DenseNet model incorporating hybrid attention mechanisms and clinical features for pancreatic cystic tumor classification.

PURPOSE: The aim of this study is to develop a deep learning model capable of discriminating between pancreatic plasma cystic neoplasms (SCN) and mucinous cystic neoplasms (MCN) by leveraging patient-specific clinical features and imaging outcomes. The intent is to offer valuable diagnostic support to clinicians in their clinical decision-making processes. METHODS: The construction of the deep learning model involved utilizing a dataset comprising abdominal magnetic resonance T2-weighted images obtained from patients diagnosed with pancreatic cystic tumors at Changhai Hospital. The dataset comprised 207 patients with SCN and 93 patients with MCN, encompassing a total of 1761 images. The foundational architecture employed was DenseNet-161, augmented with a hybrid attention mechanism module. This integration aimed to enhance the network's attentiveness toward channel and spatial features, thereby amplifying its performance. Additionally, clinical features were incorporated prior to the fully connected layer of the network to actively contribute to subsequent decision-making processes, thereby significantly augmenting the model's classification accuracy. The final patient classification outcomes were derived using a joint voting methodology, and the model underwent comprehensive evaluation. RESULTS: Using the five-fold cross validation, the accuracy of the classification model in this paper was 92.44%, with an AUC value of 0.971, a precision rate of 0.956, a recall rate of 0.919, a specificity of 0.933, and an F1-score of 0.936. CONCLUSION: This study demonstrates that the DenseNet model, which incorporates hybrid attention mechanisms and clinical features, is effective for distinguishing between SCN and MCN, and has potential application for the diagnosis of pancreatic cystic tumors in clinical practice.

Tracheal computed tomography radiomics model for prediction of the Omicron variant of severe acute respiratory syndrome coronavirus 2.

OBJECTIVES: The Omicron variant of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is highly contagious, fast-spreading, and insidious. Most patients present with normal findings on lung computed tomography (CT). The current study aimed to develop and validate a tracheal CT radiomics model to predict Omicron variant infection. MATERIALS AND METHODS: In this retrospective study, a radiomics model was developed based on a training set consisting of 157 patients with an Omicron variant infection and 239 healthy controls between 1 January and 30 April 2022. A set of morphological expansions, with dilations of 1, 3, 5, 7, and 9 voxels, was applied to the trachea, and radiomic features were extracted from different dilation voxels of the trachea. Logistic regression (LR), support vector machines (SVM), and random forests (RF) were developed and evaluated; the models were validated on 67 patients with the Omicron variant and on 103 healthy controls between 1 May and 30 July 2022. RESULTS: Logistic regression with 12 radiomic features extracted from the tracheal wall with dilation of 5 voxels achieved the highest classification performance compared with the other models. The LR model achieved an area under the curve of 0.993 (95% confidence interval [CI]: 0.987-0.998) in the training set and 0.989 (95% CI: 0.979-0.999) in the validation set. Sensitivity, specificity, and accuracy of the model for the training set were 0.994, 0.946, and 0.965, respectively, whereas those for the validation set were 0.970, 0.952, and 0.959, respectively. CONCLUSION: The tracheal CT radiomics model reliably identified the Omicron variant of SARS-CoV­2, and may help in clinical decision-making in future, especially in cases of normal lung CT findings.

Cross-Modal Consistency for Single-Modal MR Image Segmentation.

OBJECTIVE: Multi-modal magnetic resonance (MR) image segmentation is an important task in disease diagnosis and treatment, but it is usually difficult to obtain multiple modalities for a single patient in clinical applications. To address these issues, a cross-modal consistency framework is proposed for a single-modal MR image segmentation. METHODS: To enable single-modal MR image segmentation in the inference stage, a weighted cross-entropy loss and a pixel-level feature consistency loss are proposed to train the target network with the guidance of the teacher network and the auxiliary network. To fuse dual-modal MR images in the training stage, the cross-modal consistency is measured according to Dice similarity entropy loss and Dice similarity contrastive loss, so as to maximize the prediction similarity of the teacher network and the auxiliary network. To reduce the difference in image contrast between different MR images for the same organs, a contrast alignment network is proposed to align input images with different contrasts to reference images with a good contrast. RESULTS: Comprehensive experiments have been performed on a publicly available prostate dataset and an in-house pancreas dataset to verify the effectiveness of the proposed method. Compared to state-of-the-art methods, the proposed method can achieve better segmentation. CONCLUSION: The proposed image segmentation method can fuse dual-modal MR images in the training stage and only need one-modal MR images in the inference stage. SIGNIFICANCE: The proposed method can be used in routine clinical occasions when only single-modal MR image with variable contrast is available for a patient.

Smoking Behavior and its Effect on Psychiatric Symptoms of Patients with Stable Schizophrenia.

Context: Schizophrenia is a common and clinically disabling mental disorder. Many patients with schizophrenia smoke. Research on the effects of smoking on schizophrenia's symptoms are inconsistent. Objective: The study intended to investigate the smoking status of patients with stable schizophrenia to determine the effects of smoking on schizophrenia-related symptoms. Design: The research team performed an case-control study. Setting: The study took place at Beijing Huilongguan Hospital in Beijing, Changping District, China. Participants: Participants were 160 patients at the hospital who had been diagnosed with stable schizophrenia between April 2018 and March 2020. Groups: The research team divided participants into two groups based on their current smoking status: (1) a smoking group with 72 participants and (2) a nonsmoking group with 88 participants. Outcome Measures: The research team: (1) examined the types of antipsychotic drugs that participants received; (2) used a schizophrenia-related scale, the Positive and Negative Syndrome Scale (PANSS), to examine participants' status; (3) examined the smoking habits of the smoking group; and (4) analyzed the correlation between the PANSS score and the smoking group's smoking index. Results: No significant difference existed between the groups in the type of medicine used (P > .05). The smoking group's PANSS total (P = .014), positive symptom (P = .039), and negative symptom (P = .003) scores were significantly lower than those of the nonsmoking group (P < .05). No significant difference existed between the groups in the general psychopathological symptom score (P > .05). The smoking group started smoking between 13 and 24 years of age, with an mean age of 19.11 ± 4.10 years. The group smoked 10-30 cigarettes/d, with a mean smoking amount of 18.4 ± 3.1 cigarettes/d, and the smoking index was 344.7 ± 48.0. The smoking group's smoking index was significantly negatively correlated with the positive symptom, negative symptom, and total PANSS scores (all P = .000). No correlation existed between the smoking index and the general psychopathological symptom score (P > .05). Conclusions: Smoking patients with stable schizophrenia generally exhibit fewer symptoms than nonsmoking patients, which relate to the alleviation of mental tension that smoking can provide.

D-A-D type based NIR fluorescence probe for monitoring the cysteine levels in pancreatic cancer cell during ferroptosis.

Cysteine (Cys) as a crucial precursor for intracellular glutathione (GSH) synthesis, plays an important role in the redox regulation in ferroptosis, Therefore, evaluating intracellular Cys levels is worthy to better understand ferroptosis-related physiological process. In this work, we constructed a novel NIR coumarin-derived fluorescent probe (NCDFP-Cys) based on a dual-ICT system, the NCDFP-Cys can show fluorescence turn-on response at 717 nm toward Cys over other amino acids, and possess large Stokes shift (Δλ = 167 nm), low detection limit, hypotoxicity. More significantly, NCDFP-Cys has been utilized to monitor the intracellular Cys fluctuation in pancreatic cancer cells during ferroptosis induced by Erastin and RSL3 respectively, and revealing the difference of Cys levels changes in different activator-triggered ferroptosis pathways.

Automated pancreatic segmentation and fat fraction evaluation based on a self-supervised transfer learning network.

Accurate segmentation of the pancreas from abdominal computed tomography (CT) images is challenging but essential for the diagnosis and treatment of pancreatic disorders such as tumours and diabetes. In this study, a dataset with 229 sets of high-resolution CT images was generated and annotated. We proposed a novel 3D segmentation model named nnTransfer (nonisomorphic transfer learning) net, which employs generative model structure for self-supervision to facilitate the network's learning of image attributes from unlabelled data. The effectiveness for pancreas segmentation of nnTransfer was assessed using the Hausdorff distance (HD) and Dice similarity coefficient (DSC) on the dataset. Additionally, a histogram analysis with local thresholding was used to achieve automated whole-volume measurement of pancreatic fat (fat volume fraction, FVF). The proposed technique performed admirably on the dataset, with DSC: 0.937 ± 0.019 and HD: 2.655 ± 1.479. The mean pancreas volume and FVF of the pancreas were 91.95 ± 23.90 cm3 and 12.67 % ± 9.84 %, respectively. The nnTransfer functioned flawlessly and autonomously, facilitating the use of the FVF to evaluate pancreatic disease, particularly in patients with diabetes.

Application of a pyramid pooling Unet model with integrated attention mechanism and Inception module in pancreatic tumor segmentation.

BACKGROUND: The segmentation and recognition of pancreatic tumors are crucial tasks in the diagnosis and treatment of pancreatic diseases. However, due to the relatively small proportion of the pancreas in the abdomen and significant shape and size variations, pancreatic tumor segmentation poses considerable challenges. PURPOSE: To construct a network model that combines a pyramid pooling module with Inception architecture and SE attention mechanism (PIS-Unet), and observe its effectiveness in pancreatic tumor images segmentation, thereby providing supportive recommendations for clinical practitioners. MATERIALS AND METHODS: A total of 303 patients with histologically confirmed pancreatic cystic neoplasm (PCN), including serous cystic neoplasm (SCN) and mucinous cystic neoplasm (MCN), from Shanghai Changhai Hospital between March 2011 and November 2021 were included. A total of 1792 T2-weighted imaging (T2WI) slices were used to build a CNN model. The model employed a pyramid pooling Inception module with a fused attention mechanism. The attention mechanism enhanced the network's focus on local features, while the Inception module and pyramid pooling allowed the network to extract features at different scales and improve the utilization efficiency of global information, thereby effectively enhancing network performance. RESULTS: Using three-fold cross-validation, the model constructed by us achieved a dice score of 85.49 ± 2.02 for SCN images segmentation, and a dice score of 87.90 ± 4.19 for MCN images segmentation. CONCLUSION: This study demonstrates that using deep learning networks for the segmentation of PCNs yields favorable results. Applying this network as an aid to physicians in PCN diagnosis shows potential for clinical applications.

Variabilities in apparent diffusion coefficient (ADC) measurements of the spleen and the paraspinal muscle: A single center large cohort study.

Purpose: Evaluation of the variabilities in apparent diffusion coefficient (ADC) measurements of the spleen (ADCspleen) and the paraspinal muscles (ADCmuscle) to identify the reference organ for normalizing the ADC from the abdominal diffusion weighted imaging (DWI). Methods: Two MRI scanners, with 314 abdominal exams on the GE and 929 on the Siemens system, were used for MRI examinations including DWI (b-values, 50 and 800 s/mm2). For a subset of 73 exams on the Siemens system a second exam was conducted. Four regions of interest (ROIs) in each exam were placed to measure the ADCspleen and the bilateral ADCmuscle. ADC variability between patients (on each scanner separately), ADC variability due to ROI placement between the two ROIs in each organ, and variability in the subset between the first and second exams were assessed. Results: The ADCspleen was more scattered and variable than the ADCmuscle in the comparability (n = 929 and 314 for two MRI scanners, respectively) and repeatability (n = 73) datasets. The Bland-Altmann bias and limits of agreement (LoAs) for the ADCspleen (ICC, 0.47; CV, 0.070) and ADCmuscle (ICC, 0.67; CV, 0.023) in the repeatability datasets (n = 73) were -0.1 (-25.7%-25.6%) and -0.3 (-8.8%-8.1%), respectively. For the Siemens system, the Bland-Altmann bias and LoAs for the ADCspleen (ICC, 0.72; CV, 0.061) and ADCmuscle (ICC, 0.53; CV, 0.030) in the comparability datasets (n = 929) were 2.1 (-20.0%-24.2%) and 0.7 (-10.0%-11.4%), respectively. Similar findings have been found in the GE system (n = 314). The CVs for the ADCmuscle measurements were lower than those of the ADCspleen both in the repeatability and the comparability analyses (all p < 0.001). Conclusion: Paraspinal muscles demonstrate better reference characteristics than the spleen in estimating ADC variability of abdominal DWI.

Replacing secretin-enhanced MRCP with MRI radiomics model based on a fully automated pancreas segmentation for assessing pancreatic exocrine function in chronic pancreatitis.

OBJECTIVES: To develop and validate a radiomics nomogram based on a fully automated pancreas segmentation to assess pancreatic exocrine function. Furthermore, we aimed to compare the performance of the radiomics nomogram with the pancreatic flow output rate (PFR) and conclude on the replacement of secretin-enhanced magnetic resonance cholangiopancreatography (S-MRCP) by the radiomics nomogram for pancreatic exocrine function assessment. METHODS: All participants underwent S-MRCP between April 2011 and December 2014 in this retrospective study. PFR was quantified using S-MRCP. Participants were divided into normal and pancreatic exocrine insufficiency (PEI) groups using the cut-off of 200 µg/L of fecal elastase-1. Two prediction models were developed including the clinical and non-enhanced T1-weighted imaging radiomics model. A multivariate logistic regression analysis was conducted to develop the prediction models. The models' performances were determined based on their discrimination, calibration, and clinical utility. RESULTS: A total of 159 participants (mean age [Formula: see text] standard deviation, 45 years [Formula: see text] 14;119 men) included 85 normal and 74 PEI. All the participants were divided into a training set comprising 119 consecutive patients and an independent validation set comprising 40 consecutive patients. The radiomics score was an independent risk factor for PEI (odds ratio = 11.69; p < 0.001). In the validation set, the radiomics nomogram exhibited the highest performance (AUC, 0.92) in PEI prediction, whereas the clinical nomogram and PFR had AUCs of 0.79 and 0.78, respectively. CONCLUSION: The radiomics nomogram accurately predicted pancreatic exocrine function and outperformed pancreatic flow output rate on S-MRCP in patients with chronic pancreatitis. KEY POINTS: • The clinical nomogram exhibited moderate performance in diagnosing pancreatic exocrine insufficiency. • The radiomics score was an independent risk factor for pancreatic exocrine insufficiency, and every point rise in the rad-score was associated with an 11.69-fold increase in pancreatic exocrine insufficiency risk. • The radiomics nomogram accurately predicted pancreatic exocrine function and outperformed the clinical model and pancreatic flow output rate quantified by secretin-enhanced magnetic resonance cholangiopancreatography on MRI in patients with chronic pancreatitis.

Effects of Adjunctive Betahistine Therapy on Lipid Metabolism in Patients with Chronic Schizophrenia: A Randomized Double-Blind Placebo-Controlled Study.

Objective: This study aims to explore the ability of betahistine to inhibit weight gain and abnormal lipid metabolism in patients with chronic schizophrenia. Methods: A comparison study of betahistine or placebo therapy was conducted for 4 weeks in 94 patients with chronic schizophrenia, who were randomly divided into two groups. Clinical information and lipid metabolic parameters were collected. Positive and Negative Syndrome Scale (PANSS) was used to assess psychiatric symptoms. Treatment Emergent Symptom Scale (TESS) was used to evaluate treatment-related adverse reactions. The differences in lipid metabolic parameters before and after treatment between the two groups were compared. Results: Repeated measures analysis of variance (ANOVA) revealed that after 4 weeks of betahistine/placebo treatment, the interaction effect of time and group was statistically significant on low-density lipoprotein cholesterol (F = 6.453, p = 0.013) and waist-to-hip ratio (F = 4.473, p = 0.037), but did not reveal any significant interaction effect of time and group on weight, body mass index or other lipid metabolic parameters, as well as the time main effect and group main effect (all p > 0.05). Betahistine had no significant impact on PANSS, and no side effects related to betahistine were detected. Conclusion: Betahistine may delay metabolic abnormalities in patients with chronic schizophrenia. It does not affect the efficacy of the original antipsychotics. Thus, it provides new ideas for the treatment of metabolic syndrome in patients with chronic schizophrenia.

Fully automated magnetic resonance imaging-based radiomics analysis for differentiating pancreatic adenosquamous carcinoma from pancreatic ductal adenocarcinoma.

PURPOSE: To develop and validate an automated magnetic resonance imaging (MRI)-based model to preoperatively differentiate pancreatic adenosquamous carcinoma (PASC) from pancreatic ductal adenocarcinoma (PDAC). METHODS: This retrospective study included patients with surgically resected, histopathologically confirmed PASC or PDAC who underwent MRI between January 2011 and December 2020. According to time of treatment, they were divided into training and validation sets. Automated deep-learning-based artificial intelligence was used for pancreatic tumor segmentation. Linear discriminant analysis was performed with conventional MRI and radiomic features to develop clinical, radiomics, and mixed models in the training set. The models' performances were determined from their discrimination and clinical utility. Kaplan-Meier and log-rank tests were used for survival analysis. RESULTS: Overall, 389 and 123 patients with PDAC (age, 61.37 ± 9.47 years; 251 men) and PASC (age, 61.99 ± 9.82 years; 78 men) were included, respectively; they were split into the training (n = 358) and validation (n = 154) sets. The mixed model showed good performance in the training and validation sets (area under the curve: 0.94 and 0.96, respectively). The sensitivity, specificity, and accuracy were 76.74%, 93.38%, and 89.39% for the training set, respectively, and 67.57%, 97.44%, and 90.26% for the validation set, respectively. The mixed model outperformed the clinical (p = 0.001) and radiomics (p = 0.04) models in the validation set. Log-rank test revealed significantly longer survival in the predicted PDAC group than in the predicted PASC group (p = 0.003), according to the mixed model. CONCLUSION: Our mixed model, which combined MRI and radiomic features, can be used to differentiate PASC from PDAC.

Non-enhanced magnetic resonance imaging-based radiomics model for the differentiation of pancreatic adenosquamous carcinoma from pancreatic ductal adenocarcinoma.

Purpose: To evaluate the diagnostic performance of radiomics model based on fully automatic segmentation of pancreatic tumors from non-enhanced magnetic resonance imaging (MRI) for differentiating pancreatic adenosquamous carcinoma (PASC) from pancreatic ductal adenocarcinoma (PDAC). Materials and methods: In this retrospective study, patients with surgically resected histopathologically confirmed PASC and PDAC who underwent MRI scans between January 2011 and December 2020 were included in the study. Multivariable logistic regression analysis was conducted to develop a clinical and radiomics model based on non-enhanced T1-weighted and T2-weighted images. The model performances were determined based on their discrimination and clinical utility. Kaplan-Meier and log-rank tests were used for survival analysis. Results: A total of 510 consecutive patients including 387 patients (age: 61 ± 9 years; range: 28-86 years; 250 males) with PDAC and 123 patients (age: 62 ± 10 years; range: 36-84 years; 78 males) with PASC were included in the study. All patients were split into training (n=382) and validation (n=128) sets according to time. The radiomics model showed good discrimination in the validation (AUC, 0.87) set and outperformed the MRI model (validation set AUC, 0.80) and the ring-enhancement (validation set AUC, 0.74). Conclusions: The radiomics model based on non-enhanced MRI outperformed the MRI model and ring-enhancement to differentiate PASC from PDAC; it can, thus, provide important information for decision-making towards precise management and treatment of PASC.

Tumor size measurements of pancreatic cancer with neoadjuvant therapy based on RECIST guidelines: is MRI as effective as CT?

OBJECTIVES: To compare tumor size measurements using CT and MRI in pancreatic cancer (PC) patients with neoadjuvant therapy (NAT). METHODS: This study included 125 histologically confirmed PC patients who underwent NAT. The tumor sizes from CT and MRI before and after NAT were compared by using Bland-Altman analyses and intraclass correlation coefficients (ICCs). Variations in tumor size estimates between MRI and CT in relationship to different factors, including NAT methods (chemotherapy, chemoradiotherapy), tumor locations (head/neck, body/tail), tumor regression grade (TRG) levels (0-2, 3), N stages (N0, N1/N2) and tumor resection margin status (R0, R1), were further analysed. The McNemar test was used to compare the efficacy of NAT evaluations based on the CT and MRI measurements according to RECIST 1.1 criteria. RESULTS: There was no significant difference between the median tumor sizes from CT and MRI before and after NAT (P = 0.44 and 0.39, respectively). There was excellent agreement in tumor size between MRI and CT, with mean size differences and limits of agreement (LOAs) of 1.5 [-9.6 to 12.7] mm and 0.9 [-12.6 to 14.5] mm before NAT (ICC, 0.93) and after NAT (ICC, 0.91), respectively. For all the investigated factors, there was good or excellent correlation (ICC, 0.76 to 0.95) for tumor sizes between CT and MRI. There was no significant difference in the efficacy evaluation of NAT between CT and MRI measurements (P = 1.0). CONCLUSION: MRI and CT have similar performance in assessing PC tumor size before and after NAT.

Diffusion-Weighted Imaging of the Abdomen: Correction for Gradient Nonlinearity Bias in Apparent Diffusion Coefficient.

BACKGROUND: Gradient nonlinearity (GNL) introduces spatial nonuniformity bias in apparent diffusion coefficient (ADC) measurements, especially at large offsets from the magnet isocenter. PURPOSE: To investigate the effects of GNL in abdominal ADC measurements and to develop an ADC bias correction procedure. STUDY TYPE: Retrospective. PHANTOM/POPULATION: Two homemade ultrapure water phantoms/25 patients with histologically confirmed pancreatic ductal adenocarcinoma (PDAC). FIELD STRENGTH/SEQUENCE: A 3.0 T/diffusion-weighted imaging (DWI) with single-shot echo-planar imaging sequence. ASSESSMENT: ADC bias was computed in the three orthogonal directions at different offset locations. The spatial-dependent correctors of ADC bias were generated from the ADCs of phantom 1. The ADCs were estimated before and after corrections for the phantom 1 with both the proposed approach and the theoretical GNL correction method. For the patients, ADCs were measured in abdominal tissues including left and right liver lobes, PDAC, spleen, bilateral kidneys, and bilateral paraspinal muscles. STATISTICAL TEST: Friedman tests and Wilcoxon tests. RESULTS: The ADC bias measured by phantom 1 was 9.7% and 12.6% higher in the right-left and anterior-posterior directions and 9.2% lower in the superior-inferior direction at the 150 mm offsets from the magnetic isocenter. The corrected vs. the uncorrected ADCs measurements (median: 2.20 × 10-3  mm2 /sec for both the proposed method and the theoretical GNL method vs. 2.31 × 10-3  mm2 /sec, respectively) and their relative ADC errors (0.014, 0.016, and 0.054, respectively) were lower in the phantom 1. The relative ADC errors substantially decreased after correction in the phantom 2 (median: 0.048 and -0.008, respectively). The ADCs of all the abdominal tissues were lower after correction except for the left liver lobes (P = 0.13). DATA CONCLUSION: GNL bias in abdominal ADC can be measured by a DWI phantom. The proposed correction procedure was successfully applied for the bias correction in abdominal ADC. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 1.

Artificial Intelligence to Predict Lymph Node Metastasis at CT in Pancreatic Ductal Adenocarcinoma.

Background Although deep learning has brought revolutionary changes in health care, reliance on manually selected cross-sectional images and segmentation remain methodological barriers. Purpose To develop and validate an automated preoperative artificial intelligence (AI) algorithm for tumor and lymph node (LN) segmentation with CT imaging for prediction of LN metastasis in patients with pancreatic ductal adenocarcinoma (PDAC). Materials and Methods In this retrospective study, patients with surgically resected, pathologically confirmed PDAC underwent multidetector CT from January 2015 to April 2020. Three models were developed, including an AI model, a clinical model, and a radiomics model. CT-determined LN metastasis was diagnosed by radiologists. Multivariable logistic regression analysis was conducted to develop the clinical and radiomics models. The performance of the models was determined on the basis of their discrimination and clinical utility. Kaplan-Meier curves, the log-rank test, or Cox regression were used for survival analysis. Results Overall, 734 patients (mean age, 62 years ± 9 [SD]; 453 men) were evaluated. All patients were split into training (n = 545) and validation (n = 189) sets. Patients who had LN metastasis (LN-positive group) accounted for 340 of 734 (46%) patients. In the training set, the AI model showed the highest performance (area under the receiver operating characteristic curve [AUC], 0.91) in the prediction of LN metastasis, whereas the radiologists and the clinical and radiomics models had AUCs of 0.58, 0.76, and 0.71, respectively. In the validation set, the AI model showed the highest performance (AUC, 0.92) in the prediction of LN metastasis, whereas the radiologists and the clinical and radiomics models had AUCs of 0.65, 0.77, and 0.68, respectively (P < .001). AI model-predicted positive LN metastasis was associated with worse survival (hazard ratio, 1.46; 95% CI: 1.13, 1.89; P = .004). Conclusion An artificial intelligence model outperformed radiologists and clinical and radiomics models for prediction of lymph node metastasis at CT in patients with pancreatic ductal adenocarcinoma. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Chu and Fishman in this issue.

Semi-supervised interactive fusion network for MR image segmentation.

BACKGROUND: Medical image segmentation is an important task in the diagnosis and treatment of cancers. The low contrast and highly flexible anatomical structure make it challenging to accurately segment the organs or lesions. PURPOSE: To improve the segmentation accuracy of the organs or lesions in magnetic resonance (MR) images, which can be useful in clinical diagnosis and treatment of cancers. METHODS: First, a selective feature interaction (SFI) module is designed to selectively extract the similar features of the sequence images based on the similarity interaction. Second, a multi-scale guided feature reconstruction (MGFR) module is designed to reconstruct low-level semantic features and focus on small targets and the edges of the pancreas. Third, to reduce manual annotation of large amounts of data, a semi-supervised training method is also proposed. Uncertainty estimation is used to further improve the segmentation accuracy. RESULTS: Three hundred ninety-five 3D MR images from 395 patients with pancreatic cancer, 259 3D MR images from 259 patients with brain tumors, and four-fold cross-validation strategy are used to evaluate the proposed method. Compared to state-of-the-art deep learning segmentation networks, the proposed method can achieve better segmentation of pancreas or tumors in MR images. CONCLUSIONS: SFI-Net can fuse dual sequence MR images for abnormal pancreas or tumor segmentation. The proposed semi-supervised strategy can further improve the performance of SFI-Net.

Automatic multi-tissue segmentation in pancreatic pathological images with selected multi-scale attention network.

The morphology of tissues in pathological images has been used routinely by pathologists to assess the degree of malignancy of pancreatic ductal adenocarcinoma (PDAC). Automatic and accurate segmentation of tumor cells and their surrounding tissues is often a crucial step to obtain reliable morphological statistics. Nonetheless, it is still a challenge due to the great variation of appearance and morphology. In this paper, a selected multi-scale attention network (SMANet) is proposed to segment tumor cells, blood vessels, nerves, islets and ducts in pancreatic pathological images. The selected multi-scale attention module is proposed to enhance effective information, supplement useful information and suppress redundant information at different scales from the encoder and decoder. It includes selection unit (SU) module and multi-scale attention (MA) module. The selection unit module can effectively filter features. The multi-scale attention module enhances effective information through spatial attention and channel attention, and combines different level features to supplement useful information. This helps learn the information of different receptive fields to improve the segmentation of tumor cells, blood vessels and nerves. An original-feature fusion unit is also proposed to supplement the original image information to reduce the under-segmentation of small tissues such as islets and ducts. The proposed method outperforms state-of-the-arts deep learning algorithms on our PDAC pathological images and achieves competitive results on the GlaS challenge dataset. The mDice and mIoU have reached 0.769 and 0.665 in our PDAC dataset.

Distinguishing pancreatic solid serous cystadenomas from nonfunctional pancreatic neuroendocrine tumors by computed tomography: A propensity score analysis.

This study aims to evaluate the utility of calculated computed tomography (CT) attenuation value ratio (AVR) and enhancement pattern in distinguishing pancreatic solid serous cystadenomas (SCAs) from nonfunctional pancreatic neuroendocrine tumors (NF-pNETs). A total of 142 consecutive patients with 22 solid SCAs and 120 NF-pNETs confirmed by pathology were included in this retrospective study. All patients underwent preoperative contrast-enhanced CT and were categorized into 2 groups, solid SCA and NF-pNET groups. Patients with NF-pNETs were matched to patients with solid SCAs via propensity scores. AVR was measured and defined as: attenuation value of tumor/attenuation value of normal pancreas. AVR and enhancement pattern performance were assessed according to the discriminative abilities of patients. After matching, 29 patients were allocated to the NF-pNET group. Before matching, sex, age, and the peak enhanced value phase were significantly different between solid SCA and NF-pNET patients (P < .05). After matching, no significant difference was observed between both groups (P > .05). Solid SCAs AVRs were significantly smaller than NF-pNETs AVRs in all unenhanced, arterial, portal venous, and delayed phases (P < .05). Solid SCAs showed significantly more wash-in and wash-out enhancement patterns than NF-pNETs (P < .05). For unenhanced, arterial, portal venous, and delayed phases, and enhancement pattern, the area under the curve (AUC) values were 0.96, 0.72, 0.80, 0.85, and 0.86, respectively. Low AVR on unenhanced CT and wash-in and wash-out enhancement patterns were useful for differentiating solid SCAs from NF-pNETs and may be useful for clinical decisions, a clearer opinion will be formed with further studies to be conducted with larger patient numbers.