Objective To observe the performance of ConvNeXt architecture model(SC2-Net)integrated with feedforward attention(FA)for segmentation of pancreas from abdominal CT images.Methods 3D abdominal CT images of 80 healthy adults(Dataset 1)and 68 patients with pancreatic lesions(Dataset 2)were included.ConvNeXt network model was established and enhanced by introducing a FA mechanism,a scalable convolution block(SCB)and a feature gating(FG)module into the encoder section.The performance of the model for segmenting pancreas were comparatively evaluated with other models(Swin UNETR,nnFormer,UNETR,TransBTS models based on Transformer and 3D UX-NET model based on ConvNeXt),while conduct ablation experiments were performed on the added modules.Results SC2-Net accurately segmented pancreas from abdominal CT images,with Dice similarity coefficient(DSC),95％Hausdorff distance(HD95)and the mean surface distance(MSD)of 0.92±0.01,(1.08±0.05)mm and(2.12±0.01)mm in Dataset 1,respectively.The DSC and HD95 of SC2-Net segmentation of pancreas were both superior to those of other models.In Dataset 2,SC 2-Net achieved DSC,HD95 and MSD of 0.82±0.03,(3.35±0.36)mm and(0.87±0.15)mm,respectively,surpassing all other models.SC2-Net achieved complete pancreas segmentation in both datasets,whereas other models demonstrated under-segmentation or mis-segmentation.FA module significantly improved segmentation performance when integrated into the baseline network.Conclusion SC2-Net could improve segmentation of pancreas from abdominal CT images.

Objective To investigate the feasibility and accuracy of convolutional neural networks for automatically delineating the pancreatic cancer gross target volume(GTV)in pancreatic enhanced CT.Methods The localizable enhanced CT images of 114 patients with pancreatic cancer were retrospectively selected,in which the GTV was manually delineated using AccuContour.The imaging data were then import to AccuLearning and randomly divided as the training set,validation set and test set at a ratio of 8:1:1.Flex and Segresnet were used to train the automatic segmentation model,with each network structure trained continuously 3 times using fixed training parameters.The model was evaluated in terms of Dice similarity coefficient(DSC),95%Hausdorff distance(HD95),average symmetric surface distance(ASSD)and relative volume difference(RVD).Results In the model training phase,Flex-3 test results in Flex group were the worst,with a minimum DSC of 0.14%and an average DSC of 56.30%,while Flex-1 performed well,achieving a minimum DSC of 47.90%and an average DSC of 67.35%.Meanwhile,Segresnet-2 in Segresnet group had the worst test results,with a minimum DSC of 0.00%and an average DSC of 42.46%,while Segresnet-3 test results were better,with a minimum DSC of 42.65%and an average DSC of 63.28%.In the fixed testing phase,the best results among all were as follows:average DSC and RVD values of 63.88%and 29.41%in Segresnet-3 group,average ASSD value of 4.43 mm in Segresnet-2 group,and average HD95 value of 12.87 mm in Segresnet-1 group.Conclusion Both Flex and Segresnet architectures of convolutional neural network can be used for the automatic pancreatic tumor GTV segmentation training,with Segresnet performing better in comprehensive evaluation.

In recent years, the application of artificial intelligence technology in rectal cancer radiotherapy has become increasingly significant. By constructing models from patient clinical information, accurate prediction of dose distribution, treatment effect, and toxic side effects of rectal cancer can be achieved. This allows optimizing the radiotherapy plan, ensuring the dose is focused on the tumor target area while reducing the radiation damage to the bladder, rectum, and other surrounding tissues. Thus, it can achieve precision and personalization in radiotherapy. In this review, the construction method of artificial intelligence predictive models was described, and the value of different predictive factors to the model was systematically analyzed, including patient clinical data, radiomics, and dosimetry. Moreover, the application and limitations of artificial intelligence predictive models in radiotherapy were summarized. This information can serve as a reference for the clinical application of artificial intelligence predictive models in rectal cancer radiotherapy.

Objective To investigate the feasibility and accuracy of convolutional neural networks for automatically delineating the pancreatic cancer gross target volume(GTV)in pancreatic enhanced CT.Methods The localizable enhanced CT images of 114 patients with pancreatic cancer were retrospectively selected,in which the GTV was manually delineated using AccuContour.The imaging data were then import to AccuLearning and randomly divided as the training set,validation set and test set at a ratio of 8:1:1.Flex and Segresnet were used to train the automatic segmentation model,with each network structure trained continuously 3 times using fixed training parameters.The model was evaluated in terms of Dice similarity coefficient(DSC),95%Hausdorff distance(HD95),average symmetric surface distance(ASSD)and relative volume difference(RVD).Results In the model training phase,Flex-3 test results in Flex group were the worst,with a minimum DSC of 0.14%and an average DSC of 56.30%,while Flex-1 performed well,achieving a minimum DSC of 47.90%and an average DSC of 67.35%.Meanwhile,Segresnet-2 in Segresnet group had the worst test results,with a minimum DSC of 0.00%and an average DSC of 42.46%,while Segresnet-3 test results were better,with a minimum DSC of 42.65%and an average DSC of 63.28%.In the fixed testing phase,the best results among all were as follows:average DSC and RVD values of 63.88%and 29.41%in Segresnet-3 group,average ASSD value of 4.43 mm in Segresnet-2 group,and average HD95 value of 12.87 mm in Segresnet-1 group.Conclusion Both Flex and Segresnet architectures of convolutional neural network can be used for the automatic pancreatic tumor GTV segmentation training,with Segresnet performing better in comprehensive evaluation.

Objective To observe the performance of ConvNeXt architecture model(SC2-Net)integrated with feedforward attention(FA)for segmentation of pancreas from abdominal CT images.Methods 3D abdominal CT images of 80 healthy adults(Dataset 1)and 68 patients with pancreatic lesions(Dataset 2)were included.ConvNeXt network model was established and enhanced by introducing a FA mechanism,a scalable convolution block(SCB)and a feature gating(FG)module into the encoder section.The performance of the model for segmenting pancreas were comparatively evaluated with other models(Swin UNETR,nnFormer,UNETR,TransBTS models based on Transformer and 3D UX-NET model based on ConvNeXt),while conduct ablation experiments were performed on the added modules.Results SC2-Net accurately segmented pancreas from abdominal CT images,with Dice similarity coefficient(DSC),95％Hausdorff distance(HD95)and the mean surface distance(MSD)of 0.92±0.01,(1.08±0.05)mm and(2.12±0.01)mm in Dataset 1,respectively.The DSC and HD95 of SC2-Net segmentation of pancreas were both superior to those of other models.In Dataset 2,SC 2-Net achieved DSC,HD95 and MSD of 0.82±0.03,(3.35±0.36)mm and(0.87±0.15)mm,respectively,surpassing all other models.SC2-Net achieved complete pancreas segmentation in both datasets,whereas other models demonstrated under-segmentation or mis-segmentation.FA module significantly improved segmentation performance when integrated into the baseline network.Conclusion SC2-Net could improve segmentation of pancreas from abdominal CT images.

Objective To analyze the correlation between morphological and dosimetric parameters in patients with esophageal cancer at different locations using multi-to-multi double screening stepwise regression method,and to make simple predictions.Methods A retrospective analysis was conducted on 105 patients with advanced esophageal cancer who underwent radiotherapy at the First Affiliated Hospital of Fujian Medical University from 2019 to 2021.Morphological parameters of organs-at-risk were collected from CT images,and intensity-modulated radiotherapy plans were developed using Raystation4.7.The prescription doses for PTV-G and PTV-C were 60 Gy/30 F and 54 Gy/30 F,respectively.Multi-to-multi double screening stepwise regression method was employed to analyze the correlation between morphological and dosimetric parameters in esophageal cancer patients,and some preliminary predictions were provided.Results The dosimetric volume parameters of the lungs and heart were correlated with PTV-G volume,PTV-G length,PTV-G cross-sectional area,left and right lung volumes,lung length and total lung volume(P<0.05).For upper thoracic esophageal cancer,dosimetric volume parameters of the lungs and heart were correlated with PTV-G volume,PTV-G length,and right lung volume(P<0.05).For middle thoracic esophageal cancer,dosimetric volume parameters of the lungs,heart,and spinal cord were correlated with PTV-G volume,PTV-G length,PTV-G cross-sectional area,left and right lung volumes,and lung length(P<0.05).For lower thoracic esophageal cancer,dosimetric volume parameters of the lungs,heart,and spinal cord were correlated with PTV-G volume,PTV-G length,right lung volume,and lung length(P<0.05).Conclusion For patients with tumors at different locations,both overall and segmental analyses should be considered to balance therapeutic effect and side effects of radiotherapy,thereby maximizing the benefits for tumor patients.

Objective To analyze the correlation between morphological and dosimetric parameters in patients with esophageal cancer at different locations using multi-to-multi double screening stepwise regression method,and to make simple predictions.Methods A retrospective analysis was conducted on 105 patients with advanced esophageal cancer who underwent radiotherapy at the First Affiliated Hospital of Fujian Medical University from 2019 to 2021.Morphological parameters of organs-at-risk were collected from CT images,and intensity-modulated radiotherapy plans were developed using Raystation4.7.The prescription doses for PTV-G and PTV-C were 60 Gy/30 F and 54 Gy/30 F,respectively.Multi-to-multi double screening stepwise regression method was employed to analyze the correlation between morphological and dosimetric parameters in esophageal cancer patients,and some preliminary predictions were provided.Results The dosimetric volume parameters of the lungs and heart were correlated with PTV-G volume,PTV-G length,PTV-G cross-sectional area,left and right lung volumes,lung length and total lung volume(P<0.05).For upper thoracic esophageal cancer,dosimetric volume parameters of the lungs and heart were correlated with PTV-G volume,PTV-G length,and right lung volume(P<0.05).For middle thoracic esophageal cancer,dosimetric volume parameters of the lungs,heart,and spinal cord were correlated with PTV-G volume,PTV-G length,PTV-G cross-sectional area,left and right lung volumes,and lung length(P<0.05).For lower thoracic esophageal cancer,dosimetric volume parameters of the lungs,heart,and spinal cord were correlated with PTV-G volume,PTV-G length,right lung volume,and lung length(P<0.05).Conclusion For patients with tumors at different locations,both overall and segmental analyses should be considered to balance therapeutic effect and side effects of radiotherapy,thereby maximizing the benefits for tumor patients.