,and to better understand the facilitators and barriers encountered throughout this process.Methods:Qualitative interviews were conducted by telephone with 14 participants from across Canada(age 40-72,Mdn=55.5,7 female).All participants learned braille in adulthood(starting age 33-60,Mdn=46.5).Recorded interviews were transcribed verbatim and later analyzed by two independent researchers based on the phenomenological method of analysis.Results:Personal,social and environmental factors were shown to influence the braille-learning process.In particular,prior learning experiences(both positive and negative)were shown to impact both the decision to learn braille and participant training experiences.Participants highlighted the importance of perceived support from family and friends,and reactions towards braille from the general public posed a barrier to those still adjusting to vision loss.Participants also highlighted the value of knowing others who learned braille later in life.Conversely,a reoccurring environmental barrier that emerged was the perceived response from rehabilitation practitioners who believed that clients should not or could not learn braille due to their age,as well as the lack of available resources to facilitate training.Conclusions:Results highlight the need to explore the influence of stereotypes associated with aging,blindness and braille,and the degree to which this may impact opportunities for clients who may benefit from braille training.Though previous learning experiences appear to influence successful rehabilitation outcomes,these factors are not routinely considered during the braille assessment process.Our findings will therefore contribute to future research and the development of assessments to better meet the needs of older adults who pursue braille training.

Tone Task Proves Blind Hear Better--Early vision loss leads to keener hearing

盲人音乐家的才华洋溢是纯属巧合,还是由于潜能的凝聚呢?加拿大的科学家们发现:年幼失明的盲者对于乐音变化的察觉能力确实远远强于后天失明者及普通人。

一种轻量型果园环境果实检测方法

果园环境下柑橘的快速准确检测是自主采摘机器人作业的关键.针对现有的模型过于冗余、检测速度与精度不平衡等问题,提出一种轻量型果园环境果实检测方法.在YOLOv4算法的基础上引入焦点损失函数(Focal Loss)来提高模型在二分类检测任务中的负样本挖掘能力,并针对模型参数冗余等问题提出一种优化的模型剪枝方法.试验结果表明:提出的方法在果园环境中柑橘果实数据集检测得到的平均精度均值(mean average precision,M_(AP))达到94.22%,相较于YOLOv4模型提高了1.18%,模型参数减小了95.22%,模型尺寸为原来的4.84%,检测速度为原来的4.03倍.

武汉市中小学生身心健康现况及其影响因素分析

目的 探讨武汉市中小学生身心健康现况及其影响因素。方法 于2020年9-11月对武汉市4~12年级63 159名学生进行问卷调查,并获取体质健康检查数据。采用Logistic回归模型分析身心健康的主要影响因素。结果 总人群中自报视力下降率为60.0%,体检视力不良率为53.0%。随学段增加,重度视力不良率(视力≤4.5)明显上升,小学为17.8%、初高中则分别激增至45.1%和51.4%。超重/肥胖率达28.0%,小学最高(30.9%)。情绪不良率为23.5%,随学段的增加而上升,初高中分别为31.4%和35.8%(均Ptrend<0.001)。Logistic回归分析显示,观看电子屏幕≥1 h/d、桌椅高度不合适、健康教育改善认知不好与视力下降呈正向关联;每周喝含糖饮料1~4次、睡眠<7 h/d与超重/肥胖率增高有关;父母关系变差、受欺凌、睡眠<7 h/d与情绪不良率上升有关(均P<0.05)。结论 武汉市中小学生视力下降、超重/肥胖和情绪不良问题突出,需家庭、学校和政府联合干预。

AB030.Usability of ArtontheBrain:A visual-arts based mobile health solution to promote quality of life in older adults with low vision due to age-related macular degeneration

Background:Taking part in productive and enjoyable recreational activities has been shown improve quality of life for people of all ages and capabilities.However,vision loss can have a significant impact on participation in important leisure activities.This is especially the case for the elderly,whose lowered mobility is further impacted by vision loss.Technology can offer solutions to bridging some barriers caused by these deficits by bringing leisure activities to the user in the form of mobile applications.As such,the purpose of the present study was to evaluate the accessibility and usability of the ArtOnTheBrain application,a visual art based mobile health solution to promote brain health and well-being,by older adults with low vision due to age-related macular degeneration(AMD).Methods:There were a total of 16 participants(age range,65-93 years,M=79,10 males).All had a diagnosis of AMD with visual acuities in the better eye between 20/60 and 20/200.Additionally,all participants had accessed rehabilitation services and most had experience using a computer at home.Using an Apple iPad Air(2013),they were asked to interact with the ArtOnTheBrain website’s Learn(e.g.,Listen to artwork description)and Play(e.g.,complete a word-search game)features with either the Safari or Google Chrome Internet apps.Using the Concurrent Think Aloud method,participants were asked to continuously comment on their activities and experiences with the app and verbalize their internal monologue while being audio and video recorded.These recordings were later transcribed verbatim and analysed using qualitative description and thematic analysis.Results:Participants’behaviours and verbal feedback were divided based on whether they presented as barriers or facilitators.Beginning with barriers,these were mostly related to the accessibility of the visual aspects of the applications interface(i.e.,contrast and font size),whereby participants requested additional control over the magnification options,both for text as well as images of the artwork.The main facilitator was the audio option built into the Learn tab,which allowed participants to listen to the artwork descriptions.Facilitators also included the aesthetics of the app,the perceived boost in confidence in interacting with technology,and the educational and leisure benefits.Conclusions:Older adults with low vision are faced with a decreased ability to engage in leisure activities.The development of technologies aimed at increasing the accessibility of leisure activities for these individuals is an important step in increasing their quality of life.As such,despite some of the accessibility challenges,the majority of participants viewed ArtontheBrain positively.Identifying the barriers and facilitators to its use is an important step in the development of this application in order to optimize its accessibility for older adults with low vision.Future work to be done with this application will be to evaluate the cognitive impact of ArtontheBrain on low vision users.

OMC框架下的行人多目标跟踪算法研究

多目标跟踪是计算机视觉领域被广泛研究的重要方向,但是在实际应用中,目标的快速移动、光照变化、遮挡等问题会导致跟踪性能变差,因此以多目标跟踪模型OMC为基础框架展开研究,以实现跟踪性能的进一步提升。针对多目标跟踪过程中存在的目标特征质量层次不齐的问题,对特征提取器进行优化,在主干网络集成了GAM注意力机制并在Neck网络部分更换了上采样方式;针对现有方法中存在的检测任务和重识别任务之间的“竞争问题”,构建了递归交叉相关网络,使得模型可以学习不同任务的特性和共性。此处针对两个子任务分别进行了优化,一是设计了新的通道注意力HS-CAM优化了重识别网络;二是更换了检测部分的边界回归损失函数,采用EIoU损失函数。实验表明,在MOT16数据集上MOTA指标可达73.5%,IDF1可达70.4%,MLgt为11.7%,相比较OMC算法减少了1.5个百分点。

视网膜血管病伴白质脑病和多系统损害的临床特点(附1例报告)

目的 探讨视网膜血管病伴白质脑病和多系统损害(RVCL-S)的临床特征,为临床诊治RVCLS提供思路。方法 总结2022年2月我科收治的1例RVCL-S的患者临床、影像及遗传学资料,同时系统性回顾国内外文献,分析其临床特征、治疗及预后等情况。结果 该患者为51岁女性,首发神经系统症状为言语不利及左侧肢体偏瘫。病程中出现认知功能下降、抑郁及共济失调。头颅MRI示双侧脑白质病变及脑萎缩。头颅CT示颅内多发点状钙化灶。家中其他成员无相关临床症状。全基因组测序发现患者3′端修复核酸外切酶1(TREX1)基因存在1个杂合移码变异,即c. 703dup:(p. Val235Glyfs*6)。Sanger测序证实其女儿存在相同杂合变异。根据美国医学遗传学与基因组学学会指南评级为可能致病性变异。该突变位点为既往国外已报道的致病突变。结论 对视力减退合并局灶性神经功能障碍、认知下降的中青年患者,头颅MRI示双侧脑白质病变,需考虑TREX1基因突变所致RVCL-S的可能。

Retinoprotective compounds,current efficacy,and future prospective

Retinal dysfunction is the most common cause of vision loss in several retinal disorders.It has been estimated a great increase in these pathologies that are becoming more globally widespread and numerous over time,also supported by the life expectancy increment.Among different types of retinopathies,we can account some that share causes,symptoms,and treatment including diabetic retinopathy,age-related macular degeneration,glaucoma,and retinitis pigmentosa.Molecular changes,environmental factors,and genetic predisposition might be some of the main causes that drive retinal tissue to chronic inflammation and neurodegeneration in these retinopathies.The treatments available on the market contain compounds that efficiently ameliorate some of the important clinical features of these pathologies like stabilization of the intraocular pressure,reduction of eye inflammation,control of eye oxidative stress which are considered the major molecular mechanisms related to retinal dysfunction.Indeed,the most commonly used drugs are antiinflammatories,such as corticosteroids,antioxidant,hypotonic molecules and natural neuroprotective compounds.Unfortunately,these drugs,which are fundamental to treating disease symptoms,are not capable to cure the pathologies and so they are not life-changing for patients.This review provided an overview of current treatments on the market,but more interestingly,wants to be a quick window on the new treatments that are now in clinical trials.Additionally,it has been here highlighted that the recent technical enhancement of the investigation methods to identify the various retinopathies causes might be used as a sort of“precise medicine”approach to tailor the identification of molecular pathways involved and potentially study a dedicated treatment for each patient.This approach includes the use of cutting-edge technologies like gene therapy and metabolomics.

Mitochondrial phylogeography and molecular evolution of the rhodopsin visual pigment in troglobitic populations of Astyanax mexicanus(De Filippi,1853)

Cave-adapted animals provide a unique opportunity to study the evolutionary mechanisms underlying phenotypic,metabolic,behavioral,and genetic evolution in response to cave environments.The Mexican tetra(Astyanax mexicanus)is considered a unique model system as it shows both surface and cave-dwelling morphs.To date,at least 33 different cave populations have been identified,with phylogenetic studies suggesting an origin from at least two independent surface lineages,thereby providing a unique opportunity to study parallel evolution.In the present study,we carried out the most exhaustive phylogeographic study of A.mexicanus to date,including cave and surface localities,using two mitochondrial markers(cytochrome b(cyt b)and cytochrome c oxidase subunit I(COI))and nuclear rhodopsin visual pigment(rho).Additionally,we inferred the molecular evolution of rho within the two contrasting environments(cave and surface)and across three geographic regions(Sierra de El Abra,Sierra de Guatemala,and Micos).In total,267 individuals were sequenced for the two mitochondrial fragments and 268 individuals were sequenced for the rho visual pigment from 22 cave and 46 surface populations.Phylogeographic results based on the mitochondrial data supported the two-lineage hypothesis,except for the Pachón and Chica caves,whose introgression has been largely documented.The Sierra de El Abra region depicted the largest genetic diversity,followed by the Sierra de Guatemala region.Regarding the phylogeographic patterns of rho,we recovered exclusive haplogroups for the Sierra de El Abra(Haplogroup I)and Sierra de Guatemala regions(Haplogroup IV).Moreover,a 544 bp deletion in the rho gene was observed in the Escondido cave population from Sierra de Guatemala,reducing the protein from seven to three intramembrane domains.This change may produce a loss-of-function(LOF)but requires further investigation.Regarding nonsynonymous(dN)and synonymous(dS)substitution rates(omega valuesω),our results revealed the prevailing influence of purifying selection upon the rho pigment for both cave and surface populations(ω<1),but relaxation at the El Abra region.Notably,in contrast to the other two regions,we observed an increase in the number of dN mutations for Sierra de El Abra.However,given that a LOF was exclusively identified in the Sierra de Guatemala region,we cannot dismiss the possibility of a pleiotropic effect on the Rho protein.

A Semantic Adversarial Network for Detection and Classification of Myopic Maculopathy

The diagnosis of eye disease through deep learning (DL) technologyis the latest trend in the field of artificial intelligence (AI). Especially indiagnosing pathologic myopia (PM) lesions, the implementation of DL is adifficult task because of the classification complexity and definition system ofPM. However, it is possible to design an AI-based technique that can identifyPM automatically and help doctors make relevant decisions. To achieve thisobjective, it is important to have adequate resources such as a high-qualityPM image dataset and an expert team. The primary aim of this research isto design and train the DLs to automatically identify and classify PM intodifferent classes. In this article, we have developed a new class of DL models(SAN-FSL) for the segmentation and detection of PM through semanticadversarial networks (SAN) and few-short learning (FSL) methods, respectively.Compared to DL methods, the conventional segmentation methodsuse supervised learning models, so they (a) require a lot of data for trainingand (b) fixed weights are used after the completion of the training process.To solve such problems, the FSL technique was employed for model trainingwith few samples. The ability of FSL learning in UNet architectures is beingexplored, and to fine-tune the weights, a few new samples are being providedto the UNet. The outcomes show improvement in the detection area andclassification of PM stages. Betterment in the result is observed by sensitivity(SE) of 95%, specificity (SP) of 96%, and area under the receiver operatingcurve (AUC) of 98%, and the higher F1-score is achieved using 10-fold crossvalidation.Furthermore, the obtained results confirmed the superiority of theSAN-FSL method.

基于类相似特征扩充与中心三元组损失的哈希图像检索

现有的深度哈希图像检索方法主要采用卷积神经网络,提取的深度特征的相似性表征能力不足.此外,三元组深度哈希主要从小批量数据中构建局部三元组样本,样本数量较少,数据分布缺失全局性,使网络训练不够充分且收敛困难.针对上述问题,文中提出基于类相似特征扩充与中心三元组损失的哈希图像检索模型(Hash Image Retrieval Based on Category Similarity Feature Expansion and Center Triplet Loss,HRFT-Net).设计基于Vision Transformer的哈希特征提取模块(Hash Feature Extraction Module Based on Vision Transformer,HViT),利用Vision Transformer提取表征能力更强的全局特征信息.为了扩充小批量训练样本的数据量,提出基于类约束的相似特征扩充模块(Similar Feature Expansion Based on Category Constraint,SFEC),利用同类样本间的相似性生成新特征,丰富三元组训练样本.为了增强三元组损失的全局性,提出基于Hadamard的中心三元组损失函数(Central Triplet Loss Function Based on Hadamard,CTLH),利用Hadamard为每个类建立全局哈希中心约束,通过增添局部约束与全局中心约束的中心三元组加速网络的学习和收敛,提高图像检索的精度.在CIFAR10、NUS-WIDE数据集上的实验表明,HRFT-Net在不同长度比特位哈希码检索上的平均精度均值较优,由此验证HRFT-Net的有效性.

Detection of Diabetic Retinopathy from Retinal Images Using DenseNet Models

A prevalent diabetic complication is Diabetic Retinopathy(DR),which can damage the retina’s veins,leading to a severe loss of vision.If treated in the early stage,it can help to prevent vision loss.But since its diagnosis takes time and there is a shortage of ophthalmologists,patients suffer vision loss even before diagnosis.Hence,early detection of DR is the necessity of the time.The primary purpose of the work is to apply the data fusion/feature fusion technique,which combines more than one relevant feature to predict diabetic retinopathy at an early stage with greater accuracy.Mechanized procedures for diabetic retinopathy analysis are fundamental in taking care of these issues.While profound learning for parallel characterization has accomplished high approval exactness’s,multi-stage order results are less noteworthy,especially during beginning phase sickness.Densely Connected Convolutional Networks are suggested to detect of Diabetic Retinopathy on retinal images.The presented model is trained on a Diabetic Retinopathy Dataset having 3,662 images given by APTOS.Experimental results suggest that the training accuracy of 93.51%0.98 precision,0.98 recall and 0.98 F1-score has been achieved through the best one out of the three models in the proposed work.The same model is tested on 550 images of the Kaggle 2015 dataset where the proposed model was able to detect No DR images with 96%accuracy,Mild DR images with 90%accuracy,Moderate DR images with 89%accuracy,Severe DR images with 87%accuracy and Proliferative DR images with 93%accuracy.

基于YOLOv5s改进的口罩佩戴检测算法

由于新型冠状病毒肺炎的爆发,口罩成为人们日常生活中必需品。为了识别与检测人们是否佩戴口罩,提出了一种基于改进的YOLOv5s口罩佩戴检测算法。通过在YOLOv5s主干网络引入改进的自适应的协调注意力机制模块(Coordinate attention-activate or not,CA-A)提升网络的特征提取能力,解决了错误检测和漏检的问题。以新的损失函数AD-CIoU代替CIoU损失函数,作为回归损失函数,提升了边界框的定位精确度。实验表明,与原始模型算法相比,所提出的模型算法平均精度mAP值达到96.1%,提升了1.7%,具有较好的检测精度,可以满足目标检测应用需求。

基于CIoU改进边界框损失函数的目标检测方法

损失函数对于目标检测任务的检测精度和模型收敛速度具有重要作用,而损失函数中的边界框损失函数是影响检测结果和模型收敛速度的重要因素。针对传统模型定位精度低和训练时模型收敛慢的问题,本文在CIoU边界框损失函数的基础上提出一种改进的边界框损失函数,解决了CIoU损失函数求导过程中由边界框宽高比带来的梯度爆炸问题和模型提前退化的问题,并且引入重叠区域与目标框的宽高关系和中心点之间的归一化距离作为附加的惩罚项,提高了模型的检测精度和收敛速度,这种损失函数称为BCIoU(Better CIoU)。在PASACL VOC 2007数据集上的实验结果表明,改进的BCIoU边界框损失函数在YOLOv3网络下相对于IoU损失的mAP50指标相对提升了2.09%,AP指标相对提升了6.88%;相对于CIoU损失的mAP50指标相对提升了1.64%,AP指标相对提升了5.35%。模型的收敛速度也有一定程度的提升。本文提出的BCIoU损失函数提高了模型的检测精度和模型收敛速度,并且可以很方便地纳入到当前目标检测算法中。

急性视网膜动脉缺血的防治研究进展

急性短暂性或永久性视网膜动脉缺血是需要及时诊断和治疗的眼部和全身急症。单眼一过性黑矇属于短暂性视网膜动脉缺血,并不会遗留永久性视功能损害。视网膜中央动脉阻塞和视网膜分支动脉阻塞在大部分患者均会遗留永久性视功能损害。目前的治疗方式包括降眼压、扩血管、高压氧、经静脉或动脉溶栓治疗等,但仍缺乏标准治疗方案。对高危人群应早期一级预防。急性视网膜动脉缺血患者发生继发卒中和不良心血管事件的风险高,应及时发现相关危险因素,积极治疗原发病,采取适当的二级预防措施改善预后。本文对急性视网膜动脉缺血事件的治疗和预防研究进展进行综述,以便为该类疾病的防治和研究提供参考。

功能性视力下降临床特点及治疗分析

目的探讨功能性视力下降临床特点及各类型患眼的视功能训练效果。方法回顾门诊接诊的功能性视力下降患儿临床资料,按照是否接受训练分为接受双眼视功能训练(包括调节及集合训练)的观察组,和仅予以屈光矫正的对照组,随访两组视力及双眼视功能。结果4周后观察组36眼矫正视力均达到1.0,对照组31眼中有16眼(51.6%)达到1.0,差异具有统计学意义(P<0.001)。两组患者治疗后调节功能各项指标均较治疗前改善,且观察组较对照组更为明显(P<0.05)。观察组中单纯调节异常者视力提升快于合并集合不足或集合过度者。停止训练后1个月,合并集合过度者4眼视力均有下降,其余32眼可维持1.0。结论功能性视力下降以单纯调节异常所致最常见;视功能训练可有效提高患眼视力,单纯调节异常治疗效果最佳;合并集合过度者需更长的训练时间,且停止训练后容易复发。

融合交互性特征信息的余弦度量行人重识别

针对行人重识别中难以表达特征间相关性信息,以及欧氏距离度量的损失函数忽略了特征向量角度影响的问题,提出一种融合交互性特征的余弦距离度量行人重识别网络。通过聚合分组注意力模块提取的特征,实现不同子特征跨通道的整合交互。训练阶段采用余弦度量的三元组损失,结合特征空间的批量规一化操作,消除模长波动的影响,从角度维度判别行人差异。采用广义平均池化,保留更完整的特征信息。在Market-1501和DukeMTMC-reID两个数据集的实验结果表明,网络能有效提升行人重识别精度。

支持抗光照目标检测的改进YOLO算法

针对现有的深度学习目标检测算法中存在的复杂光照场景下检测精度不高、实时性差等问题,提出了一种基于YOLO算法的抗光照目标检测网络模型YOLO-RLG。首先,将输入模型的RGB数据转换为HSV数据,从HSV数据分离出抗光照能力强的S通道,并与RGB数据合并生成RGBS数据,使输入数据具备抗光照能力;其次,将YOLOV4的主干网络替换成Ghostnet网络,并对其在普通卷积与廉价卷积的模型分配比例上进行调整,在保证检测精度的同时提高检测速度;最后,用EIoU替换CIoU改进模型的损失函数,提高了目标检测精度和算法鲁棒性。基于KITTI与VOC数据集的实验结果表明,与原网络模型比较,FPS提高了22.54与17.84 f/s,模型降低了210.3 M,精确度(AP)提升了0.83%与1.31%,且算法的抗光照能力得到显著增强。