[Exploration and practice of artificial intelligence assisted primary vision health management].

It has attracted much attention worldwide that the application of artificial intelligence (AI) in primary screening and clinical diagnosis and treatment of eye diseases. In recent years, this technology has also been widely used in various grass-roots eye disease management, effectively improving the current situation of weak eye disease diagnosis ability and shortage of human resources in primary medical institutions. At present, there is no reference standard or guideline for the management mode, implementation content and management method of vision health management based on this technology, which are in urgent need of standardization. The article described the work mode exploration of AI-assisted grass-roots visual health management in Shanghai and shared practical experience. The aim is to provide reference for other provinces in China to carry out relevant work.

[Analysis of community intervention effects for diabetic eye diseases in Shanghai Xinjing community from 2016 to 2018].

Objective: To analyze the effects of community-based interventions for diabetic eye diseases in Xinjing community, Shanghai from 2016 to 2018. Methods: Based on the project of "Establishment of Service Model for Comprehensive Prevention and Treatment of Diabetic Eye Diseases in Shanghai", the participants were not suffering diabetic retinopathy (DR) in Xinjing community in 2016 before interventions and received community-based interventions for diabetic eye diseases. The incidence of DR, visual acuity and awareness of DR were used as evaluation indicators to analyze the effects of interventions for diabetic eye diseases in the community. Results: A total of 537 patients were included in this study, the incidence of DR among diabetic patients in Xinjing community was 7.6% after interventions. The duration of diabetes (OR= 1.065) and HbA1c (OR= 1.090) were the risk factors of DR. Before and after the interventions, the patients with monocular low vision and binocular low vision were 27 cases (5.0%), 8 cases (1.5%), 19 cases (3.5%) and 7 cases (1.3%) respectively. After interventions, the awareness on the prevention and treatment of DR increased significantly, and the proportion of regular visits to ophthalmology examination, diet control and physical exercise also increased significantly. Conclusion: Community-based interventions for diabetic eye diseases are helpful to improve the awareness of DR prevention and control, reduce the incidence of DR.

[Two cases of Wolfram syndrome].

Two adolescents with T1DM participated in the Shanghai Children and Adolescent DM Eye study (SCADE) 2017-2018. The previous T1DM history of the 2 children were 12 years and 4 years respectively. The history of optic atrophy were 8 years and 4 years respectively. The ophthalmic examination showed that there were slight posterior capsule cataract and atrophy of optic atrophy in both patients. Optical Coherance Tomographic examination showed that the choroidal thickness of the inferior inner ring of the optic nerve and RNFL layer thickness decreased markedly, perfusion density and vascular density of the optic disc area and macular decreased markedly, electrophysiological examination showed that amplitude decreased markedly, and visual field examination showed binocular extinction or tubular vision field. Now the 2 adolescents were in closely followed-up. We suggest that children with T1DM whose best corrected visual acuity decreased recently to 0.6 or lower, in addition to routine ophthalmic examination, choroidal thickness, retinal nerve fiber layer thickness, retinal blood flow, ERG and visual field examination are recommended to distinguish wolfram syndrome from T1DM. Genetic sequencing examination of WFS-1 is also recommended if possible.

[Challenges of artificial intelligence used for eye disease screening in recent China communities].

Due to factors such as medical resources, public awareness, funding for general screening, or optimized screening models, community-based screening is far from meeting the demand. Artificial intelligence (AI) can replace some of the medical work and combine it with the "Internet+" model to transfer medical resources to improve accessibility and availability. However, the application of AI technology to community-based screening still faces many challenges, such as most AI products cannot be directly applied to community-based screening, the inability to integrate multimodal information such as medical history, symptoms, and images, and the lack of relevant regulations and health policies for productization and implementation. Therefore, we suggest that the relevant departments take actions: (1) to build standardized big data sets, unlock data barriers, and accelerate the development and application of AI technology; (2) to train "AI+" medical staffs as soon as possible; (3) to establish relevant laws and regulations; (4) to establish relevant R&D plans and quality standards and regulatory frameworks for AI products; (5) to encourage more investment in medical AI infrastructure in the central and western regions and remote and poor areas.

[Two-year follow-up study in type 2 diabetic patients with mild visual impairment].

Objective: To determine the 2-year visual prognosis in Chinese type 2 diabetic patients with mild visual impairment and identify the predictors factors. Method: This was a 2-year population-based cohort study. The study population consisted of 650 type 2 diabetic patients with bilateral mild visual impairment in 2014 who were followed up in 2016. The demographic information, systemic and ophthalmological examination results for each participant was collected. Mild visual impairment was defined as best-corrected visual acuity (BCVA)<20/25 to ≥20/63, moderate and severe visual impairment was defined as BCVA<20/63 to ≥20/400, blindness was defined as BCVA<20/400 following the International Council of Ophthalmology (ICO) 2002 definition. The two-year visual prognosis was divided into three groups: visual impairment regression, progression, and stable. Using chi-square test or independent t-test, the predictor factors of visual prognosis and the leading causes of visual impairment were studied. Results: 605 patients completed the follow-up in 2016. Among them, 477 were still bilateral mild visual impairment, accounting for 78.8% (477/605). The level of unilateral or bilateral visual impairment regressed in 8.1% (49/605), while progressed in 13.1% (79/605). Young age (t=2.7, P<0.05), short duration of diabetes (t=2.5, P<0.05), low blood glycemic hemoglobin (t=2.5, P<0.05) and total cholesterol (t=2.8, P<0.05) were associated with regression of visual impairment. Low levels of education (t=5.2, P<0.05), high blood glycemic hemoglobin (t=2.4, P<0.05) and total cholesterol (t=2.4, P<0.05)were associated with progression of visual impairment. Cataracts and diabetic retinopathy were the first and second cause of mild, moderate severe visual impairment or blindness. Conclusions: The percentage of 2-year visual acuity decline is relatively high in type 2 diabetic patients with mild visual impairment. We should strengthen blood glucose and cholesterol control to reduce the progress of visual impairment.(Chin J Ophthalmol, 2021, 57: 766-771).

[Forcasting the prevalence of myopia among students aged 6-18 years in China from 2021 to 2030].

Objective: To predict the prevalence of myopia among Chinese students aged 6-18 years under different intervention scenarios from 2021 to 2030. Methods: The multi-state Markov model was developed based on the transition process of study stages and myopia statuses. The development of myopia was simplified into two statuses: non-myopia and myopia. Students aged 6-18 years were also divided according to their study stages including senior kindergarten, primary school (from Grade 1 to 6), junior school (from Grade 1 to 3) and high school (from Grade 1 to 3). The parameters were extracted from the National Myopia Investigation in 2018 and published articles of cohort studies. The transition probability was applied to simulate the intervention scenarios, and sensitivity analysis was carried out. Results: The cumulative incidence of myopia among Chinese school-aged children and adolescents would increase consistently. It would be 91.3% (min to max: 83.7% to 96.7%) upon graduation from high school. Without any intervention, the myopia prevalence would increase to 61.8% (min to max: 55.4% to 69.5%) by 2030 among Chinese school-aged children and adolescents. And the myopia prevalence among students in primary schools, junior schools and high schools would be 45.6% (min to max: 40.2% to 54.3%), 81.3% (min to max: 72.6% to 91.0%) and 90.5% (min to max: 82.4% to 96.7%), respectively, all higher than the national target. If the interventions could achieve 70% of the desired effect, the myopia prevalence would be lower than the national target at each stage. Conclusions: Without effective interventions, the prevalence of myopia among students aged 6-18 years may keep increasing in the next ten years. If the interventions achieve the desired effect, the national target for myopia prevention and control could be reached. It is urgent to identify more effective interventions and call on the whole society to participate in the myopia prevention action to achieve the national goal by 2030. (Chin J Ophthalmol, 2021, 57: 261-267).

[Paying attention to adopting modern epidemiological methods for the prevention and treatment of infectious eye diseases].

Carrying out research on infectious epidemics is an important work contents stipulated in the Law on the Prevention and Control of Infectious Diseases in China. Nowadays, Based on the recent Chinese prevention and treatment system of infectious eye diseases, it is difficult to meet the needs of data collection, etiology determination, and spatiotemporal analysis. The present article explores how to use modern social epidemiology methods to clarify the role of individual factors, social factors, and structural factors in the three links of eye disease epidemic, how to use spatial epidemiology methods to describe, quantify, and explain the spatial distribution characteristics of eye diseases in the population, and how to use infodemiology and digital epidemiology methods to promptly and efficiently mine and process massive amounts of electronic data of eye diseases. These may be used to improve the scientific nature of Chinese eye disease epidemiology research, and strengthen the China's infectious eye disease prevention and control system.(Chin J Ophthalmol, 2020, 56:567-570).

[Analysis of cataract surgery status in public hospitals of Shanghai from 2013 to 2015].

Objective: To analyze the reasons that restrict the growth of cataract surgery service capacity in public hospitals in Shanghai in recent years. Methods: The status of surgeries performed in public hospitals are analyzed based on the data related to cataract surgery collected from the database of Shanghai Eye Disease Treatment. Meanwhile, the surgeries performed by ophthalmologists working in the public hospitals are studied based on the National ophthalmology service capacity questionnaire. Results: The cataract surgery volume performed in public hospitals of Shanghai increased from 45 480 in 2013 to 51 941 in 2015. In 2014, the year on year growth rate of cataract surgery volume in tertiary hospitals was 8.54%, while in 2015, it was -0.21% on an annual basis. More than 70% cataract surgeries were performed in tertiary public hospitals. For those performed in tertiary public hospitals, 80% were performed in urban area. The actual surgeons in tertiary account for 70% of the actual surgeons in all public hospitals. Among all cataract surgeries performed in secondary hospitals, half were performed in urban areas. The volume of cataract surgery by cataract surgeon and the number of the ophthalmologist were higher than those in secondary hospitals. The average cataract surgery volume of tertiary hospitals in urban areas and the average annual cataract surgery volume of the actual surgeons are much higher than those of the secondary hospitals in the urban areas, but it is contrary in exurban areas. Conclusion: The excessive density of tertiary hospitals in urban area and poor ophthalmology service capacity in secondary hospitals in suburban and exurban areas have restricted the rapid growth rate of cataract surgery and even a decline in Shanghai public hospitals. (Chin J Ophthalmol, 2020, 56: 615-620).

[Development and application of a fundus image quality assessment system based on computer vision technology].

Objective: To develop a fundus image quality assessment system based on computer vision technology and to verify its accuracy by comparing the results of artificial discrimination and using this system. Methods: The process of image evaluation was divided into four modules: fundus image preprocessing, fundus image quality evaluation, fundus image content detection and evaluation result output. The system was designed to automatically evaluate the image quality of each fundus image, identify the optic disc and macula, and judge whether the image was qualified or not according to the image quality discrimination rules. A total of 2 397 fundus images of 787 type 2 diabetes patients were selected as the test data set. The average age of the patients, including 384 males and 403 females, was (69.65±19.09) years old. The images were taken by the staff of community health service centers in Shanghai with a fundus camera. The fundus image quality assessment system was used to conduct quality control and classification of the data set. At the same time, 12 professional fundus picture readers were employed to conduct manual quality control and classification of this data set. The system quality control results and artificial quality discrimination results were compared and analyzed. Results: The fundus image quality assessment system automatically recognized left and right eyes and eye positions on the input fundus images. The quality control interface included four indicator lights, which respectively corresponded to the images with the optic disc or macula as the center of the left or right eye. Evaluation of each fundus image was completed within 1 second, and the results were automatically displayed on the user interface. The 2 397 fundus photos were identified manually as 1 846 qualified photos and 551 unqualified photos. Among the unqualified images, 62 (11.27%) were too dark, 51 (9.27%) were too bright, 59 (10.73%) were not clear in the macular area, 36 (6.54%) showed no macula or optic disc, 125 (22.73%) could not present the fundus structure, 175 (31.82%) were blurred, and 42 (7.64%) were blocked. The results of the system and manual assessment were consistent in 1 788 qualified images (96.86%) and 550 unqualified images (99.82%), with an overall consistency rate of 97.54%. Conclusion: The fundus image quality assessment system can achieve highly consistent results with the professional judgment of ophthalmologists and has the characteristics of objectivity. (Chin J Ophthalmol, 2020, 56:920-927).

[Rasch analysis of the Chinese Version of the Low Vision Quality of Life Questionnaire].

Objective: To investigate the unidimensionality, reliability and validity of the Chinese version of the Low Vision Quality of Life Questionnaire (CLVQOL) on the basis of Rasch analysis. Methods: All participants completed the CLVQOL. The psychometric properties of the CLVQOL were tested using Rasch analysis,including the unidimensionality,person separation index (PSI),person reliability (PR), infit mean square MNSQ and differential item functioning (DIF). Results: One hundred and forty seven residents,including 58 males (39.46%) and 89 females (60.54%),were enrolled. The mean age of the participants was (67±10) years old. The CLVQOL had 4 original scales. The first scale was found to be multidimensional and was subsequently divided into 2 scales. A new 5-scale CLVQOL was then developed, and all 5 scales were shown to be unidimensional. All PSIs exceeded 2,and all PRs exceeded 0.8,indicating the favorable discriminant ability of the questionnaire. Notable DIF was not found for any item across different age or gender subgroups. Conclusions: A Rasch model can be used in improving the rating questionnaires for evaluating vision-related quality of life (VRQoL). All 25 items demonstrated an acceptable fit in Rasch analysis,and a new 5-scale CLVQOL was developed.(Chin J Ophthalmol, 2019, 55: 582-588).

[Analysis of cataract surgery cases from 2013 to 2015 in Shanghai].

Objective: To analyze the changes of cataract surgery volume and related influencing factors in Shanghai during 2013 and 2015. Methods: The information of Cataract patients during the period of 2013 and 2015 was collected from Shanghai Cataract Operations Database. The cataract surgery volume and CSR of Shanghai residents and non-registered ones were compared. The cataract surgery volume and CSR of registered Shanghai residents in urban, suburban and exurban areas were calculated, respectively. In addition, the sources of surgery cost, hospitals where surgeries were performed were analyzed. Results: In Shanghai, the cataract surgery volume grew from 63 915 to 114 688 and the total CSR grow from 2 686 to 4 729 during the years between 2013 and 2015. In this period, the cataract surgery volume of registered Shanghai residents in urban, suburban and exurban areas increased from 27 203, 21 921, 8 847 to 46 415, 40 463, 18 171, respectively. The CSR1 (CSR of registered Shanghai residents) grew rapidly from 4 081 to 7 363, while the CSR2 (CSR of non-registered Shanghai residents) grew slowly from 619 to 949. According to statistics, 93.08% of the registered Shanghai residents paid the medical expenses with their medical insurance, while more than 80% non-registered Shanghai residents afforded it at their own expenses. The number of total CSR in urban, suburban and exurban areas increased from 5 085, 3 600, 3 205 to 3 600, 6 588, 6 513 respectively. In 2015, the number of cataract surgery cases in non-public hospitals exceeded that in public hospital, accounting for 54.71% of the total cataract surgeries. In the same year, 55.44% of the total cases come from suburban and exurban residents. Conclusions: From 2013 to 2015, there had been a dramatic increase in cataract surgery volume in nonpublic hospitals and exurban area in Shanghai, which effectively solved the problem of low CSR in suburban and exurban area, however the 1ow level of CSR of non-registered Shanghai residents was the key factor that significantly reduced the total CSR in Shanghai. (Chin J Ophthalmol, 2018, 54:580-585).

[The problems and challenges in the current eye health management in China].

The core goal of "Thirteenth Five-Year Plan for National Eye Health (2016 to 2020)" is to obtain eye health. At present, there are still several problems hampering Chinese eye health management system development, including the lack of top policy design, system protection and financial investment, weak support from primary health care institutions, other medical disciplines and social forces, and insufficient digital information. To achieve effective eye health management, it is necessary to address the sustainability of eye health management, the integrity of eye health digital information data and the feasibility of eye health payment. It is recommended that eye health-related medical workers take positive actions to strengthen the digital information construction and service capacity construction of eye disease prevention and control network, carry out health promotion extensively, and conduct eye health management mode research and practice.(Chin J Ophthalmol, 2017, 53: 481-483).

[Critical elements of cataract prevention work in China].

Surgical treatment of cataract is the emphasis of prevention of blindness in China. In the past ten years, cataract prevention work has made great strides. However, the cataract surgery rate (CSR) in one million people is still far away from the developed countries, and not enough to eliminate the blindness of cataract in our country. The CSR is related to many factors. In this article, the issue of the CSR increase is commented from four aspects, including improving the quantity and quality of cataract surgery, increasing the allocation of ophthalmic resources to primary hospitals, strengthening the cooperation between the urban and backward areas in ophthalmology, and the economic factors. It is hoped that our ophthalmic practitioners pay more attention to the prevention of blindness, and the quantity and quality of cataract surgery could be increased. (Chin J Ophthalmol, 2017, 53: 801-804).

[A cross-sectional study of cataract in residents with type 2 diabetes living in Xinjing Town, Shanghai].

Objective: To investigate the prevalence, subtypes and risk factors of cataract in type 2 diabetic individuals. Methods: Geographically defined cluster sampling method was used in this population-based, cross-sectional study. There were 7 756 type 2 diabetes individuals in Xinjing Town with a residential population of 88 864. The type 2 diabetic individuals were randomly selected from 20 basic sample units in Xinjing Town of Changning District from April to June 2016. All participants received visual acuity measurement and eye examination. The standard of lens opacity assessment was according to the Lens Opacities Classification SystemⅡ(LOCSⅡ). The prevalence of cataract in diabetic individuals was calculated with LOCS≥2. Multivariate Logistic regression analysis was used to explore the relevant factors of cataract in type 2 diabetic individuals. Results: A total of 1 719 type 2 diabetic individuals were included in the analysis. There were 682 men (39.67%) and 1 037 women (60.33%) in this population. There were 434 cataract individuals and the prevalence of cataract was 25.25%. There were 269 cases of nuclear type (15.7%), 38 cases of cortical type (2.2%), 2 cases of posterior subcapsular type (0.12%) and 42 cases of mixed type (2.4%) in the 1 719 individuals. Multivariate Logistic regression analysis showed that age (P<0.001), duration of diabetes (P<0.001), education (P=0.005), fasting blood glucose (P(6.1-7.7mmol/L)=0.025, P(7.8-24.3mmol/L)=0.022, compared with 3.6-6.0 mmol/L of fasting blood glucose), and ocular axial length (P<0.001) were associated with cataract. Conclusion: Cataract is a common ophthalmic disease in adults with type 2 diabetes in Xinjing Town. Regular screening of diabetes in the high risk population and intensive control of both glucose and blood pressure in diabetic patients are recommended to prevent and delay the development of cataract. (Chin J Ophthalmol, 2017, 53: 489-494).

[Active screening and early intervention of diabetic retinopathy in community service].

In China, the number of diabetic retinopathy (DR) patients is increasing rapidly. Prevention is the most effective method to reduce blindness and visual impairment caused by DR. The ophthalmic professionals who are interested in the DR prevention and treatment should promptly go to the communities, work together with general practitioners and community health personnel, make full use of community residents health archives, and gradually establish community DR screening protocols. Thus the efficiency of screening and early intervention of DR will be improved, and finally the high incidence and high prevalence of blindness related to DR will be controlled. (Chin J Ophthalmol, 2016, 52: 801-804).

[Research advances in diabetic retinopathy telescreening systems].

Diabetic retinopathy has become the main cause of the sight impairment and blindness among the adult population. Early detection of diabetic retinopathy helps to prevent and reduce the damage to eyesight. The development of diabetic retinopathy telescreening systems has been rapid. The operation modes, key technologies, economic benefits and new progression of diabetic retinopathy telescreening systems are reviewed. (Chin J Ophthalmol, 2016, 52: 868-871).

[The study of establishment of the " Chinese provincial Blindness prevention technical guidance group performance evaluation system"].

Objective: To establish an assessment system, including indexes and scoring methods, that can be used for performance evaluation of the provincial blindness prevention technical guidance group properly and effectively . Methods: The indexes and scoring methods were set based on the core content of The " National Plan of Prevention and Treatment of Blindness (2012-2015)" , the specific requirement and target of the World Health Organization (WHO) "For the General Eye Health: Global plan of Action (2014-2019)" , and the current situation of the China's provinces and autonomous regions. These indexes should be of effectiveness, feasibility, comparability, guidance and advancing. Formed by a literature review of candidate indicators, the framework of the system is built by qualitative assessment. With the Delphi method, the system was further revised and improved. Empirical pilot study was then used to prove the feasibility, followed by the final qualitative analysis that establish the " Chinese provincial Blindness prevention technical guidance group performance evaluation system" . Results: Through the literature review and qualitative assessment, a six dimensional system framework was built, including 6 first-level indicators, 16 second-level indicators, and 29 third-level indicators through Delphi method evaluation. With the variation coefficient method, the coeffiences of the first-level index weight were calculated as: Organization and management 0.15, Development and implementation of blindness prevention plans 0.15, Implementation of blindness prevention projects 0.14, Training 0.17, Health education 0.18, and Cooperation and exchanges 0.21. The specific scoring method for this system is confirmed as: data and files check, field interview, and record interview, sampling investigation. Empirical pilot study was conducted in the Jilin, Guizhou and Gansu provinces, and the self-assessment results from local experts were consistent with the scores from the systems. Conclusion: This system established is appropriate at current time, and it can effectively evaluate the performance of the Chinese provincial Blindness prevention technical guidance group. (Chin J Ophthalmol, 2016, 52:814-824).

[A cross-sectional study of moderate or severe visual impairment and blindness in residents with type 2 diabetes living in Xinjing Town, Shanghai].

Objective: To investigate the prevalence, underlying causes and risk factors of moderate or severe visual impairment and blindness in a population with type 2 diabetes in Xinjing Town, Shanghai, China. Methods: A cross-sectional survey among local Han adult residents, who were previously diagnosed as type 2 diabetes, was conducted between October 2014 and January 2015. The survey was preceded by a pilot study; operational methods were refined and quality assurance evaluation was carried out. The best corrected visual acuity was recorded and classified following the modified World Health Organization grading system. Assigned ophthalmic doctors assured the leading causes of every blind or visually impaired eye. Binary logistic regression analysis was used to determine the related factors of blindness and moderate or severe visual impairment. Results: A total of 2 216 type 2 diabetic residents were enrolled, and 166 eyes (3.7%, 166/4 432) were blind. Cataract was the leading cause of blindness (39.8%); macular degeneration (18.0%) and eyeball atrophy (11.4%) were the second and third leading causes of blindness, respectively. Moderate or severe visual impairment was found in 376 eyes (8.5%, 376/4 432), and the most frequent cause was cataract (65.7%), followed by diabetic retinopathy (9.8%) and macular degeneration (9.4% ). Older age, female gender, earlier onset diabetes and a lower spherical equivalent in the better eye were associated with best corrected visual acuity<20/63 in the better eye. Conclusion: The prevalences of moderate or severe visual impairment and blindness in our population with type 2 diabetes were high. (Chin J Ophthalmol, 2016, 52: 825-830).

Expressed sequence tag-PCR markers for identification of alien barley chromosome 2H in wheat.

We developed EST-PCR markers specific to barley chromosome 2H, for the purpose of effectively tracing alien chromosomes or chromosome parts in the wheat genetic background. The target alien chromosome 2H confers high resistance to pre-harvest sprouting, which is a worldwide natural disaster in wheat. A total of 120 primer pairs were selected by combining the wheat group 2 chromosomes of the EST database and the genome sequences of the new model plant Brachypodium distachyon. Seventy-seven of 120 primer pairs were polymorphic and 31 of 120 primer pairs were monomorphic between a set of wheat-barley chromosome 2H disomic addition/substitution lines and their parents by agarose gel electrophoresis and polyacrylamide gel electrophoresis. Thirty of 77 polymorphic primer pairs including primer pair P120 derived from the basi gene were chromosome 2H-specific. These markers are expected to be valuable in screening of wheat-barley chromosome 2H recombination lines and pre-harvest sprouting resistant varieties.

Cloning and characterization of a ß-amyrin synthase gene from the medicinal tree Aralia elata (Araliaceae).

Aralia elata is an important medicinal plant in China; it produces large amounts of oleanane type triterpene saponins. A full-length cDNA encoding ß-amyrin synthase (designated as AeAS) was isolated from young leaves of A. elata by reverse transcription-PCR. The full-length cDNA of AeAS was found to have a 2292-bp open reading frame, encoding a protein with 763 amino acid residues. The deduced amino acid sequence of AeAS showed the highest identity (97%) to Panax ginseng ß-amyrin synthase. When AeAS cDNA was expressed in Escherichia coli, an 87.8-kDa recombinant protein was detected by SDS-PAGE and Western blotting. The sequence was also heterologously expressed in the yeast Pichia pastoris, and production of ß-amyrin was detected by HPLC. Tissue expression pattern analysis by real-time reverse transcription-PCR revealed that AeAS is strongly expressed in leaves and stems, and weakly expressed in roots and flowers.