Guidelines on clinical research evaluation of artificial intelligence in ophthalmology(2023)

With the upsurge of artificial intelligence(AI)technology in the medical field,its application in ophthalmology has become a cutting-edge research field.Notably,machine learning techniques have shown remarkable achievements in diagnosing,intervening,and predicting ophthalmic diseases.To meet the requirements of clinical research and fit the actual progress of clinical diagnosis and treatment of ophthalmic AI,the Ophthalmic Imaging and Intelligent Medicine Branch and the Intelligent Medicine Committee of Chinese Medicine Education Association organized experts to integrate recent evaluation reports of clinical AI research at home and abroad and formed a guideline on clinical research evaluation of AI in ophthalmology after several rounds of discussion and modification.The main content includes the background and method of developing this guideline,an introduction to international guidelines on the clinical research evaluation of AI,and the evaluation methods of clinical ophthalmic AI models.This guideline introduces general evaluation methods of clinical ophthalmic AI research,evaluation methods of clinical ophthalmic AI models,and commonly-used indices and formulae for clinical ophthalmic AI model evaluation in detail,and amply elaborates the evaluation methods of clinical ophthalmic AI trials.This guideline aims to provide guidance and norms for clinical researchers of ophthalmic AI,promote the development of regularization and standardization,and further improve the overall level of clinical ophthalmic AI research evaluations.

A Nationwide Evaluation of the State of Practice of Performance Measurements for Intelligent Transportation Systems

State departments of transportation’s (DOTs) decisions to invest resources to expand or implement intelligent transportation systems (ITS) programs or even retire existing infrastructure need to be based on performance evaluations. Nonetheless, an apparent gap exists between the need for ITS performance measurements and the actual implementation. The evidence available points to challenges in the ITS performance measurement processes. This paper evaluated the state of practice of performance measurement for ITS across the US and provided insights. A comprehensive literature review assessed the use of performance measures by DOTs for monitoring implemented ITS programs. Based on the gaps identified through the literature review, a nationwide qualitative survey was used to gather insights from key stakeholders on the subject matter and presented in this paper. From the data gathered, performance measurement of ITS is fairly integrated into ITS programs by DOTs, with most agencies considering the process beneficial. There, however, exist reasons that prevent agencies from measuring ITS performance to greater detail and quality. These include lack of data, fragmented or incomparable data formats, the complexity of the endeavor, lack of data scientists, and difficulty assigning responsibilities when inter-agency collaboration is required. Additionally, DOTs do not benchmark or compare their ITS performance with others for reasons that include lack of data, lack of guidance or best practices, and incomparable data formats. This paper is relevant as it provides insights expected to guide DOTs and other agencies in developing or reevaluating their ITS performance measurement processes.

Model of Land Suitability Evaluation Based on Computational Intelligence

A novel model of land suitability evaluation is built based on computational intelligence (CI). A fuzzy neural network (FNN) is constructed by the integration of fuzzy logic and artificial neural network (ANN). The structure and process of this network is clear. Fuzzy rules (knowledge) are expressed in the model explicitly, and can be self-adjusted by learning from samples. Genetic algorithm (GA) is employed as the learning algorithm to train the network, and makes the training of the model efficient. This model is a self-learning and self-adaptive system with a rule set revised by training.

The Role of Computational Intelligence in Sensory Evaluation

Sensory evaluation is the evaluation of signals that a human receives via its senses of sight, smell, taste, touch and hearing. In today’s industrial companies, sensory evaluation is widely used in quality inspection of products, in marketing study and in many other fields such as risk evaluation, investment evaluation and safety evaluation. In practice, setting up a suitable mathematical formulation, an efficient working procedure and a pertinent computing method for sensory evaluation is quite difficult because of uncertainty and imprecision in sensory panels and their results involving linguistic expressions, non normalized data, data reliability, etc. At the present a prime problem of the practitioner is not the lack of useful methods but the lack of transparency in this area. In this tutorial lecture, we briefly describe some of the technology in the computational intelligence (CI) areas that has been developed for application to sensory evaluation and related fields. Moreover, we will illustrate the role of CI in sensory evaluation related applications from some recent publications.

Efficient Explanation and Evaluation Methodology Based on Hybrid Feature Dropout

AI-related research is conducted in various ways,but the reliability of AI prediction results is currently insufficient,so expert decisions are indispensable for tasks that require essential decision-making.XAI(eXplainable AI)is studied to improve the reliability of AI.However,each XAI methodology shows different results in the same data set and exact model.This means that XAI results must be given meaning,and a lot of noise value emerges.This paper proposes the HFD(Hybrid Feature Dropout)-based XAI and evaluation methodology.The proposed XAI methodology can mitigate shortcomings,such as incorrect feature weights and impractical feature selection.There are few XAI evaluation methods.This paper proposed four evaluation criteria that can give practical meaning.As a result of verifying with the malware data set(Data Challenge 2019),we confirmed better results than other XAI methodologies in 4 evaluation criteria.Since the efficiency of interpretation is verified with a reasonable XAI evaluation standard,The practicality of the XAI methodology will be improved.In addition,The usefulness of the XAI methodology will be demonstrated to enhance the reliability of AI,and it helps apply AI results to essential tasks that require expert decision-making.

The Economic Impact of Artificial Intelligence on Healthcare: A Literature Review

Background: This review delves into the effects of artificial intelligence (AI) on healthcare, which is a crucial aspect considering the increasing costs of healthcare worldwide. While there is potential for AI to enhance healthcare delivery and efficiency, there are still uncertainties surrounding its effectiveness, value, and broader adoption. This comprehensive literature review aims to explore and synthesize existing knowledge on the economic impact of AI in healthcare. The primary objective of this review is to understand the potential cost savings and efficiency improvements associated with the deployment of AI in healthcare settings. By highlighting the economic implications of AI, this review seeks to offer insights into the value proposition of investing in AI technologies for stakeholders such as healthcare providers, payers, and policymakers. Methods: To conduct this review, we conducted a search of literature from 2020 to 2023 across three databases: PubMed, Scopus and Google Scholar. We specifically focused on studies that discuss the impacts of AI in healthcare and include cost evaluations, using combinations of keywords related to AI, economics, healthcare, and cost evaluation. The inclusion criteria were studies that conducted some form of economic evaluation related to AI in healthcare settings, while exclusion criteria were studies without a cost evaluation component. Data extraction and quality assessment using the CASP checklist were undertaken on the final set of included studies. Results: After screening studies, we identified 10 out of a total of 28 studies and reports that met our criteria of outlining any form of economic impact and evaluation of AI in healthcare settings. Based on our findings, implementing AI in healthcare could potentially lead to cost savings. Several studies suggest savings ranging from $200 billion to $360 billion in the United States alone. The use of AI in healthcare sectors such as ophthalmology, radiology and disease screening has shown positive economic impacts. Conclusion: While AI has potential for cost savings and efficiency improvements, in healthcare settings, it’s crucial to conduct detailed context specific cost evaluations to optimize the adoption and implementation strategies of AI.

Effect Evaluation and Intelligent Prediction of Power Substation Project Considering New Energy

The evaluation of the implementation effect of the power substation project can find out the problems of the project more comprehensively,which has important practical significance for the further development of the power substation project.To ensure accuracy and real-time evaluation,this paper proposes a novel hybrid intelligent evaluation and prediction model based on improved TOPSIS and Long Short-Term Memory(LSTM)optimized by a Sperm Whale Algorithm(SWA).Firstly,under the background of considering the development of new energy,the influencing factors of power substation project implementation effect are analyzed from three aspects of technology,economy and society.Moreover,an evaluation model based on improved TOPSIS is constructed.Then,an intelligent prediction model based on SWA optimized LSTM is designed.Finally,the scientificity and accuracy of the proposed model are verified by empirical analysis,and the important factors affecting the implementation effect of power substation projects are pointed out.

An Introduction to Intelligent Sensory Evaluation

Sensory evaluation is the evaluation of signals that a buman receives via its sensory organs. Nowadays sensory evaluation is widely used in quality inspection and quality control of products. and many other fields. Actually sensory evaluation always give. uncertain and inprecise results, therefore it derivates many problems. we reviews in detail these problem and give some cumputing methods to resolve them.

Providing Performance Evaluation Indicators for Intelligent Transportation Systems (The Case Study of Tehran-Karaj Freeway Located in Iran)

Intelligent Transportation Systems (ITS) play a fundamental role in reducing traffic congestion and increasing safety during daily transportation. These systems can also be useful in improving social welfare leading to general satisfaction. Proper performance evaluation can be efficient in improving the performance of these systems, and providing a scientific assessment index system can assist decision-makers in smart communities to plan for the development of ITS. However, the evaluation of these systems requires identifying appropriate indicators of performance evaluation that are consistent with the views of the beneficiaries of these systems. In this paper, performance evaluation indicators of ITS have been identified, and three indicators entitled “environmental and safety”, “assistance in reducing traffic congestion” and “attractive public transport” are presented to evaluate the performance of these systems. Moreover, the intelligent transport systems of the Tehran-Karaj Freeway in Iran are studied, and inferential statistical methods are employed to test the research hypotheses. It is worth noticing that in this study, a one-sample T-test method is used for hypotheses assessment and the SPSS software was used to analyze the findings. Also, the results demonstrated that the performance of ITS in the Tehran-Karaj Freeway regarding the indicators, such as “Declaration of route blocking information due to maintenance or reconstruction” and “Declaration of path geometry conditions” has not been acceptable.

Sustainability Intelligent Evaluation of Regional Microgrid Interconnection System Based on Combination Entropy Weight Rank Order-TOPSIS and NILA-KELM

Sustainability evaluation of regional microgrid interconnection system is conducive to a profound and comprehensive understanding of the impact of interconnection system projects.In order to realize the comprehensive and scientific intelligent evaluation of the system,this paper proposes an evaluation model based on combination entropy weight rank order-technique for order preference by similarity to an ideal solution(TOPSIS)and Niche Immune Lion Algorithm-Extreme Learning Machine with Kernel(NILAKELM).Firstly,the sustainability evaluation indicator system of the regional microgrid interconnection system is constructed fromfour aspects of economic,environmental,social,and technical characteristics,and the evaluation indicators are explained.Then,the classical evaluationmodel based on TOPSIS is constructed,and the entropy weight method and rank order method(RO)are coupled to obtain the indicator weight.The niche immune algorithm is used to improve the lion algorithm,and the improved lion algorithm is used to optimize the parameters of KELM,and the intelligent evaluation model based on NILA-KELM is obtained to realize fast real-time calculation.Finally,the scientificity and accuracy of themodel proposed in this paper are verified.The model proposed in this paper has the lowest RMSE,MAE and RE values,indicating that its intelligent evaluation results are the most accurate.This study is conducive to the horizontal comparison of the overall performance of regional microgrid interconnection system projects,helps investors to choose the most promising project scheme,and helps the government to find feasible project.

Research on the Application of Intelligent Evaluation and Teacher-student Cooperation Assessment System in Teaching English Writing

Low efficiency in teaching and time-consuming in writing evaluation are two big problems for college English teachers.Therefore,it is necessary to create a new teaching model to solve these problems existing in traditional classroom-based teaching.This research adopts the research methods of test comparison before and after the students’composition experiment,questionnaire and semi-open interviews.Empirical research on a new teaching model that integrates the intelligent composition review and reform system represented by Piangai.com and the collaborative evaluation of teachers and students is conducted.The research results show that the new writing teaching model improves the quality of students’writing,promotes students’learning initiative,and enhances students’writing self-efficacy.This writing teaching model provides ideas for solving the problem of time-consuming and inefficient English writing teaching in large classes.

Hybrid Prediction Method for Solar Power Using Different Computational Intelligence Algorithms

Computational Intelligence (CI) holds the key to the development of smart grid to overcome the challenges of planning and optimization through accurate prediction of Renewable Energy Sources (RES). This paper presents an architectural framework for the construction of hybrid intelligent predictor for solar power. This research investigates the applicability of heterogeneous regression algorithms for 6 hour ahead solar power availability forecasting using historical data from Rockhampton, Australia. Real life solar radiation data is collected across six years with hourly resolution from 2005 to 2010. We observe that the hybrid prediction method is suitable for a reliable smart grid energy management. Prediction reliability of the proposed hybrid prediction method is carried out in terms of prediction error performance based on statistical and graphical methods. The experimental results show that the proposed hybrid method achieved acceptable prediction accuracy. This potential hybrid model is applicable as a local predictor for any proposed hybrid method in real life application for 6 hours in advance prediction to ensure constant solar power supply in the smart grid operation.

A smart productivity evaluation method for shale gas wells based on 3D fractal fracture network model

The generation method of three-dimensional fractal discrete fracture network(FDFN)based on multiplicative cascade process was developed.The complex multi-scale fracture system in shale after fracturing was characterized by coupling the artificial fracture model and the natural fracture model.Based on an assisted history matching(AHM)using multiple-proxy-based Markov chain Monte Carlo algorithm(MCMC),an embedded discrete fracture modeling(EDFM)incorporated with reservoir simulator was used to predict productivity of shale gas well.When using the natural fracture generation method,the distribution of natural fracture network can be controlled by fractal parameters,and the natural fracture network generated coupling with artificial fractures can characterize the complex system of different-scale fractures in shale after fracturing.The EDFM,with fewer grids and less computation time consumption,can characterize the attributes of natural fractures and artificial fractures flexibly,and simulate the details of mass transfer between matrix cells and fractures while reducing computation significantly.The combination of AMH and EDFM can lower the uncertainty of reservoir and fracture parameters,and realize effective inversion of key reservoir and fracture parameters and the productivity forecast of shale gas wells.Application demonstrates the results from the proposed productivity prediction model integrating FDFN,EDFM and AHM have high credibility.

Artificial Intelligence and the Future of Education： Big Promises -Bigger Challenges

The history of educational technology in the last 50 years contains few instances of dramatic improvements in learning based on the adoption of a particular technology.An example involving artificial intelligence occurred in the 1990s with the development of intelligent tutoring systems( ITSs). What happened with ITSs was that their success was limited to well-defined and relatively simple declarative and procedural learning tasks(e. g.,learning how to write a recursive function in LISP; doing multi-column addition),and improvements that were observed tended to be more limited than promised(e. g.,one standard deviation improvement at best rather than the promised standard deviation improvement).Still,there was some progress in terms of how to conceptualize learning. A seldom documented limitation was the notion of only viewing learning from only content and cognitive perspectives( i. e.,in terms of memory limitations,prior knowledge,bug libraries,learning hierarchies and sequences etc.). Little attention was paid to education conceived more broadly than developing specific cognitive skills with highly constrained problems. New technologies offer the potential to create dynamic and multi-dimensional models of a particular learner,and to track large data sets of learning activities,resources,interventions,and outcomes over a great many learners. Using those data to personalize learning for a particular learner developing knowledge,competence and understanding in a specific domain of inquiry is finally a real possibility. While the potential to make significant progress is clearly possible,the reality is less not so promising. There are many as yet unmet challenging some of which will be mentioned in this paper. A persistent worry is that educational technologists and computer scientists will again promise too much,too soon at too little cost and with too little effort and attention to the realities in schools and universities.

Hazard-evaluation-oriented Moving Horizon Parallel Steering Control for Driver-Automation Collaboration During Automated Driving

Prompted by emerging developments in connected and automated vehicles, parallel steering control, one aspect of parallel driving, has become highly important for intelligent vehicles for easing the burden and ensuring the safety of human drivers. This paper presents a parallel steering control framework for an intelligent vehicle using moving horizon optimization.The framework considers lateral stability, collision avoidance and actuator saturation and describes them as constraints, which can blend the operation of a human driver and a parallel steering controller effectively. Moreover, the road hazard and the steering operation error are employed to evaluate the operational hazardous of an intelligent vehicle. Under the hazard evaluation,the intelligent vehicle will be mainly operated by the human driver when the vehicle operates in a safe and stable manner.The automated steering driving objective will play an active role and regulate the steering operations of the intelligent vehicle based on the hazard evaluation. To verify the effectiveness of the proposed hazard-evaluation-oriented moving horizon parallel steering control approach, various validations are conducted, and the results are compared with a parallel steering scheme that does not consider automated driving situations. The results illustrate that the proposed parallel steering controller achieves acceptable performance under both conventional conditions and hazardous conditions.

Application of diversified evaluation in middle school english teaching

Most of the current evaluations are for selecting and screening students, paying too much attention to students' scores and rankings, ignoring the true feedback on the students' actual learning process, not to mention incentives, regulation,and development of the student's daily study. Most schools still use the single final evaluation method of the test to evaluate students' academic performance.

Construction and Practice of Higher Vocational Colleges’ Quality Assurance System Based on Evaluation

Higher vocational colleges pay more attention to the establishment and improvement of the internal quality assurance system. An objective and reasonable evaluation system was established by analyzing the law of the development of higher education and vocational education in the world. Through the integration of the evaluation system and the information platform, it showed the advantages of the efficient flat structure of the information platform and the zero-distance advantages of the participants. And teachers and students were served efficiently. By analyzing all kinds of data in real time, the problems were discovered in each link in time. A normal diagnosis and improvement working mechanism were gradually formed, and internal quality assurance system of sustainable development was constructed for school governance services based on objective data, realizing the continuous improvement of the quality of higher vocational talents training.

Intelligence-Aided Seismic Zonation System IASHES

In this paper,the structure of the intelligence-aided seismic zonation system IASHES and its validation are briefly introduced.Emphasis is placed on the two rank scheme of potential seismic source areas; an expert subsystem for estimating the seismicity trends of rank A source areas; an expert subsystem for delineation of rank B source areas; an expert subsystem for judgment of upper limit of magnitude of rank B source areas,and an improved procedure for determination of weighting factors of rank B source areas,which is specially suitable to ES(Expert systems).

Intelligent Parking Management System by Multi-Agent Approach:The Case of Urban Area of Tunis

By coordination and cooperation between multi-agents, this paper proposes the network of intelligent agents which can reduce the search time needed to finding a parking place. Based on multi-agent model, the fined solution is designed to help drivers in finding a parking space at anytime and anywhere. Three services are offered: the search for a vacant place, directions to a parking space and booking a place for parking. The results of this study generated by the platform MATSim transport simulation, show that our approach optimizes the operation of vehicles in a parking need with the aim of reducing congestion, and improve traffic flow in urban area. A comparison between the first method where the vehicles are random and the second method where vehicles are steered to vacant parking spaces shows that the minimization of time looking for a parking space could improve circulation by reducing the number of cars in the morning of 2% and 0.7% of the evening. In addition, the traffic per hour per day was reduced by approximately 4.17%.