Radiomics for the Diagnosis and Differentiation of Pancreatic Cystic Lesions

Machine Learning-Driven Radiomics Analysis for Distinguishing Mucinous and Non-Mucinous Pancreatic Cystic Lesions: A Multicentric Study

The increasing use of high-resolution cross-sectional imaging has significantly enhanced the detection of pancreatic cystic lesions (PCLs), including pseudocysts and neoplastic entities such as IPMN, MCN, and SCN. However, accurate categorization of PCLs remains a challenge. This study aims to improve PCL evaluation by developing and validating a radiomics-based software tool leveraging machine learning (ML) for lesion classification. The model categorizes PCLs into mucinous and non-mucinous types using a custom dataset of 261 CT examinations, with 156 images for training and 105 for external validation. Three experienced radiologists manually delineated the images, extracting 38 radiological and 214 radiomic features using the Pyradiomics module in Python 3.13.2. Feature selection was performed using Least Absolute Shrinkage and Selection Operator (LASSO) regression, followed by classification with an Adaptive Boosting (AdaBoost) model trained on the optimized feature set. The proposed model achieved an accuracy of 89.3% in the internal validation cohort and demonstrated robust performance in the external validation cohort, with 90.2% sensitivity, 80% specificity, and 88.2% overall accuracy. Comparative analysis with existing radiomics-based studies showed that the proposed model either outperforms or performs on par with the current state-of-the-art methods, particularly in external validation scenarios. These findings highlight the potential of radiomics-driven machine learning approaches in enhancing PCL diagnosis across diverse patient populations.

Radiomics in early detection of bilio-pancreatic lesions: A narrative review

Radiomics is transforming the field of early detection of bilio-pancreatic lesions, offering significant advancements in diagnostic accuracy and personalized treatment planning. By extracting high-dimensional data from medical images such as computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET), radiomics reveals complex patterns that remain undetectable through traditional imaging evaluation. This review synthesizes recent developments in radiomics, particularly its application to early detection of pancreatic cancer (PC) and biliary duct cancer (BDC). It highlights the role of machine learning algorithms and multi-parametric models in improving diagnostic performance and discusses challenges such as standardization, reproducibility, and the need for larger, multicenter datasets. The integration of radiomics with genomic data and liquid biopsies also presents future opportunities for more individualized patient care.

Radiomics workflow definition & challenges - German priority program 2177 consensus statement on clinically applied radiomics

OBJECTIVES

Achieving a consensus on a definition for different aspects of radiomics workflows to support their translation into clinical usage. Furthermore, to assess the perspective of experts on important challenges for a successful clinical workflow implementation.

MATERIALS AND METHODS

The consensus was achieved by a multi-stage process. Stage 1 comprised a definition screening, a retrospective analysis with semantic mapping of terms found in 22 workflow definitions, and the compilation of an initial baseline definition. Stages 2 and 3 consisted of a Delphi process with over 45 experts hailing from sites participating in the German Research Foundation (DFG) Priority Program 2177. Stage 2 aimed to achieve a broad consensus for a definition proposal, while stage 3 identified the importance of translational challenges.

RESULTS

Workflow definitions from 22 publications (published 2012-2020) were analyzed. Sixty-nine definition terms were extracted, mapped, and semantic ambiguities (e.g., homonymous and synonymous terms) were identified and resolved. The consensus definition was developed via a Delphi process. The final definition comprising seven phases and 37 aspects reached a high overall consensus (> 89% of experts "agree" or "strongly agree"). Two aspects reached no strong consensus. In addition, the Delphi process identified and characterized from the participating experts' perspective the ten most important challenges in radiomics workflows.

CONCLUSION

To overcome semantic inconsistencies between existing definitions and offer a well-defined, broad, referenceable terminology, a consensus workflow definition for radiomics-based setups and a terms mapping to existing literature was compiled. Moreover, the most relevant challenges towards clinical application were characterized.

CRITICAL RELEVANCE STATEMENT

Lack of standardization represents one major obstacle to successful clinical translation of radiomics. Here, we report a consensus workflow definition on different aspects of radiomics studies and highlight important challenges to advance the clinical adoption of radiomics.

KEY POINTS

Published radiomics workflow terminologies are inconsistent, hindering standardization and translation. A consensus radiomics workflow definition proposal with high agreement was developed. Publicly available result resources for further exploitation by the scientific community.

Risk Assessment and Radiomics Analysis in Magnetic Resonance Imaging of Pancreatic Intraductal Papillary Mucinous Neoplasms (IPMN)

Intraductal papillary mucinous neoplasms (IPMNs) are a very common incidental finding during patient radiological assessment. These lesions may progress from low-grade dysplasia (LGD) to high-grade dysplasia (HGD) and even pancreatic cancer. The IPMN progression risk grows with time, so discontinuation of surveillance is not recommended. It is very important to identify imaging features that suggest LGD of IPMNs, and thus, distinguish lesions that only require careful surveillance from those that need surgical resection. It is important to know the management guidelines and especially the indications for surgery, to be able to point out in the report the findings that suggest malignant degeneration. The imaging tools employed for diagnosis and risk assessment are Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) with contrast medium. According to the latest European guidelines, MRI is the method of choice for the diagnosis and follow-up of patients with IPMN since this tool has a highest sensitivity in detecting mural nodules and intra-cystic septa. It plays a key role in the diagnosis of worrisome features and high-risk stigmata, which are associated with IPMNs malignant degeneration. Nowadays, the main limit of diagnostic tools is the ability to identify the precursor of pancreatic cancer. In this context, increasing attention is being given to artificial intelligence (AI) and radiomics analysis. However, these tools remain in an exploratory phase, considering the limitations of currently published studies. Key limits include noncompliance with AI best practices, radiomics workflow standardization, and clear reporting of study methodology, including segmentation and data balancing. In the radiological report it is useful to note the type of IPMN so as the morphological features, size, rate growth, wall, septa and mural nodules, on which the indications for surveillance and surgery are based. These features should be reported so as the surveillance time should be suggested according to guidelines.

Radiomics Based on Contrast-Enhanced Ultrasound Images for Diagnosis of Pancreatic Serous Cystadenoma

OBJECTIVE

The purpose of the study was to develop and validate a radiomics model by using contrast-enhanced ultrasound (CEUS) data for pre-operative differential diagnosis of pancreatic cystic neoplasms (PCNs), especially pancreatic serous cystadenoma (SCA).

METHODS

Patients with pathologically confirmed PCNs who underwent CEUS examination at Chinese PLA hospital from May 2015 to August 2022 were retrospectively collected. Radiomic features were extracted from the regions of interest, which were obtained based on CEUS images. A support vector machine algorithm was used to construct a radiomics model. Moreover, based on the CEUS image features, the CEUS and the combined models were constructed using logistic regression. The performance and clinical utility of the optimal model were evaluated by area under the receiver operating characteristic curve (AUC), sensitivity, specificity and decision curve analysis.

RESULTS

A total of 113 patients were randomly split into the training (n = 79) and test cohorts (n = 34). These patients were pathologically diagnosed with SCA, mucinous cystadenoma, intraductal papillary mucinous neoplasm and solid-pseudopapillary tumor. The radiomics model achieved an AUC of 0.875 and 0.862 in the training and test cohorts, respectively. The sensitivity and specificity of the radiomics model were 81.5% and 86.5% in the training cohort and 81.8% and 91.3% in the test cohort, respectively, which were higher than or comparable with that of the CEUS model and the combined model.

CONCLUSION

The radiomics model based on CEUS images had a favorable differential diagnostic performance in distinguishing SCA from other PCNs, which may be beneficial for the exploration of personalized management strategies.

Current state of radiomics in hepatobiliary and pancreatic malignancies

Rising in incidence, hepatobiliary and pancreatic (HPB) cancers continue to exhibit dismal long-term survival. The overall poor prognosis of HPB cancers is reflective of the advanced stage at which most patients are diagnosed. Late diagnosis is driven by the often-asymptomatic nature of these diseases, as well as a dearth of screening modalities. Additionally, standard imaging modalities fall short of providing accurate and detailed information regarding specific tumor characteristics, which can better inform surgical planning and sequencing of systemic therapy. Therefore, precise therapeutic planning must be delayed until histopathological examination is performed at the time of resection. Given the current shortcomings in the management of HPB cancers, investigations of numerous noninvasive biomarkers, including circulating tumor cells and DNA, proteomics, immunolomics, and radiomics, are underway. Radiomics encompasses the extraction and analysis of quantitative imaging features. Along with summarizing the general framework of radiomics, this review synthesizes the state of radiomics in HPB cancers, outlining its role in various aspects of management, present limitations, and future applications for clinical integration. Current literature underscores the utility of radiomics in early detection, tumor characterization, therapeutic selection, and prognostication for HPB cancers. Seeing as single-center, small studies constitute the majority of radiomics literature, there is considerable heterogeneity with respect to steps of the radiomics workflow such as segmentation, or delineation of the region of interest on a scan. Nonetheless, the introduction of the radiomics quality score (RQS) demonstrates a step towards greater standardization and reproducibility in the young field of radiomics. Altogether, in the setting of continually improving artificial intelligence algorithms, radiomics represents a promising biomarker avenue for promoting enhanced and tailored management of HPB cancers, with the potential to improve long-term outcomes for patients.

Preoperative differentiation of pancreatic cystic neoplasm subtypes on computed tomography radiomics

BACKGROUND

Serous cystic neoplasm (SCN), mucinous cystic neoplasm (MCN), and intraductal papillary mucinous neoplasm (IPMN) comprise a large proportion of pancreatic cystic neoplasms (PCNs). Patients with MCN and IPMN require surgery due to the potential of malignant transformation, whereas those with SCN require periodic surveillance. However, the differential diagnosis of patients with PCNs before treatment remains a great challenge for all surgeons. Therefore, the establishment of a reliable diagnostic tool is urgently required for the improvement of precision diagnostics.

METHODS

Between February 2015 and December 2020, 143 consecutive patients with PCNs who were confirmed by postoperative pathology were retrospectively included in the study cohort, then randomized into development and test cohorts at a ratio of 7:3. The predictors of preoperative clinical-radiologic parameters were evaluated by univariate and multivariable logistic regression analyses. A total of 1,218 radiomics features were computationally extracted from the enhanced computed tomography (CT) scans of the tumor region, and a radiomics signature was established by the random forest algorithm. In the development cohort, multi- and binary-class radiomics models integrating preoperative variables and radiomics features were constructed to distinguish between the 3 types of PCNs. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used to evaluate the predictive efficiency of the model. An independent internal test cohort was applied to validate the classification models.

RESULTS

All preoperative prediction models were built by integrating the radiomics signature with 13 diagnosis-related radiomics features and 3 important clinical-radiologic parameters: age, sex, and tumor diameter. The multiclass prediction model presented an overall accuracy of 0.804 in the development cohort and 0.707 in the test cohort. The binary-class prediction models displayed higher overall accuracies of 0.853, 0.866, and 0.928 in the development dataset and 0.750, 0.839, and 0.889 in the test dataset. In the test cohort, the binary-class radiomics models showed better predictive performances {AUC =0.914 [95% confidence interval (CI): 0.786 to 1.000], 0.863 (95% CI: 0.714 to 0.941), and 0.926 (95% CI: 0.824 to 1.000)} than the multiclass radiomics model [AUC =0.850 (95% CI: 0.696 to 1.000)], with a large net benefit in the decision curve analysis (DCA). The radiomics-based nomogram provided the correct predicted probability for the diagnosis of PCNs.

CONCLUSIONS

The proposed radiomics models with clinical-radiologic parameters and radiomics features help to predict the accurate diagnosis among PCNs to advance personalized medicine.

The Role of Magnetic Resonance Imaging (MRI) in the Diagnosis of Pancreatic Cystic Lesions (PCLs)

Pancreatic cystic lesions (PCLs) are a common incidental finding on cross-sectional imaging. Given the high signal to noise and contrast resolution, multi-parametric capability and lack of ionizing radiation, magnetic resonance imaging (MRI) has become the non-invasive method of choice to predict cyst type, risk stratify the presence of neoplasia, and monitor changes during surveillance. In many patients with PCLs, the combination of MRI and the patient's history and demographics will suffice to stratify lesions and guide treatment decisions. In other patients, especially those with worrisome or high-risk features, a multimodal diagnostic approach that includes endoscopic ultrasound (EUS) with fluid analysis, digital pathomics, and/or molecular analysis is often necessary to decide on management options. The application of radiomics and artificial intelligence in MRI may improve the ability to non-invasively stratify PCLs and better guide treatment decisions. This review will summarize the evidence on the evolution of MRI for PCLs, the prevalence of PCLs using MRI, and the MRI features to diagnose specific PCL types and early malignancy. We will also describe topics such as the utility of gadolinium and secretin in MRIs of PCLs, the limitations of MRI for PCLs, and future directions.

Radiomics model-based algorithm for preoperative prediction of pancreatic ductal adenocarcinoma grade

OBJECTIVE

To develop diagnostic radiomic model-based algorithm for pancreatic ductal adenocarcinoma (PDAC) grade prediction.

METHODS

Ninety-one patients with histologically confirmed PDAC and preoperative CT were divided into subgroups based on tumor grade. Two histology-blinded radiologists independently segmented lesions for quantitative texture analysis in all contrast enhancement phases. The ratio of densities of PDAC and unchanged pancreatic tissue, and relative tumor enhancement (RTE) in arterial, portal venous, and delayed phases of the examination were calculated. Principal component analysis was used for multivariate predictor analysis. The selection of predictors in the binary logistic model was carried out in 2 stages: (1) using one-factor logistic models (selection criterion was p < 0.1); (2) using regularization (LASSO regression after standardization of variables). Predictors were included in proportional odds models without interactions.

RESULTS

There were significant differences in 4, 16, and 8 texture features out of 62 for the arterial, portal venous, and delayed phases of the study, respectively (p < 0.1). After selection, the final diagnostic model included such radiomics features as DISCRETIZED HU standard, DISCRETIZED HUQ3, GLCM Correlation, GLZLM LZLGE for the portal venous phase of the contrast enhancement, and CONVENTIONAL_HUQ3 for the delayed phase of CT study. On its basis, a diagnostic model was built, showing AUC for grade ≥ 2 of 0.75 and AUC for grade 3 of 0.66.

CONCLUSION

Radiomics features vary in PDAC of different grades and increase the accuracy of CT in preoperative diagnosis. We have developed a diagnostic model, including texture features, which can be used to predict the grade of PDAC.

KEY POINTS

• A diagnostic algorithm based on CT texture features for preoperative PDAC grade prediction was developed. • The assumption that the scanning protocol can influence the results of texture analysis was confirmed and assessed. • Our results show that tumor differentiation grade can be assessed with sufficient diagnostic accuracy using CT texture analysis presented in this study.

Risk Assessment and Pancreatic Cancer: Diagnostic Management and Artificial Intelligence

Pancreatic cancer (PC) is one of the deadliest cancers, and it is responsible for a number of deaths almost equal to its incidence. The high mortality rate is correlated with several explanations; the main one is the late disease stage at which the majority of patients are diagnosed. Since surgical resection has been recognised as the only curative treatment, a PC diagnosis at the initial stage is believed the main tool to improve survival. Therefore, patient stratification according to familial and genetic risk and the creation of screening protocol by using minimally invasive diagnostic tools would be appropriate. Pancreatic cystic neoplasms (PCNs) are subsets of lesions which deserve special management to avoid overtreatment. The current PC screening programs are based on the annual employment of magnetic resonance imaging with cholangiopancreatography sequences (MR/MRCP) and/or endoscopic ultrasonography (EUS). For patients unfit for MRI, computed tomography (CT) could be proposed, although CT results in lower detection rates, compared to MRI, for small lesions. The actual major limit is the incapacity to detect and characterize the pancreatic intraepithelial neoplasia (PanIN) by EUS and MR/MRCP. The possibility of utilizing artificial intelligence models to evaluate higher-risk patients could favour the diagnosis of these entities, although more data are needed to support the real utility of these applications in the field of screening. For these motives, it would be appropriate to realize screening programs in research settings.

Classification prediction of pancreatic cystic neoplasms based on radiomics deep learning models

BACKGROUND

Preoperative prediction of pancreatic cystic neoplasm (PCN) differentiation has significant value for the implementation of personalized diagnosis and treatment plans. This study aimed to build radiomics deep learning (DL) models using computed tomography (CT) data for the preoperative differential diagnosis of common cystic tumors of the pancreas.

METHODS

Clinical and CT data of 193 patients with PCN were collected for this study. Among these patients, 99 were pathologically diagnosed with pancreatic serous cystadenoma (SCA), 55 were diagnosed with mucinous cystadenoma (MCA) and 39 were diagnosed with intraductal papillary mucinous neoplasm (IPMN). The regions of interest (ROIs) were obtained based on manual image segmentation of CT slices. The radiomics and radiomics-DL models were constructed using support vector machines (SVMs). Moreover, based on the fusion of clinical and radiological features, the best combined feature set was obtained according to the Akaike information criterion (AIC) analysis. Then the fused model was constructed using logistic regression.

RESULTS

For the SCA differential diagnosis, the fused model performed the best and obtained an average area under the curve (AUC) of 0.916. It had a best feature set including position, polycystic features (≥6), cystic wall calcification, pancreatic duct dilatation and radiomics-DL score. For the MCA and IPMN differential diagnosis, the fused model with AUC of 0.973 had a best feature set including age, communication with the pancreatic duct and radiomics score.

CONCLUSIONS

The radiomics, radiomics-DL and fused models based on CT images have a favorable differential diagnostic performance for SCA, MCA and IPMN. These findings may be beneficial for the exploration of individualized management strategies.

Update on quantitative radiomics of pancreatic tumors

Radiomics is a newer approach for analyzing radiological images obtained from conventional imaging modalities such as computed tomography, magnetic resonance imaging, endoscopic ultrasonography, and positron emission tomography. Radiomics involves extracting quantitative data from the images and assessing them to identify diagnostic or prognostic features such as tumor grade, resectability, tumor response to neoadjuvant therapy, and survival. The purpose of this review is to discuss the basic principles of radiomics and provide an overview of the current clinical applications of radiomics in the field of pancreatic tumors.

Role of endoscopic ultrasound and cyst fluid tumor markers in diagnosis of pancreatic cystic lesions

BACKGROUND

Pancreatic cystic lesions (PCLs) are common in clinical practice. The accurate classification and diagnosis of these lesions are crucial to avoid unnecessary treatment of benign lesions and missed opportunities for early treatment of potentially malignant lesions.

AIM

To evaluate the role of cyst fluid analysis of different tumor markers such as cancer antigens [e.g., cancer antigen (CA)19-9, CA72-4], carcinoembryonic antigen (CEA), serine protease inhibitor Kazal-type 1 (SPINK1), interleukin 1 beta (IL1-β), vascular endothelial growth factor A (VEGF-A), and prostaglandin E2 (PGE2)], amylase, and mucin stain in diagnosing pancreatic cysts and differentiating malignant from benign lesions.

METHODS

This study included 76 patients diagnosed with PCLs using different imaging modalities. All patients underwent endoscopic ultrasound (EUS) and EUS-fine needle aspiration (EUS-FNA) for characterization and sampling of different PCLs.

RESULTS

The mean age of studied patients was 47.4 ± 11.4 years, with a slight female predominance (59.2%). Mucin stain showed high statistical significance in predicting malignancy with a sensitivity of 87.1% and specificity of 95.56%. It also showed a positive predictive value and negative predictive value of 93.1% and 91.49%, respectively (P < 0.001). We found that positive mucin stain, cyst fluid glucose, SPINK1, amylase, and CEA levels had high statistical significance (P < 0.0001). In contrast, IL-1β, CA 72-4, VEGF-A, VEGFR2, and PGE2 did not show any statistical significance. Univariate regression analysis for prediction of malignancy in PCLs showed a statistically significant positive correlation with mural nodules, lymph nodes, cyst diameter, mucin stain, and cyst fluid CEA. Meanwhile, logistic multivariable regression analysis proved that mural nodules, mucin stain, and SPINK1 were independent predictors of malignancy in cystic pancreatic lesions.

CONCLUSION

EUS examination of cyst morphology with cytopathological analysis and cyst fluid analysis could improve the differentiation between malignant and benign pancreatic cysts. Also, CEA, glucose, and SPINK1 could be used as promising markers to predict malignant pancreatic cysts.

Deficient Rnf43 potentiates hyperactive Kras-mediated pancreatic preneoplasia initiation and malignant transformation

BACKGROUND

Largely due to incidental detection, asymptomatic pancreatic cystic lesions (PCLs) have become prevalent in recent years. Among them, intraductal papillary mucinous neoplasm (IPMN) infrequently advances to pancreatic ductal adenocarcinoma (PDAC). Conservative surveillance versus surgical intervention is a difficult clinical decision for both caregivers and PCL patients. Because RNF43 loss-of-function mutations and KRAS gain-of-function mutations concur in a subset of IPMN and PDAC, their biological significance and therapeutic potential should be elucidated.

METHODS

Pancreatic Rnf43 knockout and Kras activated mice (Rnf43-/-; KrasG12D) were generated to evaluate their clinical significance in pancreatic pre-neoplastic initiation and malignant transformation.

RESULTS

Loss of Rnf43 potentiated the occurrence and severity of IPMN and PDAC in oncogenic Kras mice. The Wnt/β-catenin signaling pathway was activated in pancreatic KrasG12D and Rnf43 knockout mice and the PORCN inhibitor LGK974 blocked pancreatic IPMN initiation and progression to PDAC accordingly.

CONCLUSIONS

Rnf43 is a tumor suppressor in the prevention of pancreatic malignant transformation. This genetically reconstituted autochthonous pancreatic Rnf43-/-; KrasG12D preclinical cancer model recapitulates the pathological process from pancreatic cyst to cancer in humans and can be treated with inhibitors of Wnt/β-catenin signaling. Since the presence of RNF43 and KRAS mutations in IPMNs predicts future development of advanced neoplasia from PCLs, patients with these genetic anomalies warrant surveillance, surgery, and/or targeted therapeutics such as Wnt/β-catenin inhibitors.

Ultrasound Radiomics Nomogram Integrating Three-Dimensional Features Based on Carotid Plaques to Evaluate Coronary Artery Disease

This study aimed to explore the feasibility of ultrasound radiomics analysis before invasive coronary angiography (ICA) for evaluating the severity of coronary artery disease (CAD) quantified by the SYNTAX score (SS). This study included 105 carotid plaques from 105 patients (64 low-SS patients, 41 intermediate-high-SS patients). The clinical characteristics and three-dimensional ultrasound (3D-US) features before ICA were assessed. Ultrasound images of carotid plaques were used for radiomics analysis. Least absolute shrinkage and selection operator (LASSO) regression, which generated several nonzero coefficients, was used to select features that could predict intermediate-high SS. Based on those coefficients, the radiomics score (Rad-score) was calculated. The selected clinical characteristics, 3D-US features, and Rad-score were finally integrated into a radiomics nomogram. Among the clinical characteristics and 3D-US features, high-density lipoprotein (HDL), apolipoprotein B (Apo B), and plaque volume were identified as predictors for distinguishing between low SS and intermediate-high SS. During the radiomics process, 8 optimal radiomics features most capable of identifying intermediate-high SS were selected from 851 candidate radiomics features. The differences in Rad-score between the training and the validation set were significant (p = 0.016 and 0.006). The radiomics nomogram integrating HDL, Apo B, plaque volume, and Rad-score showed excellent results in the training set (AUC, 0.741 (95% confidence interval (CI): 0.646–0.835)) and validation set (AUC, 0.939 (95% CI: 0.860–1.000)), with good calibration (mean absolute errors of 0.028 and 0.059 in training and validation sets, respectively). Decision curve analysis showed that the radiomics nomogram could identify patients who could obtain the most benefit. We concluded that the radiomics nomogram based on carotid plaque ultrasound has favorable value for the noninvasive prediction of intermediate-high SS. This radiomics nomogram has potential value for the risk stratification of CAD before ICA and provides clinicians with a noninvasive diagnostic tool.

Pancreatic Serous Cystic Neoplasms and Mucinous Cystic Neoplasms: Differential Diagnosis by Combining Imaging Features and Enhanced CT Texture Analysis

Objective

To establish a diagnostic model by combining imaging features with enhanced CT texture analysis to differentiate pancreatic serous cystadenomas (SCNs) from pancreatic mucinous cystadenomas (MCNs).

Materials and Methods

Fifty-seven and 43 patients with pathology-confirmed SCNs and MCNs, respectively, from one center were analyzed and divided into a training cohort (n = 72) and an internal validation cohort (n = 28). An external validation cohort (n = 28) from another center was allocated. Demographic and radiological information were collected. The least absolute shrinkage and selection operator (LASSO) and recursive feature elimination linear support vector machine (RFE_LinearSVC) were implemented to select significant features. Multivariable logistic regression algorithms were conducted for model construction. Receiver operating characteristic (ROC) curves for the models were evaluated, and their prediction efficiency was quantified by the area under the curve (AUC), 95% confidence interval (95% CI), sensitivity and specificity.

Results

Following multivariable logistic regression analysis, the AUC was 0.932 and 0.887, the sensitivity was 87.5% and 90%, and the specificity was 82.4% and 84.6% with the training and validation cohorts, respectively, for the model combining radiological features and CT texture features. For the model based on radiological features alone, the AUC was 0.84 and 0.91, the sensitivity was 75% and 66.7%, and the specificity was 82.4% and 77% with the training and validation cohorts, respectively.

Conclusion

This study showed that a logistic model combining radiological features and CT texture features is more effective in distinguishing SCNs from MCNs of the pancreas than a model based on radiological features alone.

Pancreatic Cancer Surveillance and Novel Strategies for Screening

Individuals with a genetic susceptibility to pancreatic ductal adenocarcinoma (PDAC) may benefit from surveillance to increase the likelihood of early detection. Currently, candidates for surveillance are identified based on genetic test results and family history of PDAC, and surveillance is accomplished through imaging of the pancreas (endoscopic ultrasound or MRI). Novel methods that incorporate personalized risk, biomarkers, and radiomics are being investigated in an attempt to improve identification of at-risk individuals and to increase detection of precursor and early-stage lesions.

High performance in risk stratification of intraductal papillary mucinous neoplasms by confocal laser endomicroscopy image analysis with convolutional neural networks (with video)

BACKGROUND AND AIMS

EUS-guided needle-based confocal laser endomicroscopy (EUS-nCLE) can differentiate high-grade dysplasia/adenocarcinoma (HGD-Ca) in intraductal papillary mucinous neoplasms (IPMNs) but requires manual interpretation. We sought to derive predictive computer-aided diagnosis (CAD) and artificial intelligence (AI) algorithms to facilitate accurate diagnosis and risk stratification of IPMNs.

METHODS

A post hoc analysis of a single-center prospective study evaluating EUS-nCLE (2015-2019; INDEX study) was conducted using 15,027 video frames from 35 consecutive patients with histopathologically proven IPMNs (18 with HGD-Ca). We designed 2 CAD-convolutional neural network (CNN) algorithms: (1) a guided segmentation-based model (SBM), where the CNN-AI system was trained to detect and measure papillary epithelial thickness and darkness (indicative of cellular and nuclear stratification), and (2) a reasonably agnostic holistic-based model (HBM) where the CNN-AI system automatically extracted nCLE features for risk stratification. For the detection of HGD-Ca in IPMNs, the diagnostic performance of the CNN-CAD algorithms was compared with that of the American Gastroenterological Association (AGA) and revised Fukuoka guidelines.

RESULTS

Compared with the guidelines, both n-CLE-guided CNN-CAD algorithms yielded higher sensitivity (HBM, 83.3%; SBM, 83.3%; AGA, 55.6%; Fukuoka, 55.6%) and accuracy (SBM, 82.9%; HBM, 85.7%; AGA, 68.6%; Fukuoka, 74.3%) for diagnosing HGD-Ca, with comparable specificity (SBM, 82.4%; HBM, 88.2%; AGA, 82.4%; Fukuoka, 94.1%). Both CNN-CAD algorithms, the guided (SBM) and agnostic (HBM) models, were comparable in risk stratifying IPMNs.

CONCLUSION

EUS-nCLE-based CNN-CAD algorithms can accurately risk stratify IPMNs. Future multicenter validation studies and AI model improvements could enhance the accuracy and fully automatize the process for real-time interpretation.