Search

Background In the development of artificial intelligence in ophthalmology, the ophthalmic AI-related recognition issues are prominent, but there is a lack of research into people’s familiarity with and their attitudes toward ophthalmic AI. This survey aims to assess medical workers’ and other professional technicians’ familiarity with, attitudes toward, and concerns about AI in ophthalmology. Methods This is a cross-sectional study design study. An electronic questionnaire was designed through the app Questionnaire Star, and was sent to respondents through WeChat, China’s version of Facebook or WhatsApp. The participation was voluntary and anonymous. The questionnaire consisted of four parts, namely the respondents’ background, their basic understanding of AI, their attitudes toward AI, and their concerns about AI. A total of 562 respondents were counted, with 562 valid questionnaires returned. The results of the questionnaires are displayed in an Excel 2003 form. Results There were 291 medical workers and 271 other professional technicians completed the questionnaire. About 1/3 of the respondents understood AI and ophthalmic AI. The percentages of people who understood ophthalmic AI among medical workers and other professional technicians were about 42.6 % and 15.6 %, respectively. About 66.0 % of the respondents thought that AI in ophthalmology would partly replace doctors, about 59.07 % having a relatively high acceptance level of ophthalmic AI. Meanwhile, among those with AI in ophthalmology application experiences (30.6 %), above 70 % of respondents held a full acceptance attitude toward AI in ophthalmology. The respondents expressed medical ethics concerns about AI in ophthalmology. And among the respondents who understood AI in ophthalmology, almost all the people said that there was a need to increase the study of medical ethics issues in the ophthalmic AI field. Conclusions The survey results revealed that the medical workers had a higher understanding level of AI in ophthalmology than other professional technicians, making it necessary to popularize ophthalmic AI education among other professional technicians. Most of the respondents did not have any experience in ophthalmic AI but generally had a relatively high acceptance level of AI in ophthalmology, and there was a need to strengthen research into medical ethics issues.

In a large number of bidding supplier groups, it is difficult to accurately find suppliers with unreasonable bidding behavior. In order to solve the problem of precise positioning of massive abnormal bidding behavior groups of diverse and widely distributed suppliers, the authors design a detector framework of abnormal bidding behavior based on supplier portrait. This paper mainly focuses on three abnormal bidding behaviors which harmful to the tenderers—“affiliated operation,” “subcontracting behavior,” and “colluding behavior.” Based on the bidding behavior records of suppliers, this paper establishes supplier portraits in four dimensions. In order to solve the problem that the detection algorithm under the unlabeled data is difficult to verify, this research establishes a new evaluation framework based on the bid base price formula and benefit map database of the supplier. The experiment verifies that the framework can effectively detect most suppliers with abnormal bidding behavior and can significantly change the benchmark price after eliminating abnormal suppliers.

A two-category model and a segmentation model of pterygium were proposed to assist ophthalmologists in establishing the diagnosis of ophthalmic diseases. A total of 367 normal anterior segment images and 367 pterygium anterior segment images were collected at the Affiliated Eye Hospital of Nanjing Medical University. AlexNet, VGG16, ResNet18, and ResNet50 models were used to train the two-category pterygium models. A total of 150 normal and 150 pterygium anterior segment images were used to test the models, and the results were compared. The main evaluation indicators, including sensitivity, specificity, area under the curve, kappa value, and receiver operator characteristic curves of the four models, were compared. Simultaneously, 367 pterygium anterior segment images were used to train two improved pterygium segmentation models based on PSPNet. A total of 150 pterygium images were used to test the models, and the results were compared with those of the other four segmentation models. The main evaluation indicators included mean intersection over union (MIOU), IOU, mean average precision (MPA), and PA. Among the two-category models of pterygium, the best diagnostic result was obtained using the VGG16 model. The diagnostic accuracy, kappa value, diagnostic sensitivity of pterygium, diagnostic specificity of pterygium, and F1-score were 99%, 98%, 98.67%, 99.33%, and 99%, respectively. Among the pterygium segmentation models, the double phase-fusion PSPNet model had the best results, with MIOU, IOU, MPA, and PA of 86.57%, 78.1%, 92.3%, and 86.96%, respectively. This study designed a pterygium two-category model and a pterygium segmentation model for the images of the normal anterior and pterygium anterior segments, which could help patients self-screen easily and assist ophthalmologists in establishing the diagnosis of ophthalmic diseases and marking the actual scope of surgery.

Introduction In April 2018, the US Food and Drug Administration (FDA) approved the world’s first artificial intelligence (AI) medical device for detecting diabetic retinopathy (DR), the IDx-DR. However, there is a lack of evaluation systems for DR intelligent diagnostic technology. Methods Five hundred color fundus photographs of diabetic patients were selected. DR severity varied from grade 0 to 4, with 100 photographs for each grade. Following that, these were diagnosed by both ophthalmologists and the intelligent technology, the results of which were compared by applying the evaluation system. The system includes primary, intermediate, and advanced evaluations, of which the intermediate evaluation incorporated two methods. Main evaluation indicators were sensitivity, specificity, and kappa value. Results The AI technology diagnosed 93 photographs with no DR, 107 with mild non-proliferative DR (NPDR), 107 with moderate NPDR, 108 with severe NPDR, and 85 with proliferative DR (PDR). The sensitivity, specificity, and kappa value of the AI diagnoses in the primary evaluation were 98.8%, 88.0%, and 0.89, respectively. According to method 1 of the intermediate evaluation, the sensitivity of AI diagnosis was 98.0%, specificity 97.0%, and the kappa value 0.95. In method 2 of the intermediate evaluation, the sensitivity of AI diagnosis was 95.5%, the specificity 99.3%, and kappa value 0.95. In the advanced evaluation, the kappa value of the intelligent diagnosis was 0.86. Conclusions This article proposes an evaluation system for color fundus photograph-based intelligent diagnostic technology of DR and demonstrates an application of this system in a clinical setting. The results from this evaluation system serve as the basis for the selection of scenarios in which DR intelligent diagnostic technology can be applied.

Background: In the development of artificial intelligence in ophthalmology, the ophthalmic AI-related recognition issues are prominent, but there is a lack of research into people's familiarity with and their attitudes toward ophthalmic AI. This survey aims to assess medical workers’ and other professional technicians’ familiarity with, attitudes toward, and concerns about AI in ophthalmology.Methods: An electronic questionnaire was designed through the app Questionnaire Star, and was sent to participants through WeChat, China’s version of Facebook or WhatsApp. The participation was voluntary and anonymous. The questionnaire consisted of four parts, namely the participant’s background, their basic understanding of AI, their attitudes toward AI, and their concerns about AI. A total of 562 participants were counted, with 562 valid questionnaires returned. The results of the questionnaires are displayed in an Excel 2003 form.Results: A total of 562 participants completed the questionnaire, of whom 291 were medical workers and 271 were other professional technicians. About 1/3 of the participants understood AI and ophthalmic AI. The percentages of people who understood ophthalmic AI among medical workers and other professional technicians were about 42.61% and 15.6%, respectively. About 66.01% of the participants thought that AI in ophthalmology would partly replace doctors, with about 59.07% still having a relatively high acceptance level of ophthalmic AI. Meanwhile, among those with AI in ophthalmology application experiences (30.6%), above 70% of participants held a full acceptance attitude toward AI in ophthalmology. The participants expressed medical ethics concerns about AI in ophthalmology. And among the participants who understood AI in ophthalmology, almost all the people said that there was a need to increase the study of medical ethics issues in the ophthalmic AI field.Conclusions: The survey results revealed that the medical workers had a higher understanding level of AI in ophthalmology than other professional technicians, making it necessary to popularize ophthalmic AI education among other professional technicians. Most of the participants did not have any experience in ophthalmic AI but generally had a relatively high acceptance level of AI in ophthalmology, and there was a need to strengthen research into medical ethics issues of the field.

BACKGROUND Artificial intelligence (AI) is widely applied in the medical field, especially in ophthalmology. In the development of ophthalmic artificial intelligence, some problems worthy of attention have gradually emerged, among which the ophthalmic AI-related recognition issues are particularly prominent. That is to say, currently, there is a lack of research into people's familiarity with and their attitudes toward ophthalmic AI. OBJECTIVE This survey aims to assess medical workers’ and other professional technicians’ familiarity with AI, as well as their attitudes toward and concerns of ophthalmic AI. METHODS An electronic questionnaire was designed through the Questionnaire Star APP, an online survey software and questionnaire tool, and was sent to relevant professional workers through Wechat, China’s version of Facebook or WhatsApp. The participation was based on a voluntary and anonymous principle. The questionnaire mainly consisted of four parts, namely the participant’s background, the participant's basic understanding of AI, the participant's attitude toward AI, and the participant's concerns about AI. A total of 562 participants were counted, with 562 valid questionnaires returned. The results of the questionnaires are displayed in an Excel 2003 form. RESULTS A total of 562 professional workers completed the questionnaire, of whom 291 were medical workers and 271 were other professional technicians. About 37.9% of the participants understood AI, and 31.67% understood ophthalmic AI. The percentages of people who understood ophthalmic AI among medical workers and other professional technicians were about 42.61% and 15.6%, respectively. About 66.01% of the participants thought that ophthalmic AI would partly replace doctors, with about 59.07% still having a relatively high acceptance level of ophthalmic AI. Meanwhile, among those with ophthalmic AI application experiences (30.6%), respectively about 84.25% of medical professionals and 73.33% of other professional technicians held a full acceptance attitude toward ophthalmic AI. The participants expressed concerns that ophthalmic AI might bring about issues such as the unclear definition of medical responsibilities, the difficulty of ensuring service quality, and the medical ethics risks. And among the medical workers and other professional technicians who understood ophthalmic AI, 98.39%, and 95.24%, respectively, said that there was a need to increase the study of medical ethics issues in the ophthalmic AI field. CONCLUSIONS Analysis of the questionnaire results shows that the medical workers have a higher understanding level of ophthalmic AI than other professional technicians, making it necessary to popularize ophthalmic AI education among other professional technicians. Most of the participants did not have any experience in ophthalmic AI, but generally had a relatively high acceptance level of ophthalmic AI, believing that doctors would partly be replaced by it and that there was a need to strengthen research into medical ethics issues of the field.

In a large number of bidding cases, there exists a widespread phenomenon of supplier group hustling or lowering the price to win the bid. In order to solve this problem and curb this trend, we propose a supplier group identification framework based on supplier bidding behavior, business-relationship and geo-relationship. For the supplier bidding quotation data, we adopt the method of correlation coefficient of bidding sequence to extract feature vectors and the method of clustering to find groups. For the business-relationship data, we crawl the industrial and commercial information from the industrial and commercial website to build the graph database and find the interest-association group. For geo-relationship data, we divide supplier groups based on different administrative level regions. Finally, we innovatively introduce a group-results merging algorithm based on Apriori correlation analysis to merge multiple group results. In the B-value test, after removing the group found in our framework, the average price of effective quotation can be significantly reduced and be closer to the estimated cost price by state grid. So our framework is validated effectively in supplier group recognition through experiments.

This paper discusses the design of an indoor mobile robot positioning system. The problem of indoor positioning is solved through Wi-Fi fingerprint positioning to implement a low cost deployment. A wireless fingerprint matching algorithm based on the similarity of unequal length sequences is presented. Candidate sequences selection is defined as a set of mappings, and detection errors caused by wireless hotspot stability and the change of interior pattern can be corrected by transforming the unequal length sequences into equal length sequences. The presented scheme was verified experimentally to achieve the accuracy requirements for an indoor positioning system with low deployment cost., {"references":["Gressmann B, Klimek H, Turau V. Towards ubiquitous indoor location based services and indoor navigation (C). Positioning Navigation and Communication (WPNC), 2010.","Stewart K, Glanville J L, Bennett D A. Exploring spatiotemporal and social network factors in community response to major flood disaster (J). The Professional Geographer, 2014, 66(3): 421-435.","Yadav M S, Valck K D, Henning-Thurau T, Hoffman D L. Social commerce: A contingency frameworks for assessing marketing potential (J). Journal of Interactive Marketing, 2013, 27(4): 311-323.","Lin Yiming, Luo Haiyong, Li Jintao, Zhao Fang. Dynamic Radio Map Based Particle Filter for Indoor Wireless Localization(J). Journal of Computer Research and Development, 2011, 48(1): 139-146.","Yang Guanglong, Kong Yongping, Zhong Zhimin, Zhang Yuliang. Multimode-fingerprint Matching Based Indoor Positioning System Design and Implementation (J). Computer Engineering and Design, 2013, 4(5): 1896-1901.","Schulcz R, Varga G, Toth L. Indoor location services and context-sensitive applications in wireless networks (C). Indoor Positioning and Indoor Navigation (IPIN). 2010.","Wu K S, Xiao J, Yi Y W, et al. CSI-based indoor localization (J). IEEE Transactions on Parallel and Distributed Systems, 2013, 24(7): 1300-1309.","Ville, Tommi, et al. A Comparative Survey of WLAN Location Fingerprinting Methods (J). Proc. of the 6th Workshop on Positioning, Navigation and Communication 2009, 243-251.","Deng Zhongliang. Indoor and Outdoor Wireless Positioning and Navigation (M). Beijing: Beijing University of Posts and Telecommunications Press, 2013.\n[10]\tSchulcz R, Varga G, Toth L. Indoor location services and context-sensitive applications in wireless networks (C). Indoor Positioning and Indoor Navigation(IPIN). 2010. \n[11]\tKushki A, Plataniotis K N, Venetsanopoulos A N. Intelligent Dynamic Radio Tracking in Indoor Wireless Local Area Networks (J). IEEE Transactions on Mobile Computing, 2010, 9(3): 405-419. \n[12]\tWu K S, Xiao J, Yi Y W, et al. CSI-based indoor localization (J). IEEE Transactions on Parallel and Distributed Systems, 2013, 24(7): 1300-1309. \n[13]\tZhang Mingyang, Chen Jian, Wen Yingyou, Zhao Hong. Wi-Fi Fingerprint Localization Algorithm Based on Sliding Window Combined with Longest Common Subsequence (J). Journal of Northeastern University (Natural Science), 2014, 35(10): 1390-1394. \n[14]\tLiu Yong, Jiang Yunliang, Huang Lican. Modeling Complex Architectures Based on Granular Computing on Ontology. IEEE T. Fuzzy Systems, 2010, 18: 585-598.\n[15]\tGuan Xin, Sun Guidong, Yi Xiao, Guo Qiang. Data Association Algorithm Based on Unequal Length Sequence Data Similarity Mining (J). Control and Decision, 2015, 30(6): 1033-1038.\n[16]\tLiu Yong, Huang Wenliang, Jiang Yunliang, Zeng Zhiyong. Quick attribute reduct algorithm for neighborhood rough set model. Information Science, 2014, V271: 65-81. \n[17]\tJiang Yunliang, Xu Yunxi, Liu Yong. Performance Evaluation of Feature and Matching in Stereo Visual Odometry. Neurocomputing, 2013, 120: 380-390.\n[18]\tBen Tingting, Qin Xiaolin, Xu Jianqiu. Index of Indoor Moving Objects for Multiple Queries (J). Journal of Computer Research and Development, 2015, 52(9): 2002-2013,2015. \n[19]\tHyo-Sang Lim, Kyu-Young Whang, Yang-Sae Moon. Similar sequence matching supporting variable-length and variable-tolerance continuous queries on time-series data stream (J). Information Sciences, 178(2008): 1461-1478."]}

AGM model is the notable methodology in belief revision. But there are many AGM belief change operators when contract or revise a set of formulas by a formula. In order to obtain more useful belief change operator, researchers increase some interesting postulates to AGM postulates. On the one hand, Nebel give Prioritized belief change operator, a special AGM belief change operator, based on a relation over belief base/set. On the other hand, Parikh, Kourousias and Makinson formulated local belief change by the finest splitting over belief base/set. Naturally, whether there is a prioritized belief revision operator respects AGM postulates and relevance criterion or not? In the paper, we will answer the problem negatively by an example.

Belief revision is an interesting and important and interesting topic in artificial intelligence. AGM model is the famous methodology in belief revision. But almost all original beliefs might be changed in AGM model. In reality, there are some important beliefs in entire belief set. When belief revision, we wish belief revision operator to protect their unchangeableness as much as possible. For belief set with many important subsets ($ I_1\subseteq I_2\subseteq ...\subseteq I_n$), researchers hope that the belief revision operators unchange subset $I_n$ whenever possible, otherwise falling back on $I_{n-1}$ etc. It is namely the new belief revision, fallback revision in the paper. We also show a representation theorem for the fallback revision.

Removing noises and increasing articulation are two important aims in digital image processing. But they are inconsistent in many methods. To enhance the definition and vision effect of digital images, a novel image enhancement algorithm named CT-PCA is proposed based on many methods. Firstly, on the basis of low-pass filter, adaptive gray mathematical morphology has been adopted in contourlet domain coefficients. On the other hand, image de-noising is executed on different frequency sub-images by kernel principal component analysis method. Finally, the clearest image is acquired with inverse contourlet transform. So contourlet transform, kernel principal component analysis and gray mathematical morphology are harmonious in this method. The simulation experimental results on images show that the proposed algorithm not only decreases image noise effectively, but also improves image PSNR.