The segmentation of head and neck(H&N)tumors in dual Positron Emission Tomography/Computed Tomogra-phy(PET/CT)imaging is a critical task in medical imaging,providing essential information for diagnosis,treatment p...The segmentation of head and neck(H&N)tumors in dual Positron Emission Tomography/Computed Tomogra-phy(PET/CT)imaging is a critical task in medical imaging,providing essential information for diagnosis,treatment planning,and outcome prediction.Motivated by the need for more accurate and robust segmentation methods,this study addresses key research gaps in the application of deep learning techniques to multimodal medical images.Specifically,it investigates the limitations of existing 2D and 3D models in capturing complex tumor structures and proposes an innovative 2.5D UNet Transformer model as a solution.The primary research questions guiding this study are:(1)How can the integration of convolutional neural networks(CNNs)and transformer networks enhance segmentation accuracy in dual PET/CT imaging?(2)What are the comparative advantages of 2D,2.5D,and 3D model configurations in this context?To answer these questions,we aimed to develop and evaluate advanced deep-learning models that leverage the strengths of both CNNs and transformers.Our proposed methodology involved a comprehensive preprocessing pipeline,including normalization,contrast enhancement,and resampling,followed by segmentation using 2D,2.5D,and 3D UNet Transformer models.The models were trained and tested on three diverse datasets:HeckTor2022,AutoPET2023,and SegRap2023.Performance was assessed using metrics such as Dice Similarity Coefficient,Jaccard Index,Average Surface Distance(ASD),and Relative Absolute Volume Difference(RAVD).The findings demonstrate that the 2.5D UNet Transformer model consistently outperformed the 2D and 3D models across most metrics,achieving the highest Dice and Jaccard values,indicating superior segmentation accuracy.For instance,on the HeckTor2022 dataset,the 2.5D model achieved a Dice score of 81.777 and a Jaccard index of 0.705,surpassing other model configurations.The 3D model showed strong boundary delineation performance but exhibited variability across datasets,while the 2D model,although effective,generally underperformed compared to its 2.5D and 3D counterparts.Compared to related literature,our study confirms the advantages of incorporating additional spatial context,as seen in the improved performance of the 2.5D model.This research fills a significant gap by providing a detailed comparative analysis of different model dimensions and their impact on H&N segmentation accuracy in dual PET/CT imaging.展开更多
Bladder tumor is the most common malignant tumor in urinary system and always com- panied with lymph node metastasis. The accurate staging plays a significant role in treatment for bladder tumor and prognostic evaluat...Bladder tumor is the most common malignant tumor in urinary system and always com- panied with lymph node metastasis. The accurate staging plays a significant role in treatment for bladder tumor and prognostic evaluation, and the distant metastasis predicts worse prognosis. The objective of this study was to assess the clinical significance of 18F-FDG PET/CT imaging in diagnosing bladder tumor metastasis lesions. A retrospective analysis of 60 patients with bladder tumor from October 2008 to May 2010 was done. The patients were stratified based on the imaging technique. Among all 60 cases, besides the primary lesion, 81 suspected lesions were spotted and 73 confirmed as metastasis, including 50 lymph node metastases, 22 distant metastases, and 1 bone metastasis. For PET/CT imaging, its sensitivity was 94.5%, specificity 87.5%, positive predictive value 98.6%, negative predictive value 63.6% and accuracy 93.8% respectively. For CT, its sensitivity was 82.2%, specificity 50%, positive predictive value 93.8%, negative predictive value 23.5% and accuracy 79% respectively. PET/CT im- aging was superior to CT in sensitivity, specificity and accuracy. In conclusion, 18F-FDG PET/CT imaging is more significant in diagnosing bladder tumor metastasis lesions.展开更多
Purpose: To generate parametric images of tumor hypoxia in a tumor-bearing rat model using voxel-based compartmental analysis of dynamic fluorine-18 labeled misonidazole (18F-FMISO) microPET? images, and to compare th...Purpose: To generate parametric images of tumor hypoxia in a tumor-bearing rat model using voxel-based compartmental analysis of dynamic fluorine-18 labeled misonidazole (18F-FMISO) microPET? images, and to compare the parametric images thus derived with static “late” 18F-FMISO microPET? images for the detection of tumor hypoxia. Materials and Methods: Nude rats bearing HT-29 colorectal carcinoma xenografts (≈1.5 - 2 cm in diameter) in the right hind limb were positioned in a custom-fabricated, animal-specific foam mold. Animals were injected via the tail vein with ≈55.5 MBq 18F-FMISO and continuously imaged for either 60 or 120 minutes, with additional late static images up to 3 hour post-injection. The raw list-mode data was reconstructed into 37 - 64 frames with earlier frames of shorter time durations (12 - 15 seconds) and later frames of longer durations (up to 300 seconds). Time activity curves (TACs) were generated over regions encompassing the tumor as well as an artery, the latter for use as an input function. A beta version of a compartmental modeling package (BioGuide?, Philips Healthcare) was used to generate parametric images of k3 and Ki, rate constants of entrapment and flux of 18F-FMISO, respectively. Results: Data for 7 HT-29 tumor xenografts were presented, 6 of which yielded clear areas of tumor hypoxia as defined by Ki/k3 maps. Importantly, intratumoral foci with high 18F-FMISO uptakes on the late images did not always exhibit high Ki/k3 values and may there- fore represent false-positives for radiobiologically significant hypoxia. Conclusions: This study attempts to quantify tumor hypoxia using compartmental analysis of dynamic 18F-FMISO PET images in rodent xenograft tumor models. The results demonstrate feasibility of the approach in small-animal imaging studies, and provide evidence for the possible unreliability of late-time static imaging of 18F-FMISO PET in identifying tumor hypoxia.展开更多
We conducted a retrospective analysis of 221 subjects with 256 suspected gastrointestinal lesions from2007 to 2015 to explore the detecting efficiency of dualtime-point fluorine-18 fludeoxyglucose(^(18)F-FDG) positron...We conducted a retrospective analysis of 221 subjects with 256 suspected gastrointestinal lesions from2007 to 2015 to explore the detecting efficiency of dualtime-point fluorine-18 fludeoxyglucose(^(18)F-FDG) positron emission tomography/computed tomography(PET/CT)and pathology examination. The abdominal delayed PET/CT was performed within 45 min of the conventional scan.The change in maximum standardized uptake value(ASUV_(max)) and morphological features of the suspected lesions between the conventional and dual-time-point PET/CT were compared. The sensitivity, specificity, positive predictive value, and negative predictive value(NPV) of conventional PET/CT were 81.6%(84/103), 56.2%(86/153), 55.6%(84/151), and 81.9%(86/105), respectively.Those of dual-time-point PET/CT were 94.1%(97/103),78.4%(120/153), 74.6%(97/130), and 95.2%(120/126),respectively. There was a significant difference between the conventional and dual-time-point PET/CT(P < 0.005).The SUV_(early) and the %△SUV_(max) could not present more information in differential diagnoses, but the rate of tumors with increased SUVdelay accounted for 79.6%(82/103) and more than that of nonneoplastic lesions(15.5%, 29/187)(x^2= 115.5, P < 0.01). Therefore, the dual-time-point^(18)F-FDG PET/CT had a higher sensitivity and NPV than the conventional PET/CT to detect gastrointestinal tumors.The constant morphology and increased SUV_(delay) help to detect the tumors and adding delayed imaging on the locality will be an effective method when we accidentally find a suspected gastrointestinal tumor on the conventional PET/CT images.展开更多
基金supported by Scientific Research Deanship at University of Ha’il,Saudi Arabia through project number RG-23137.
文摘The segmentation of head and neck(H&N)tumors in dual Positron Emission Tomography/Computed Tomogra-phy(PET/CT)imaging is a critical task in medical imaging,providing essential information for diagnosis,treatment planning,and outcome prediction.Motivated by the need for more accurate and robust segmentation methods,this study addresses key research gaps in the application of deep learning techniques to multimodal medical images.Specifically,it investigates the limitations of existing 2D and 3D models in capturing complex tumor structures and proposes an innovative 2.5D UNet Transformer model as a solution.The primary research questions guiding this study are:(1)How can the integration of convolutional neural networks(CNNs)and transformer networks enhance segmentation accuracy in dual PET/CT imaging?(2)What are the comparative advantages of 2D,2.5D,and 3D model configurations in this context?To answer these questions,we aimed to develop and evaluate advanced deep-learning models that leverage the strengths of both CNNs and transformers.Our proposed methodology involved a comprehensive preprocessing pipeline,including normalization,contrast enhancement,and resampling,followed by segmentation using 2D,2.5D,and 3D UNet Transformer models.The models were trained and tested on three diverse datasets:HeckTor2022,AutoPET2023,and SegRap2023.Performance was assessed using metrics such as Dice Similarity Coefficient,Jaccard Index,Average Surface Distance(ASD),and Relative Absolute Volume Difference(RAVD).The findings demonstrate that the 2.5D UNet Transformer model consistently outperformed the 2D and 3D models across most metrics,achieving the highest Dice and Jaccard values,indicating superior segmentation accuracy.For instance,on the HeckTor2022 dataset,the 2.5D model achieved a Dice score of 81.777 and a Jaccard index of 0.705,surpassing other model configurations.The 3D model showed strong boundary delineation performance but exhibited variability across datasets,while the 2D model,although effective,generally underperformed compared to its 2.5D and 3D counterparts.Compared to related literature,our study confirms the advantages of incorporating additional spatial context,as seen in the improved performance of the 2.5D model.This research fills a significant gap by providing a detailed comparative analysis of different model dimensions and their impact on H&N segmentation accuracy in dual PET/CT imaging.
文摘Bladder tumor is the most common malignant tumor in urinary system and always com- panied with lymph node metastasis. The accurate staging plays a significant role in treatment for bladder tumor and prognostic evaluation, and the distant metastasis predicts worse prognosis. The objective of this study was to assess the clinical significance of 18F-FDG PET/CT imaging in diagnosing bladder tumor metastasis lesions. A retrospective analysis of 60 patients with bladder tumor from October 2008 to May 2010 was done. The patients were stratified based on the imaging technique. Among all 60 cases, besides the primary lesion, 81 suspected lesions were spotted and 73 confirmed as metastasis, including 50 lymph node metastases, 22 distant metastases, and 1 bone metastasis. For PET/CT imaging, its sensitivity was 94.5%, specificity 87.5%, positive predictive value 98.6%, negative predictive value 63.6% and accuracy 93.8% respectively. For CT, its sensitivity was 82.2%, specificity 50%, positive predictive value 93.8%, negative predictive value 23.5% and accuracy 79% respectively. PET/CT im- aging was superior to CT in sensitivity, specificity and accuracy. In conclusion, 18F-FDG PET/CT imaging is more significant in diagnosing bladder tumor metastasis lesions.
文摘Purpose: To generate parametric images of tumor hypoxia in a tumor-bearing rat model using voxel-based compartmental analysis of dynamic fluorine-18 labeled misonidazole (18F-FMISO) microPET? images, and to compare the parametric images thus derived with static “late” 18F-FMISO microPET? images for the detection of tumor hypoxia. Materials and Methods: Nude rats bearing HT-29 colorectal carcinoma xenografts (≈1.5 - 2 cm in diameter) in the right hind limb were positioned in a custom-fabricated, animal-specific foam mold. Animals were injected via the tail vein with ≈55.5 MBq 18F-FMISO and continuously imaged for either 60 or 120 minutes, with additional late static images up to 3 hour post-injection. The raw list-mode data was reconstructed into 37 - 64 frames with earlier frames of shorter time durations (12 - 15 seconds) and later frames of longer durations (up to 300 seconds). Time activity curves (TACs) were generated over regions encompassing the tumor as well as an artery, the latter for use as an input function. A beta version of a compartmental modeling package (BioGuide?, Philips Healthcare) was used to generate parametric images of k3 and Ki, rate constants of entrapment and flux of 18F-FMISO, respectively. Results: Data for 7 HT-29 tumor xenografts were presented, 6 of which yielded clear areas of tumor hypoxia as defined by Ki/k3 maps. Importantly, intratumoral foci with high 18F-FMISO uptakes on the late images did not always exhibit high Ki/k3 values and may there- fore represent false-positives for radiobiologically significant hypoxia. Conclusions: This study attempts to quantify tumor hypoxia using compartmental analysis of dynamic 18F-FMISO PET images in rodent xenograft tumor models. The results demonstrate feasibility of the approach in small-animal imaging studies, and provide evidence for the possible unreliability of late-time static imaging of 18F-FMISO PET in identifying tumor hypoxia.
文摘We conducted a retrospective analysis of 221 subjects with 256 suspected gastrointestinal lesions from2007 to 2015 to explore the detecting efficiency of dualtime-point fluorine-18 fludeoxyglucose(^(18)F-FDG) positron emission tomography/computed tomography(PET/CT)and pathology examination. The abdominal delayed PET/CT was performed within 45 min of the conventional scan.The change in maximum standardized uptake value(ASUV_(max)) and morphological features of the suspected lesions between the conventional and dual-time-point PET/CT were compared. The sensitivity, specificity, positive predictive value, and negative predictive value(NPV) of conventional PET/CT were 81.6%(84/103), 56.2%(86/153), 55.6%(84/151), and 81.9%(86/105), respectively.Those of dual-time-point PET/CT were 94.1%(97/103),78.4%(120/153), 74.6%(97/130), and 95.2%(120/126),respectively. There was a significant difference between the conventional and dual-time-point PET/CT(P < 0.005).The SUV_(early) and the %△SUV_(max) could not present more information in differential diagnoses, but the rate of tumors with increased SUVdelay accounted for 79.6%(82/103) and more than that of nonneoplastic lesions(15.5%, 29/187)(x^2= 115.5, P < 0.01). Therefore, the dual-time-point^(18)F-FDG PET/CT had a higher sensitivity and NPV than the conventional PET/CT to detect gastrointestinal tumors.The constant morphology and increased SUV_(delay) help to detect the tumors and adding delayed imaging on the locality will be an effective method when we accidentally find a suspected gastrointestinal tumor on the conventional PET/CT images.
