After drawbacks and shortages of using conventional kV or MV imaging mode were analyzed, this study proposes a new position verification mode with using the energy larger than 15 MeV or nominal accelerating potential ...After drawbacks and shortages of using conventional kV or MV imaging mode were analyzed, this study proposes a new position verification mode with using the energy larger than 15 MeV or nominal accelerating potential greater than 25 MV X-Ray. The new position verification mode is named HMV imaging mode. Along with the comparison of theoretical analyses, phantom experiments and clinical results to the original imaging modes, this report is going to demonstrate the HMV imaging mode is superior to traditional kV and MV imaging modes. This report first theoretically analyzed three main effects of X-ray interacting with medium by numerous equations and compared their mass attenuation coefficient with different types of tissue. X-ray irradiated on a “Catphan 500” cylinder phantom with different energies to verify these theoretical results. Furthermore, based on phantom experiments’ results, we have done numerous clinical trials and comparisons with patient’s clinical results. The theoretical and experimental results illustrate that the scanned images from HMV mode have a good quality and have ability to identify different tissue components clearly. HMV imaging mode overcomes drawbacks of position verification from both kV and MV level imaging mode as well as keeping advantages of kV and MV imaging mode. The result indicates that HMV is a good position verification mode in radiotherapy.展开更多
This paper introduces the system of game-theoretic interactions,which connects both the explanation of knowledge encoded in a deep neural networks(DNN)and the explanation of the representation power of a DNN.In this s...This paper introduces the system of game-theoretic interactions,which connects both the explanation of knowledge encoded in a deep neural networks(DNN)and the explanation of the representation power of a DNN.In this system,we define two gametheoretic interaction indexes,namely the multi-order interaction and the multivariate interaction.More crucially,we use these interaction indexes to explain feature representations encoded in a DNN from the following four aspects:(1)Quantifying knowledge concepts encoded by a DNN;(2)Exploring how a DNN encodes visual concepts,and extracting prototypical concepts encoded in the DNN;(3)Learning optimal baseline values for the Shapley value,and providing a unified perspective to compare fourteen different attribution methods;(4)Theoretically explaining the representation bottleneck of DNNs.Furthermore,we prove the relationship between the interaction encoded in a DNN and the representation power of a DNN(e.g.,generalization power,adversarial transferability,and adversarial robustness).In this way,game-theoretic interactions successfully bridge the gap between“the explanation of knowledge concepts encoded in a DNN”and"the explanation of the representation capacity of a DNN"as a unified explanation.展开更多
文摘After drawbacks and shortages of using conventional kV or MV imaging mode were analyzed, this study proposes a new position verification mode with using the energy larger than 15 MeV or nominal accelerating potential greater than 25 MV X-Ray. The new position verification mode is named HMV imaging mode. Along with the comparison of theoretical analyses, phantom experiments and clinical results to the original imaging modes, this report is going to demonstrate the HMV imaging mode is superior to traditional kV and MV imaging modes. This report first theoretically analyzed three main effects of X-ray interacting with medium by numerous equations and compared their mass attenuation coefficient with different types of tissue. X-ray irradiated on a “Catphan 500” cylinder phantom with different energies to verify these theoretical results. Furthermore, based on phantom experiments’ results, we have done numerous clinical trials and comparisons with patient’s clinical results. The theoretical and experimental results illustrate that the scanned images from HMV mode have a good quality and have ability to identify different tissue components clearly. HMV imaging mode overcomes drawbacks of position verification from both kV and MV level imaging mode as well as keeping advantages of kV and MV imaging mode. The result indicates that HMV is a good position verification mode in radiotherapy.
基金supported by National Science and Technology Major Project(No.2021ZD0111602)the National Nature Science Foundation of China(Nos.62276165 and U19B2043)Shanghai Natural Science Foundation,China(Nos.21JC1403800 and 21ZR1434600).
文摘This paper introduces the system of game-theoretic interactions,which connects both the explanation of knowledge encoded in a deep neural networks(DNN)and the explanation of the representation power of a DNN.In this system,we define two gametheoretic interaction indexes,namely the multi-order interaction and the multivariate interaction.More crucially,we use these interaction indexes to explain feature representations encoded in a DNN from the following four aspects:(1)Quantifying knowledge concepts encoded by a DNN;(2)Exploring how a DNN encodes visual concepts,and extracting prototypical concepts encoded in the DNN;(3)Learning optimal baseline values for the Shapley value,and providing a unified perspective to compare fourteen different attribution methods;(4)Theoretically explaining the representation bottleneck of DNNs.Furthermore,we prove the relationship between the interaction encoded in a DNN and the representation power of a DNN(e.g.,generalization power,adversarial transferability,and adversarial robustness).In this way,game-theoretic interactions successfully bridge the gap between“the explanation of knowledge concepts encoded in a DNN”and"the explanation of the representation capacity of a DNN"as a unified explanation.
