In lung nodules there is a huge variation in structural properties like Shape, Surface Texture. Even the spatial properties vary, where they can be found attached to lung walls, blood vessels in complex non-homogenous...In lung nodules there is a huge variation in structural properties like Shape, Surface Texture. Even the spatial properties vary, where they can be found attached to lung walls, blood vessels in complex non-homogenous lung structures. Moreover, the nodules are of small size at their early stage of development. This poses a serious challenge to develop a Computer aided diagnosis (CAD) system with better false positive reduction. Hence, to reduce the false positives per scan and to deal with the challenges mentioned, this paper proposes a set of three diverse 3D Attention based CNN architectures (3D ACNN) whose predictions on given low dose Volumetric Computed Tomography (CT) scans are fused to achieve more effective and reliable results. Attention mechanism is employed to selectively concentrate/weigh more on nodule specific features and less weight age over other irrelevant features. By using this attention based mechanism in CNN unlike traditional methods there was a significant gain in the classification performance. Contextual dependencies are also taken into account by giving three patches of different sizes surrounding the nodule as input to the ACNN architectures. The system is trained and validated using a publicly available LUNA16 dataset in a 10 fold cross validation approach where a competition performance metric (CPM) score of 0.931 is achieved. The experimental results demonstrate that either a single patch or a single architecture in a one-to-one fashion that is adopted in earlier methods cannot achieve a better performance and signifies the necessity of fusing different multi patched architectures. Though the proposed system is mainly designed for pulmonary nodule detection it can be easily extended to classification tasks of any other 3D medical diagnostic computed tomography images where there is a huge variation and uncertainty in classification.展开更多
Protons interact with human tissue differently than do photons and these differences can be exploited in an attempt to improve the care of lung cancer patients. This review examines proton beam therapy(PBT) as a compo...Protons interact with human tissue differently than do photons and these differences can be exploited in an attempt to improve the care of lung cancer patients. This review examines proton beam therapy(PBT) as a component of a combined modality program for locally advanced lung cancers. It was specifically written for the non-radiation oncologist who desires greater understanding of this newer treatment modality. This review describes and compares photon(X-ray) radiotherapy(XRT) to PBT. The physical differences of these beams are described and the clinical literature is reviewed. Protons can be used to create treatment plans delivering significantly lower doses of radiation to the adjacent organs at risk(lungs, esophagus, and bone marrow) than photons. Clinically, PBT combined with chemotherapy has resulted in low rates of toxicity comparedto XRT. Early results suggest a possible improvement in survival. The clinical results of proton therapy in lung cancer patients reveal relatively low rates of toxicity and possible survival benefits. One randomized study is being performed and another is planned to clarify the clinical differences in patient outcome for PBT compared to XRT. Along with the development of better systemic therapy, newer forms of radiotherapy such as PBT should positively impact the care of lung cancer patients. This review provides the reader with the current status of this new technology in treating locally advanced lung cancer.展开更多
Background： The detection of solitary pulmonary nodules （SPNs） that may potentially develop into a malignant lesion is essential for early clinical interventions. However, grading classification based on computed t...Background： The detection of solitary pulmonary nodules （SPNs） that may potentially develop into a malignant lesion is essential for early clinical interventions. However, grading classification based on computed tomography （CT） imaging results remains a significant challenge. The 2-[^18F]-fluoro-2-deoxy-D-glucose （^18F-FDG） positron emission tomography （PET）/CT imaging produces both false-positive and false-negative findings for the diagnosis of SPNs. In this study, we compared 18F-FDG and 3-deoxy-3-[^18F]-fluorothymidine （^18F-FLT） in lung cancer PET/CT imaging. Methods： The binding ratios of the two tracers to A549 lung cancer cells were calculated. The mouse lung cancer model was established （n = 12）, and micro-PET/CT analysis using the two tracers was performed. Images using the two tracers were collected from 55 lung cancer patients with SPNs. The correlation among the cell-tracer binding ratios, standardized uptake values （SUVs）, and Ki-67 proliferation marker expression were investigated. Results： The cell-tracer binding ratio for the A549 cells using the ^18F-FDG was greater than the ratio using 18F-FLT （P 〈 0.05）. The Ki-67 expression showed a significant positive correlation with the ^18F-FLT binding ratio （r = 0.824, P〈 0.01）. The tumor-to-nontumor uptake ratio of ^18F-FDG imaging in xenografts was higher than that of ^18F-FLT imaging. The diagnostic sensitivity, specificity, and the accuracy of ^18F-FDG for lung cancer were 89%, 67%, and 73%, respectively. Moreover, the diagnostic sensitivity, specificity, and the accuracy of ^18F-FLT for lung cancer were 71%, 79%, and 76%, respectively. There was an obvious positive correlation between the lung cancer Ki-67 expression and the mean maximum SUV of ^18F-FDG and ^18F-FLT （r = 0.658, P〈 0.05 and r = 0.724, P〈 0.01, respectively）. Conclusions： The ^18F-FDG uptake ratio is higher than that of ^18F-FLT in A549 cells at the cellular level.^18F-FLT imaging might be superior for the quantitative diagnosis of lung tumor tissue and could distinguish lung cancer nodules from other SPNs.展开更多
文摘In lung nodules there is a huge variation in structural properties like Shape, Surface Texture. Even the spatial properties vary, where they can be found attached to lung walls, blood vessels in complex non-homogenous lung structures. Moreover, the nodules are of small size at their early stage of development. This poses a serious challenge to develop a Computer aided diagnosis (CAD) system with better false positive reduction. Hence, to reduce the false positives per scan and to deal with the challenges mentioned, this paper proposes a set of three diverse 3D Attention based CNN architectures (3D ACNN) whose predictions on given low dose Volumetric Computed Tomography (CT) scans are fused to achieve more effective and reliable results. Attention mechanism is employed to selectively concentrate/weigh more on nodule specific features and less weight age over other irrelevant features. By using this attention based mechanism in CNN unlike traditional methods there was a significant gain in the classification performance. Contextual dependencies are also taken into account by giving three patches of different sizes surrounding the nodule as input to the ACNN architectures. The system is trained and validated using a publicly available LUNA16 dataset in a 10 fold cross validation approach where a competition performance metric (CPM) score of 0.931 is achieved. The experimental results demonstrate that either a single patch or a single architecture in a one-to-one fashion that is adopted in earlier methods cannot achieve a better performance and signifies the necessity of fusing different multi patched architectures. Though the proposed system is mainly designed for pulmonary nodule detection it can be easily extended to classification tasks of any other 3D medical diagnostic computed tomography images where there is a huge variation and uncertainty in classification.
基金Supported by Mayo Clinic provided the authors the time to write this manuscript.Conflict of Interest Statement:None of the authors has a conflict of interest regarding this manuscript
文摘Protons interact with human tissue differently than do photons and these differences can be exploited in an attempt to improve the care of lung cancer patients. This review examines proton beam therapy(PBT) as a component of a combined modality program for locally advanced lung cancers. It was specifically written for the non-radiation oncologist who desires greater understanding of this newer treatment modality. This review describes and compares photon(X-ray) radiotherapy(XRT) to PBT. The physical differences of these beams are described and the clinical literature is reviewed. Protons can be used to create treatment plans delivering significantly lower doses of radiation to the adjacent organs at risk(lungs, esophagus, and bone marrow) than photons. Clinically, PBT combined with chemotherapy has resulted in low rates of toxicity comparedto XRT. Early results suggest a possible improvement in survival. The clinical results of proton therapy in lung cancer patients reveal relatively low rates of toxicity and possible survival benefits. One randomized study is being performed and another is planned to clarify the clinical differences in patient outcome for PBT compared to XRT. Along with the development of better systemic therapy, newer forms of radiotherapy such as PBT should positively impact the care of lung cancer patients. This review provides the reader with the current status of this new technology in treating locally advanced lung cancer.
基金This study was supported by grants from the National Natural Science Foundation of China （No. 81271607）, and the National Postdoctoral Science Foundation of China （No. 2015M572810）.
文摘Background： The detection of solitary pulmonary nodules （SPNs） that may potentially develop into a malignant lesion is essential for early clinical interventions. However, grading classification based on computed tomography （CT） imaging results remains a significant challenge. The 2-[^18F]-fluoro-2-deoxy-D-glucose （^18F-FDG） positron emission tomography （PET）/CT imaging produces both false-positive and false-negative findings for the diagnosis of SPNs. In this study, we compared 18F-FDG and 3-deoxy-3-[^18F]-fluorothymidine （^18F-FLT） in lung cancer PET/CT imaging. Methods： The binding ratios of the two tracers to A549 lung cancer cells were calculated. The mouse lung cancer model was established （n = 12）, and micro-PET/CT analysis using the two tracers was performed. Images using the two tracers were collected from 55 lung cancer patients with SPNs. The correlation among the cell-tracer binding ratios, standardized uptake values （SUVs）, and Ki-67 proliferation marker expression were investigated. Results： The cell-tracer binding ratio for the A549 cells using the ^18F-FDG was greater than the ratio using 18F-FLT （P 〈 0.05）. The Ki-67 expression showed a significant positive correlation with the ^18F-FLT binding ratio （r = 0.824, P〈 0.01）. The tumor-to-nontumor uptake ratio of ^18F-FDG imaging in xenografts was higher than that of ^18F-FLT imaging. The diagnostic sensitivity, specificity, and the accuracy of ^18F-FDG for lung cancer were 89%, 67%, and 73%, respectively. Moreover, the diagnostic sensitivity, specificity, and the accuracy of ^18F-FLT for lung cancer were 71%, 79%, and 76%, respectively. There was an obvious positive correlation between the lung cancer Ki-67 expression and the mean maximum SUV of ^18F-FDG and ^18F-FLT （r = 0.658, P〈 0.05 and r = 0.724, P〈 0.01, respectively）. Conclusions： The ^18F-FDG uptake ratio is higher than that of ^18F-FLT in A549 cells at the cellular level.^18F-FLT imaging might be superior for the quantitative diagnosis of lung tumor tissue and could distinguish lung cancer nodules from other SPNs.
