Artificial intelligence(AI)assisted ultrasound report generation represents a technology that leverages artificial intelligence to convert ultrasound imaging analysis results into structured diagnostic reports.By inte...Artificial intelligence(AI)assisted ultrasound report generation represents a technology that leverages artificial intelligence to convert ultrasound imaging analysis results into structured diagnostic reports.By integrating image recognition and natural language generation models,AI systems can automatically detect and analyze lesions or abnormalities in ultrasound images,generating textual descriptions of diagnostic conclusions(e.g.,fatty liver,liver fibrosis,automated BIRADS grading of breast lesions),imaging findings,and clinical recommendations to form comprehensive reports.This technology enhances the efficiency and accuracy of imaging diagnosis,reduces physicians’workloads,ensures report standardization and consistency,and provides robust support for clinical decisionmaking.Current state-of-the-art algorithms for automated ultrasound report generation primarily rely on vision-language models,which harness the generalization capabilities of large language models and large vision models through multimodal(language+vision)feature alignment.However,existing approaches inadequately address challenges such as numerical measurement generation,effective utilization of report templates,incorporation of historical reports,learning text-image correlations,and overfitting under limited data conditions.This paper aims to introduce the current state of research on ultrasound report generation,the existing issues,and to provide some thoughts for future research.展开更多
Radiology report generation is of significant importance.Unlike standard image captioning tasks,radiology report generation faces more pronounced visual and textual biases due to constrained data availability,making i...Radiology report generation is of significant importance.Unlike standard image captioning tasks,radiology report generation faces more pronounced visual and textual biases due to constrained data availability,making it increasingly reliant on prior knowledge in this context.In this paper,we introduce a radiology report generation network termed Dynamics Priori Networks(DPN),which leverages a dynamic knowledge graph and prior knowledge.Concretely,we establish an adaptable graph network and harness both medical domain knowledge and expert insights to enhance the model’s intelligence.Notably,we introduce an image-text contrastive module and an image-text matching module to enhance the quality of the generated results.Our method is evaluated on two widely available datasets:X-ray collection from Indiana University(IU X-ray)and Medical Information Mart for Intensive Care,Chest X-Ray(MIMIC-CXR),where it demonstrates superior performance,particularly excelling in critical metrics.展开更多
Tests involving a large number of test cases and test scenarios are always time- and effort- intensive, and use ad hoc approaches. Test management is needed to control the complexity and the qual- ity of the testing o...Tests involving a large number of test cases and test scenarios are always time- and effort- intensive, and use ad hoc approaches. Test management is needed to control the complexity and the qual- ity of the testing of large software systems. The reporting mechanism is critical for monitoring the testing progress, analyzing test results, and evaluating the test effectiveness for a disciplined testing process throughout the testing lifecycle. This paper presents an XML-based report generation method for large sys- tem testing. The service-oriented architecture enables flexible test report generation, presentation, and ex- change to facilitate collaboration in a distributed environment. The results show that proper reporting can ef- fectively improve the visibility of the testing process and that this web-based approach is critical to enhance communication among multiple testing groups.展开更多
Epigenetic reprogramming of somatic cells into induced pluripotent stem cells （iPSCs） by overexpression of defined factors holds great promise for disease modeling and regen- erative medicine （Takahashi and Yamanak...Epigenetic reprogramming of somatic cells into induced pluripotent stem cells （iPSCs） by overexpression of defined factors holds great promise for disease modeling and regen- erative medicine （Takahashi and Yamanaka, 2006; Robinton and Daley, 2012）. However, the stochastic reprogramming process often results in variable pluripotency levels of iPSC lines as measured by their in vivo developmental potential, which poses a huge challenge to the applications of high quality iPSCs （Hanna et al., 2010）. The activation status of an imprinted Dlkl-Dio3 region has been identified as a molecular marker for pluripotency （Liu et al., 2010; Stadtfeld et al.,展开更多
In medical X-ray images,multiple abnormalities may occur frequently.However,existing report generation methods cannot efficiently extract all abnormal features,resulting in incomplete disease diagnoses when generating...In medical X-ray images,multiple abnormalities may occur frequently.However,existing report generation methods cannot efficiently extract all abnormal features,resulting in incomplete disease diagnoses when generating diagnostic reports.In real medical scenarios,there are co-occurrence relations among multiple diseases.If such co-occurrence relations are mined and integrated into the feature extraction process,the issue of missing abnormal features may be addressed.Inspired by this observation,we propose a novel method to improve the extraction of abnormal features in images through joint probability graph reasoning.Specifically,to reveal the co-occurrence relations among multiple diseases,we conduct statistical analyses on the dataset,and extract disease relationships into a probability map.Subsequently,we devise a graph reasoning network for conducting correlation-based reasoning over the features of medical images,which can facilitate the acquisition of more abnormal features.Furthermore,we introduce a gating mechanism focused on cross-modal features fusion into the current text generation model.This optimization substantially improves the model’s capabilities to learn and fuse information from two distinct modalities—medical images and texts.Experimental results on the IU-X-Ray and MIMIC-CXR datasets demonstrate that our approach outperforms previous state-of-the-art methods,exhibiting the ability to generate higher quality medical image reports.展开更多
Mining the data from the huge collection that are present in the database and uncovering the relationships between the item set are one of the key aspects of data mining technologies. Itinerary planning system with pe...Mining the data from the huge collection that are present in the database and uncovering the relationships between the item set are one of the key aspects of data mining technologies. Itinerary planning system with personalization in selecting the places to the users is one of the demanding features in most of the travel plan. In this work, the system is designed in such a way to provide the customized journey plan to the users and also the effective one to the back pack travelers. Here the Points of Interests are the places to visit in each destination for the number of days chosen by the travelers. In this system, the users are allowed to specify the desired POIs to visit for the selected destination and can make their customized travel plan effectively. This proposed system is designed to choose the customized places to visit and to plan travel for K-day itineraries. The most visited itineraries are saved and updated in the database. Association rules are used to find out the frequent places visited in each destination and to provide the reputed places to the users to plan the journey. Here the Weka tool is used to evaluate the performance of the algorithm and the rules that are generated for the given travel dataset. Data set is designed by considering several attributes that can take part during travel such as source, destination, travel cost, budget, etc. Statistical analysis is done to evaluate the performance of the proposed system and the list of features that are present in the system. During the analysis part, registered users, number of logins, frequent visits, and attributes are analyzed. Thus the system can be redefined further with the help of this statistical analysis. It is mostly used at the organization end to evaluate their performance and improve the features. Report is generated once the user has chosen their customized places to visit and all detailed description of journey is presented to the user. Report could be saved at the user end and they can use it for the future reference. Thus the goal of the system is to provide the customized travel with personalization in choosing POIs and to find the frequent places visited with desired amenities.展开更多
文摘Artificial intelligence(AI)assisted ultrasound report generation represents a technology that leverages artificial intelligence to convert ultrasound imaging analysis results into structured diagnostic reports.By integrating image recognition and natural language generation models,AI systems can automatically detect and analyze lesions or abnormalities in ultrasound images,generating textual descriptions of diagnostic conclusions(e.g.,fatty liver,liver fibrosis,automated BIRADS grading of breast lesions),imaging findings,and clinical recommendations to form comprehensive reports.This technology enhances the efficiency and accuracy of imaging diagnosis,reduces physicians’workloads,ensures report standardization and consistency,and provides robust support for clinical decisionmaking.Current state-of-the-art algorithms for automated ultrasound report generation primarily rely on vision-language models,which harness the generalization capabilities of large language models and large vision models through multimodal(language+vision)feature alignment.However,existing approaches inadequately address challenges such as numerical measurement generation,effective utilization of report templates,incorporation of historical reports,learning text-image correlations,and overfitting under limited data conditions.This paper aims to introduce the current state of research on ultrasound report generation,the existing issues,and to provide some thoughts for future research.
基金supported by the Strategic Priority Research Program of Chinese Academy of Sciences(No.XDB38050100)the Shenzhen Science and Technology Program(No.SGDX20201103095603009)the Shenzhen Polytechnic Research Fund(No.6023310009K).
文摘Radiology report generation is of significant importance.Unlike standard image captioning tasks,radiology report generation faces more pronounced visual and textual biases due to constrained data availability,making it increasingly reliant on prior knowledge in this context.In this paper,we introduce a radiology report generation network termed Dynamics Priori Networks(DPN),which leverages a dynamic knowledge graph and prior knowledge.Concretely,we establish an adaptable graph network and harness both medical domain knowledge and expert insights to enhance the model’s intelligence.Notably,we introduce an image-text contrastive module and an image-text matching module to enhance the quality of the generated results.Our method is evaluated on two widely available datasets:X-ray collection from Indiana University(IU X-ray)and Medical Information Mart for Intensive Care,Chest X-Ray(MIMIC-CXR),where it demonstrates superior performance,particularly excelling in critical metrics.
基金Supported by the National Key Basic Research and Development (973) Program of China (No. 041250001)
文摘Tests involving a large number of test cases and test scenarios are always time- and effort- intensive, and use ad hoc approaches. Test management is needed to control the complexity and the qual- ity of the testing of large software systems. The reporting mechanism is critical for monitoring the testing progress, analyzing test results, and evaluating the test effectiveness for a disciplined testing process throughout the testing lifecycle. This paper presents an XML-based report generation method for large sys- tem testing. The service-oriented architecture enables flexible test report generation, presentation, and ex- change to facilitate collaboration in a distributed environment. The results show that proper reporting can ef- fectively improve the visibility of the testing process and that this web-based approach is critical to enhance communication among multiple testing groups.
基金supported by the grants from the "Strategic Priority Research Program" of the Chinese Academy of Sciences (No. XDA01020100 to Q.Z.)the China National Basic Research Program (No. 2012CBA01300 to Q.Z.)the National Science Foundation of China (No. 91319308 to Q.Z.,31201105 to L.L. and 31371516 to W.L.)
文摘Epigenetic reprogramming of somatic cells into induced pluripotent stem cells （iPSCs） by overexpression of defined factors holds great promise for disease modeling and regen- erative medicine （Takahashi and Yamanaka, 2006; Robinton and Daley, 2012）. However, the stochastic reprogramming process often results in variable pluripotency levels of iPSC lines as measured by their in vivo developmental potential, which poses a huge challenge to the applications of high quality iPSCs （Hanna et al., 2010）. The activation status of an imprinted Dlkl-Dio3 region has been identified as a molecular marker for pluripotency （Liu et al., 2010; Stadtfeld et al.,
文摘In medical X-ray images,multiple abnormalities may occur frequently.However,existing report generation methods cannot efficiently extract all abnormal features,resulting in incomplete disease diagnoses when generating diagnostic reports.In real medical scenarios,there are co-occurrence relations among multiple diseases.If such co-occurrence relations are mined and integrated into the feature extraction process,the issue of missing abnormal features may be addressed.Inspired by this observation,we propose a novel method to improve the extraction of abnormal features in images through joint probability graph reasoning.Specifically,to reveal the co-occurrence relations among multiple diseases,we conduct statistical analyses on the dataset,and extract disease relationships into a probability map.Subsequently,we devise a graph reasoning network for conducting correlation-based reasoning over the features of medical images,which can facilitate the acquisition of more abnormal features.Furthermore,we introduce a gating mechanism focused on cross-modal features fusion into the current text generation model.This optimization substantially improves the model’s capabilities to learn and fuse information from two distinct modalities—medical images and texts.Experimental results on the IU-X-Ray and MIMIC-CXR datasets demonstrate that our approach outperforms previous state-of-the-art methods,exhibiting the ability to generate higher quality medical image reports.
文摘Mining the data from the huge collection that are present in the database and uncovering the relationships between the item set are one of the key aspects of data mining technologies. Itinerary planning system with personalization in selecting the places to the users is one of the demanding features in most of the travel plan. In this work, the system is designed in such a way to provide the customized journey plan to the users and also the effective one to the back pack travelers. Here the Points of Interests are the places to visit in each destination for the number of days chosen by the travelers. In this system, the users are allowed to specify the desired POIs to visit for the selected destination and can make their customized travel plan effectively. This proposed system is designed to choose the customized places to visit and to plan travel for K-day itineraries. The most visited itineraries are saved and updated in the database. Association rules are used to find out the frequent places visited in each destination and to provide the reputed places to the users to plan the journey. Here the Weka tool is used to evaluate the performance of the algorithm and the rules that are generated for the given travel dataset. Data set is designed by considering several attributes that can take part during travel such as source, destination, travel cost, budget, etc. Statistical analysis is done to evaluate the performance of the proposed system and the list of features that are present in the system. During the analysis part, registered users, number of logins, frequent visits, and attributes are analyzed. Thus the system can be redefined further with the help of this statistical analysis. It is mostly used at the organization end to evaluate their performance and improve the features. Report is generated once the user has chosen their customized places to visit and all detailed description of journey is presented to the user. Report could be saved at the user end and they can use it for the future reference. Thus the goal of the system is to provide the customized travel with personalization in choosing POIs and to find the frequent places visited with desired amenities.
