Extension–contraction transformation network for pancreas segmentation in abdominal CT scans

Background and objectiveThe size, shape, and position of the pancreas are affected by the patient characteristics such as age, sex, adiposity. Owing to more complex anatomical structures (size, shape, and position) of the pancreas, specialists have some difficulties in the analysis of pancreatic diseases (diabetes, pancreatic cancer, pancreatitis). Therefore, the treatment of the disease requires enormous time and depends on the experience of specialists. In order to decrease the rate of pancreatic disease deaths and to assist the specialist in the analysis of pancreatic diseases, automatic pancreas segmentation techniques have been actively developed in the research article for many years. MethodsAlthough the rapid growth of deep learning and proving satisfactory performance in many medical image processing and computer vision applications, the maximum Dice Similarity Coefficients (DSC) value of these techniques related to automatic pancreas segmentation is only around 85% due to complex structure of the pancreas and other factors. Contrary to previous techniques which are required significantly higher computational power and memory, this paper suggests a novel two-phase approach for high-accuracy automatic pancreas segmentation in computed tomography (CT) imaging. The proposed approach consists of two phases; (1) Pancreas Localization, where the rough pancreas position is detected on the 2D CT slice by adopting Mask R-CNN model, (2) Pancreas Segmentation, where the segmented pancreas region is produced by refining the candidate pancreas region with 3D U-Net on the 2D sub-CT slices generated in the first phase. Both qualitative and quantitative assessments of the proposed approach are performed on the NIH data set. ResultsIn order to evaluate the achievement of the recommended approach, a total of 16 different automatic pancreas segmentation techniques reported in the literature are compared by utilizing performance assessment procedures which are Dice Similarity Coefficient (DSC), Jaccard Index (JI), Precision (PRE), Recall (REC), Pixel Accuracy (ACC), Specificity (SPE), Receiver Operating Characteristics (ROC) and Area under ROC curve (AUC). The average values of DSC, JI, REC and ACC are computed as 86.15%, 75.93%, 86.27%, 86.27% and 99.95% respectively, which are the best values among well-established studies for automatic pancreas segmentation. ConclusionIt is demonstrated with qualitative and quantitative results that our suggested two-phase approach creates more favorable results than other existing approaches.

Pancreas is one of the most challenging organs in segmentation due to its different shape, position and size in each human being. With the development of machine learning, various deep learning methods are applied to segment the pancreas among organs in the abdominal region. In this study, pancreas segmentation is performed using the U-Net model, which is one of the convolutional neural networks (CNN) models. The results of pancreas segmentation performed on the Pancreas CT data set obtained from The Cancer Imaging Archive (TCIA) database containing computed tomography images of 82 patients are presented in detail. According to the results, Dice similarity coefficient and Jaccard similarity coefficient are found to be 0.78 and 0.66, respectively.

Automated pancreas segmentation in medical images is a prerequisite for many clinical applications, such as diabetes inspection, pancreatic cancer diagnosis, and surgical planing. In this paper, we formulate pancreas segmentation in magnetic resonance imaging (MRI) scans as a graph based decision fusion process combined with deep convolutional neural networks (CNN). Our approach conducts pancreatic detection and boundary segmentation with two types of CNN models respectively: 1) the tissue detection step to differentiate pancreas and non-pancreas tissue with spatial intensity context; 2) the boundary detection step to allocate the semantic boundaries of pancreas. Both detection results of the two networks are fused together as the initialization of a conditional random field (CRF) framework to obtain the final segmentation output. Our approach achieves the mean dice similarity coefficient (DSC) 76.1% with the standard deviation of 8.7% in a dataset containing 78 abdominal MRI scans. The proposed algorithm achieves the best results compared with other state of the arts.

We propose an automated pancreas segmentation algorithm from contrast-enhanced multiphase computed tomography (CT) and verify its effectiveness in segmentation. The algorithm is characterized by three unique ideas. First, a two-stage segmentation strategy with spatial standardization of pancreas was employed to reduce variations in the pancreas shape and location. Second, patient- specific probabilistic atlas guided segmentation was developed to cope with the remaining variability in shape and location. Finally, a classifier ensemble was incorporated to refine the rough segmentation results. The effectiveness of the proposed algorithm was validated with 20 unknown CT volumes, as well as three on-site CT volumes distributed in a competition of pancreas segmentation algorithms. The experimental results indicated that the segmentation performance was enhanced by the proposed algorithm, and the Jaccard index between an extracted pancreas and a true one was 57.9%. This study verified the effectiveness of two-stage segmentation with spatial standardization of pancreas in delineating the pancreas region, patient-specific probabilistic atlas guided segmentation in reducing false negatives, and a classifier ensemble in boosting segmentation performance.

The clinical use of computed tomography (CT) scans has been steadily rising due to their ability to provide detailed three-dimensional (3D) representations of organs, along with advantages such as speed and high resolution. With CT, physicians can identify potential diseased organs and plan appropriate treatments. However, accurate quantitative evaluation is time-consuming, and precise automatic segmentation can significantly enhance the effectiveness of CT scans. Frequency linked attention with selective hierarchy (FLASH) was introduced as a novel deep learning architecture for multi-organ segmentation in abdominal CT scans. FLASH incorporated the 3D discrete wavelet transform into Transformer blocks as the backbone of the encoder. By distinguishing 3D frequency bands, the network addressed the challenge of unclear organ boundaries in abdominal CT scans caused by low contrast. Two-directional skip connections were introduced to link the multi-scale frequency features into the bottleneck and to assign adaptive weights to frequency components using attention mechanisms. By training from scratch on the organ dataset with five-fold validation, FLASH achieved the highest dice similarity coefficient = 0.826, normalized surface distance = 0.698 with the lowest standard deviation of dice similarity coefficient = 0.034, standard deviation of normalized surface distance = 0.046 compared to other networks in the experiments. With better segmentation results, FLASH had fewer parameters (96.86M vs. 139.45M) and computation time (36.3h vs. 50.4h) compared to Swin UNETR, which is more suitable in clinical use with limited resources and waiting time. Our code can be found athttps://github.com/jorden0721/FLASH-UNETR.git.

To the Editor: Despite the frequency and variety of foreign bodies ingested, secondary intestinal perforation is a rare occurrence. Preoperative diagnosis using computed tomography (CT) scan is rarely made because of the lack of radiographic contrast of nonmetallic foreign objects. A case of jejunal perforation by a fish bone is reported. Diagnosis was made using abdominal helical CT scan, and surgery was immediately performed. This clinical case allows us to discuss the several fish bone perforation sites and the patient precipitating factors. Coloenteric perforation by foreign bodies is a rare occurrence and therefore a difficult preoperative diagnosis. A case of jejunal perforation by a fish bone in an old man with bad dentition is reported. Diagnosis was made using computed tomography (CT) scan before operating. An 81-year-old male patient suffering from lower left quadrant abdominal tenderness associated with fever (>38.5°C) and elevated white blood cell count (15,000/mm3) was admitted to the emergency service. His medical history included several laparotomies for cholecystectomy, obstruction of the bowel caused by adhesions, and incisional hernia. Plain abdominal x-ray showed gas shadows in the lower left quadrant without pneumoperitoneum. The CT scan revealed jejunal perforation by a nonmetallic foreign body suggestive of a fish bone (Figure 1). The laparotomy revealed lower left quadrant peritonitis and jejunal perforation by the fish bone, as seen on the CT scan. Removal of the fish bone, suture of the jejunal perforation, debridement, and washing and drainage of the abdominal cavity were performed. The postoperative course was uneventful, and the patient was discharged 12 days after his surgery. The patient was well at his 1-year follow-up. He reported retrospectively that he had eaten fresh fish 2 days before the surgical intervention and did not feel any pain during his lunch. Abdominal computed tomography scan (frontal view) showing the fishbone (white arrow) perforating the small bowel and the macroscopic view of the fishbone. This case is particularly interesting because the peritonitis was discovered preoperatively using a CT scan. The patient's bad dentition certainly contributed to this complication. Ingestion of nonmetallic foreign bodies leading to intestinal perforation is a rare occurrence because most foreign bodies pass spontaneously per anus.1 Perforation by foreign bodies can occur anywhere in the gastrointestinal tract. Fish bones are more often responsible for perforation of the esophagus, complicated by mediastinitis.1 Other parts of the upper digestive tract, such as the stomach or the duodenum, may be perforated.2 Fish bones may also pass through the gastrointestinal tract and enter other organs of the abdominal cavity such as the liver and the pancreas. One case of a fish bone totally integrated into the hepatic parenchyma has been recently reported.3 The small bowel4 or the large bowel5,6 may be perforated too, and cases of Meckel's diverticulum7 perforation have been described. The clinical manifestations of gastrointestinal perforation by foreign bodies are variable. Symptoms may be acute, showing signs of peritonitis associated with pneumoperitoneum visible on plain abdominal x-ray,7 although the patient may present more-indolent symptoms attributable to carcinoma or chronic disease. One study reported a clinical case of an 80-year-old woman presenting a painless, hard epigastric mass on physical examination 8 days after fish ingestion.5 She received an antibiotic treatment for abdominal pain secondary to the initially unsuspected perforation of the large bowel by a fish bone. Diagnosis was made at laparotomy. In the case of late diagnosis, a fish bone located inside the duodenum and pancreas was found several months after ingestion.2 As with the current patient, elderly persons are more at risk because of the bad quality of their dentition, which is responsible for a lack of mastication.5,8 Some cases of intestinal perforation by fish bone discovered using helical CT scan have been previously described.1,4,6 Ileal or jejunal perforation by fish bone is a rare occurrence of gastroenteric perforation. Preoperative diagnosis may be made using abdominal helical CT scan in an emergency. It is necessary to carefully remove the bones from fish that elderly persons eat because of the lack of mastication and the bad quality of dentition. Financial Disclosure: None. Author Contributions: Brice Dugas: preparation of manuscript. Bénédicte Bernard: preparation of manuscript. Pierre-Jean Valette: acquisition of subjects and data and analysis and interpretation of data. Emmanuel Odet: preparation of manuscript. Alain Naouri: preparation of manuscript. Pierre Bernard: acquisition of subjects and data, analysis and interpretation of data, and preparation of manuscript. Sponsor's Role: none.

Deep neural networks have been widely used in medical image segmentation and they can achieve good results in segmentation of some big organs. However, for some small organs, such as the pancreas in 3D CT images, the segmentation results are usually not satisfactory due to the low proportion. In this paper, we present a novel network --- Dense Attentional Network (DA-Net), to improve the pancreas segmentation in abdominal CT scans. In DA-Net, dense connection is used to combine low-level features of encoder with corresponding features of decoder, which can help to improve the utilization of feature maps (FMs). In addition, a new module for recombination and recalibration of feature maps (RRFM) and a new attentional mechanism --- deep attentional features (DAF), are used, which can excite the most discriminating features. Pancreas is segmented from 3D CT images by a coarse-to-fine mode, in which pancreas is firstly located from 3D CT images by a coarse segmentation network, and then pancreas is further finely segmented by the DA-Net. We evaluate the proposed method with 129 CT images from NIH pancreas dataset and BTCV segmentation challenge, and compare it with several mainstream segmentation networks. Comparing with these networks, our DA-Net has the higher mean DSC of 81.39%. This shows the effectiveness and advantage of the proposed method.

Robust and efficient abdominal CT segmentation using shape constrained multi-scale attention network

The volume of pelvic hematoma at CT has been shown to be the strongest independent predictor of major arterial injury requiring angioembolization in trauma victims with pelvic fractures, and also correlates with transfusion requirement and mortality. Measurement of pelvic hematomas (unopacified extraperitoneal blood accumulated from time of injury) using semi-automated seeded region growing is time-consuming and requires trained experts, precluding routine measurement at the point of care. Pelvic hematomas are markedly variable in shape and location, have irregular ill-defined margins, have low contrast with respect to viscera and muscle, and reside within anatomically distorted pelvises. Furthermore, pelvic hematomas occupy a small proportion of the entire volume of a chest, abdomen, and pelvis (C/A/P) trauma CT. The challenges are many, and no automated methods for segmentation and volumetric analysis have been described to date. Traditional approaches using fully convolutional networks result in coarse segmentations and class imbalance with suboptimal convergence. In this study, we implement a modified coarse-to-fine deep learning approach-the Recurrent Saliency Transformation Network (RSTN) for pelvic hematoma volume segmentation. RSTN previously yielded excellent results in pancreas segmentation, where low contrast with adjacent structures, small target volume, variable location, and fine contours are also problematic. We have curated a unique single-institution corpus of 253 C/A/P admission trauma CT studies in patients with bleeding pelvic fractures with manually labeled pelvic hematomas. We hypothesized that RSTN would result in sufficiently high Dice similarity coefficients to facilitate accurate and objective volumetric measurements for outcome prediction (arterial injury requiring angioembolization). Cases were separated into five combinations of training and test sets in an 80/20 split and fivefold cross-validation was performed. Dice scores in the test set were 0.71 (SD ± 0.10) using RSTN, compared to 0.49 (SD ± 0.16) using a baseline Deep Learning Tool Kit (DLTK) reference 3D U-Net architecture. Mean inference segmentation time for RSTN was 0.90min (± 0.26). Pearson correlation between predicted and manual labels was 0.95 with p< 0.0001. Measurement bias was within 10mL. AUC of hematoma volumes for predicting need for angioembolization was 0.81 (predicted) versus 0.80 (manual). Qualitatively, predicted labels closely followed hematoma contours and avoided muscle and displaced viscera. Further work will involve validation using a federated dataset and incorporation into a predictive model using multiple segmented features.

Combinations of data augmentation methods and deep learning architectures for automatic pancreas segmentation on CT images are proposed and evaluated. Images from a public CT dataset of pancreas segmentation were used to evaluate the models. Baseline U-net and deep U-net were chosen for the deep learning models of pancreas segmentation. Methods of data augmentation included conventional methods, mixup, and random image cropping and patching (RICAP). Ten combinations of the deep learning models and the data augmentation methods were evaluated. Four-fold cross validation was performed to train and evaluate these models with data augmentation methods. The dice similarity coefficient (DSC) was calculated between automatic segmentation results and manually annotated labels and these were visually assessed by two radiologists. The performance of the deep U-net was better than that of the baseline U-net with mean DSC of 0.703–0.789 and 0.686–0.748, respectively. In both baseline U-net and deep U-net, the methods with data augmentation performed better than methods with no data augmentation, and mixup and RICAP were more useful than the conventional method. The best mean DSC was obtained using a combination of deep U-net, mixup, and RICAP, and the two radiologists scored the results from this model as good or perfect in 76 and 74 of the 82 cases.

The shape of pancreas in different patients is very different and the boundary is fuzzy, so the reliable automatic segmentation of pancreas is an important and difficult task. In this paper, we propose a multi-scale hybrid attention cascade network for challenging pancreas segmentation. First, the original CT images are coarsely segmented through the first-level network, then, the coarse segmentation results are clipped and sent to the second-level network for further training to obtain the fine segmentation results. The network adopts full convolutional network (FCN) integrating multi-scale mixed attention. The cut CT image eliminates the irrelevant background interference and reduces the input size of the secondary network, thus improving the segmentation accuracy. An extensive evaluation of 82 open data sets was performed by quadruple cross validation. The experimental results showed that the Dice coefficient was 84.62±4.20% compared with several advanced methods. In addition, this method has excellent performance in three measurement indexes of precision, Jaccard and recall.

Early detection and timely surgical intervention are crucial in reducing mortality rates associated with clinically significant prostate cancer (csPCa). Currently, clinical diagnostics primarily depend on magnetic resonance imaging (MRI) and nuclear medicine, with the potential diagnostic value of abdominal computed tomography (CT) remaining underexplored. This study aims to evaluate the effectiveness of multi-task deep learning neural networks in identifying early-stage prostate cancer using CT scans. In this study, we enrolled 539 patients from the Department of Radiology (N=461) and Nuclear Medicine (N=78). We utilized a multi-task deep learning network model (MTDL), based on the 3DUnet architecture, to segment and analyze the collected abdominal plain CT images. The predictive performance of this model was compared with a radiomics model and a single-task deep learning model using ResNet18. A diagnostic nomogram was then developed using the multi-task deep learning approach, incorporating prediction results and PSAD, age. The diagnostic performance of the different models was evaluated using the receiver operating characteristic (ROC) curve and the area under the curve (AUC). The 461 patients from the Department of Radiology were divided into training and test sets at a ratio of 6:4, while the patients from the Department of Nuclear Medicine formed the validation set. Our MTDL nomogram demonstrated AUCs of 0.941 (95% confidence interval [CI]: 0.905valceedi 0.912 (95% CI: 0.904valceedi and 0.932 (95% CI: 0.883valceed in the training, test, and validation cohorts, respectively. This study indicates that combining abdominal CT with a multi-task neural network model effectively diagnoses csPCa, offering superior diagnostic performance compared to clinical models. Additionally, the multi-task neural network model outperformed both the single-task neural network model and the radiomics model in diagnostic accuracy. Our study demonstrated that the MTDL nomogram can accurately predict the presence of prostate cancer using abdominal CT scans, offering significant value for the early diagnosis of prostate cancer.

Objective: Abdominal computed tomography has proven accurate for the detection of pediatric solid organ injuries following blunt abdominal trauma but is less reliable in detecting blunt bowel and mesenteric injuries (BBMI). The purpose of this study was to determine the significance of nonspecific findings on abdominal computed tomography (CT) scan in children at risk for BBMI. Design: Retrospective chart review. Setting: Regional pediatric trauma center. Patients: All patients who received an abdominal CT scan as part of their evaluation following blunt abdominal trauma over a 10-yr period (September 1991 to September 2001). Interventions: None. Measurements and Main Results: Individual records were reviewed by one of the authors and analyzed for age, gender, mechanism of injury, diagnostic studies and procedures, results of initial CT scan, treatment, surgical procedures, complications, and outcome. Patients were excluded if they suffered penetrating trauma, had an abdominal CT scan performed at another institution, had a surgical procedure before CT scan, or had incomplete records. A total of 2,114 patients met inclusion criteria. Sixty-five percent were male and ages ranged from 3 wks to 18 yrs. There were 178 patients who had at least one nonspecific finding on abdominal CT scan suggestive of BBMI; 151 patients had one finding, 24 patients had two findings, and three had three findings. The risk of BBMI increased with the number of nonspecific findings (positive predictive value = 11% with at least one finding and 44% with two or more findings). This increase in positive predictive value, however, was accompanied by a reduction in sensitivity (62% and 37%, respectively). A total of 32 patients had surgically proven BBMI. Of these, eight had a single nonspecific finding on CT scan, ten had two findings, and two had three findings (12 patients had no CT findings suggestive of BBMI). There were complications in four of the 32 patients with BBMI and one death (due to laceration of the superior mesenteric artery). The complications appeared to occur independent of the time to surgical intervention. Conclusions: The presence of multiple nonspecific findings on abdominal CT scan does not reliably predict BBMI in children. Children also appear to suffer complications from BBMI less frequently than adults, regardless of the time to surgery. Therefore, nonspecific findings alone do not warrant surgical exploration. The decision to operate should instead be based on clinical data that include serial physical examinations.

Automatic pancreas segmentation in abdominal CT images plays an important role in clinical applications. It can provide doctors with quantitative and qualitative information. Due to the small size, unclear edges, and the high anatomical differences between patients, it is a challenging task to accurately segment the pancreas with diseases. In this paper, we propose a new method to automatically segment the pancreas in abdominal CT images. First, we propose a contrast enhancement block. The block generates edge information and uses gating mechanism to enhance edge details of the pancreas. Second, we leverage a reverse attention block. This block utilizes the decoder feature map to guide the network to mine complementary discriminative regions. The proposed method is trained on 63 3D CT images, validated on 15 3D CT images, and tested on 28 3D CT images. Compared with manual segmentation, the mean Dice similarity coefficient can reach 86.11±8.02%. Experimental results show that our method can obtain more accurate segmentation results compared with existing segmentation methods.

In this study, we modified the previously proposed X2CT-GAN to build a 2Dto3D-GAN of the spine. This study also incorporated the radiologist’s perspective in the adjustment of input signals to prove the feasibility of the automatic production of three-dimensional (3D) structures of the spine from simulated bi-planar two-dimensional (2D) X-ray images. Data from 1012 computed tomography (CT) studies of 984 patients were retrospectively collected. We tested this model under different dataset sizes (333, 666, and 1012) with different bone signal conditions to observe the training performance. A 10-fold cross-validation and five metrics—Dice similarity coefficient (DSC) value, Jaccard similarity coefficient (JSC), overlap volume (OV), and structural similarity index (SSIM)—were applied for model evaluation. The optimal mean values for DSC, JSC, OV, SSIM_anteroposterior (AP), and SSIM_Lateral (Lat) were 0.8192, 0.6984, 0.8624, 0.9261, and 0.9242, respectively. There was a significant improvement in the training performance under empirically enhanced bone signal conditions and with increasing training dataset sizes. These results demonstrate the potential of the clinical implantation of GAN for automatic production of 3D spine images from 2D images. This prototype model can serve as a foundation in future studies applying transfer learning for the development of advanced medical diagnostic techniques.