Classification of pancreatic cystic neoplasms using radiomic feature analysis is equivalent to an experienced academic radiologist: a step toward computer-augmented diagnostics for radiologists

The Role of AI in the Evaluation of Neuroendocrine Tumors: Current State of the Art

Advancements in Artificial Intelligence (AI) are driving a paradigm shift in the field of medical diagnostics, integrating new developments into various aspects of the clinical workflow. Neuroendocrine neoplasms are a diverse and heterogeneous group of tumors that pose significant diagnostic and management challenges due to their variable clinical presentations and biological behavior. Innovative approaches are essential to overcome these challenges and improve the current standard of care. AI-driven applications, particularly in imaging workflows, hold promise for enhancing tumor detection, classification, and grading by leveraging advanced radiomics and deep learning techniques. This article reviews the current and emerging applications of AI computer vision in the care of neuroendocrine neoplasms, focusing on its integration into imaging workflows, diagnostics, prognostic modeling, and therapeutic planning.

Arterial phase CT radiomics for non-invasive prediction of Ki-67 proliferation index in pancreatic solid pseudopapillary neoplasms

BACKGROUND

This study aimed to preoperatively predict Ki-67 proliferation levels in patients with pancreatic solid pseudopapillary neoplasm (pSPN) using radiomics features extracted from arterial phase helical CT images.

METHODS

We retrospectively analyzed 92 patients (Ningbo Medical Center Lihuili Hospital: n = 64, Taizhou Central Hospital: n = 28) with pathologically confirmed pSPN from June 2015 to June 2023. Ki-67 positivity > 3% was considered high. Radiomics features were extracted using PyRadiomics, with patients from training cohort (n = 64) and validation cohort (n = 28). A radiomics signature was constructed, and a CT radiomics score (CTscore) was calculated. Deep learning models were employed for prediction, with early stopping to prevent overfitting.

RESULTS

Seven key radiomics features were selected via LASSO regression with cross-validation. The deep learning model demonstrated improved accuracy with demographics and CTscore, with key features such as Morphology and CTscore contributing significantly to predictive accuracy. The best-performing models, including GBM and deep learning algorithms, achieved high predictive performance with an AUC of up to 0.946 in the training cohort.

CONCLUSIONS

We developed a robust deep learning-based radiomics model using arterial phase CT images to predict Ki-67 levels in pSPN patients, identifying CTscore and Morphology as key predictors. This non-invasive approach has potential utility in guiding personalized preoperative treatment strategies.

CLINICAL TRIAL NUMBER

Not applicable.

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.

Artificial Intelligence in Pancreatic Intraductal Papillary Mucinous Neoplasm Imaging: A Systematic Review

Background

Based on the Fukuoka and Kyoto international consensus guidelines, the current clinical management of intraductal papillary mucinous neoplasm (IPMN) largely depends on imaging features. While these criteria are highly sensitive in detecting high-risk IPMN, they lack specificity, resulting in surgical overtreatment. Artificial Intelligence (AI)-based medical image analysis has the potential to augment the clinical management of IPMNs by improving diagnostic accuracy.

Methods

Based on a systematic review of the academic literature on AI in IPMN imaging, 1041 publications were identified of which 25 published studies were included in the analysis. The studies were stratified based on prediction target, underlying data type and imaging modality, patient cohort size, and stage of clinical translation and were subsequently analyzed to identify trends and gaps in the field.

Results

Research on AI in IPMN imaging has been increasing in recent years. The majority of studies utilized CT imaging to train computational models. Most studies presented computational models developed on single-center datasets (n=11,44%) and included less than 250 patients (n=18,72%). Methodologically, convolutional neural network (CNN)-based algorithms were most commonly used. Thematically, most studies reported models augmenting differential diagnosis (n=9,36%) or risk stratification (n=10,40%) rather than IPMN detection (n=5,20%) or IPMN segmentation (n=2,8%).

Conclusion

This systematic review provides a comprehensive overview of the research landscape of AI in IPMN imaging. Computational models have potential to enhance the accurate and precise stratification of patients with IPMN. Multicenter collaboration and datasets comprising various modalities are necessary to fully utilize this potential, alongside concerted efforts towards clinical translation.

Radiomics machine learning algorithm facilitates detection of small pancreatic neuroendocrine tumors on CT

PURPOSE

The purpose of this study was to develop a radiomics-based algorithm to identify small pancreatic neuroendocrine tumors (PanNETs) on CT and evaluate its robustness across manual and automated segmentations, exploring the feasibility of automated screening.

MATERIALS AND METHODS

Patients with pathologically confirmed T1 stage PanNETs and healthy controls undergoing dual-phase CT imaging were retrospectively identified. Manual segmentation of pancreas and tumors was performed, then automated pancreatic segmentations were generated using a pretrained neural network. A total of 1223 radiomics features were independently extracted from both segmentation volumes, in the arterial and venous phases separately. Ten final features were selected to train classifiers to identify PanNETs and controls. The cohort was divided into training and testing sets, and performance of classifiers was assessed using area under the receiver operator characteristic curve (AUC), specificity and sensitivity, and compared against two radiologists blinded to the diagnoses.

RESULTS

A total of 135 patients with 142 PanNETs, and 135 healthy controls were included. There were 168 women and 102 men, with a mean age of 55.4 ± 11.6 (standard deviation) years (range: 20-85 years). Median PanNET size was 1.3 cm (Q1, 1.0; Q3, 1.5; range: 0.5-1.9). The arterial phase LightGBM model achieved the best performance in the test set, with 90 % sensitivity (95 % confidence interval [CI]: 80-98), 76 % specificity (95 % CI: 62-88) and an AUC of 0.87 (95 % CI: 0.79-0.94). Using features from the automated segmentations, this model achieved an AUC of 0.86 (95 % CI: 0.79-0.93). In comparison, the two radiologists achieved a mean 50 % sensitivity and 100 % specificity using arterial phase CT images.

CONCLUSION

Radiomics features identify small PanNETs, with stable performance when extracted using automated segmentations. These models demonstrate high sensitivity, complementing the high specificity of radiologists, and could serve as opportunistic screeners.

Advances for Managing Pancreatic Cystic Lesions: Integrating Imaging and AI Innovations

Pancreatic cystic lesions (PCLs) represent a spectrum of non-neoplasms and neoplasms with varying malignant potential, posing significant challenges in diagnosis and management. While some PCLs are precursors to pancreatic cancer, others remain benign, necessitating accurate differentiation for optimal patient care. Conventional approaches to PCL management rely heavily on radiographic imaging, and endoscopic ultrasound (EUS) guided fine-needle aspiration (FNA), coupled with clinical and biochemical data. However, the observer-dependent nature of image interpretation and the complex morphology of PCLs can lead to diagnostic uncertainty and variability in patient management strategies. This review critically evaluates current PCL diagnosis and surveillance practices, showing features of the different lesions and highlighting the potential limitations of conventional methods. We then explore the potential of artificial intelligence (AI) to transform PCL management. AI-driven strategies, including deep learning algorithms for automated pancreas and lesion segmentation, and radiomics for analyzing heterogeneity, can improve diagnostic accuracy and risk stratification. These advanced techniques can provide more objective and reproducible assessments, aiding clinicians in decision-making regarding follow-up intervals and surgical interventions. Early results suggest that AI-driven methods can significantly improve patient outcomes by enabling earlier detection of high-risk lesions and reducing unnecessary procedures for benign cysts. Finally, this review emphasizes that AI-driven approaches could potentially reshape the landscape of PCL management, ultimately leading to improved pancreatic cancer prevention.

Imaging of pancreatic serous cystadenoma and common imitators

Pancreatic cystic neoplasms are lesions comprised of cystic components that show different biological behaviors, epidemiology, clinical manifestations, imaging features, and malignant potential and management. Benign cystic neoplasms include serous cystic neoplasms (SCAs). Other pancreatic cystic lesions have malignant potential, such as intraductal papillary mucinous neoplasms and mucinous cystic neoplasms. SCAs can be divided into microcystic (classic appearance), honeycomb, oligocystic/macrocystic, and solid patterns based on imaging appearance. They are usually solitary but may be multiple in von Hippel-Lindau disease, which may depict disseminated involvement. The variable appearances of SCAs can mimic other types of pancreatic cystic lesions, and cross-sectional imaging plays an important role in their differential diagnosis. Endoscopic ultrasonography has helped in improving diagnostic accuracy of pancreatic cystic lesions by guiding tissue sampling (biopsy) or cyst fluid analysis. Immunohistochemistry and newer techniques such as radiomics have shown improved performance for preoperatively discriminating SCAs and their mimickers.

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.

Utilization of texture features of volumetric ADC maps in differentiating between serous cystadenoma and intraductal papillary neoplasms

INTRODUCTION

The rising incidence of incidental detection of pancreatic cystic neoplasms has compelled radiologists to determine new diagnostic methods for the differentiation of various kinds of lesions. We aim to demonstrate the utility of texture features extracted from ADC maps in differentiating intraductal papillary mucinous neoplasms (IPMN) from serous cystadenomas (SCA).

METHODS

This retrospective study was performed on 136 patients (IPMN = 87, SCA = 49) split into testing and training datasets. A total of 851 radiomics features were extracted from volumetric contours drawn by an expert radiologist on ADC maps of the lesions. LASSO regression analysis was used to determine the most predictive set of features and a radiomics score was developed based on their respective coefficients. A hyper-optimized support vector machine was then utilized to classify the lesions based on their radiomics score.

RESULTS

A total of four Wavelet features (LHL/GLCM/LCM2, HLL/GLCM/LCM2, /LLL/First Order/90percent, /LLL/GLCM/MCC) were selected from all of the features to be included in our classifier. The classifier was optimized by altering hyperparameters and the trained model was applied to the validation dataset. The model achieved a sensitivity of 92.8, specificity of 90%, and an AUC of 0.97 in the training data set, and a sensitivity of 83.3%, specificity of 66.7%, and AUC of 0.90 in the testing dataset.

CONCLUSION

A support vector machine model trained and validated on volumetric texture features extracted from ADC maps showed the possible beneficence of these features in differentiating IPMNs from SCAs. These results are in line with previous regarding the role of ADC maps in classifying cystic lesions and offers new evidence regarding the role of texture features in differentiation of potentially neoplastic and benign lesions.

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.

Accurate non-invasive grading of nonfunctional pancreatic neuroendocrine tumors with a CT derived radiomics signature

PURPOSE

The purpose of this study was to develop a radiomics-signature using computed tomography (CT) data for the preoperative prediction of grade of nonfunctional pancreatic neuroendocrine tumors (NF-PNETs).

MATERIALS AND METHODS

A retrospective study was performed on patients undergoing resection for NF-PNETs between 2010 and 2019. A total of 2436 radiomic features were extracted from arterial and venous phases of pancreas-protocol CT examinations. Radiomic features that were associated with final pathologic grade observed in the surgical specimens were subjected to joint mutual information maximization for hierarchical feature selection and the development of the radiomic-signature. Youden-index was used to identify optimal cutoff for determining tumor grade. A random forest prediction model was trained and validated internally. The performance of this tool in predicting tumor grade was compared to that of EUS-FNA sampling that was used as the standard of reference.

RESULTS

A total of 270 patients were included and a fusion radiomic-signature based on 10 selected features was developed using the development cohort (n = 201). There were 149 men and 121 women with a mean age of 59.4 ± 12.3 (standard deviation) years (range: 23.3-85.0 years). Upon internal validation in a new set of 69 patients, a strong discrimination was observed with an area under the curve (AUC) of 0.80 (95% confidence interval [CI]: 0.71-0.90) with corresponding sensitivity and specificity of 87.5% (95% CI: 79.7-95.3) and 73.3% (95% CI: 62.9-83.8) respectively. Of the study population, 143 patients (52.9%) underwent EUS-FNA. Biopsies were non-diagnostic in 26 patients (18.2%) and could not be graded due to insufficient sample in 42 patients (29.4%). In the cohort of 75 patients (52.4%) in whom biopsies were graded the radiomic-signature demonstrated not different AUC as compared to EUS-FNA (AUC: 0.69 vs. 0.67; P = 0.723), however greater sensitivity (i.e., ability to accurately identify G2/3 lesion was observed (80.8% vs. 42.3%; P < 0.001).

CONCLUSION

Non-invasive assessment of tumor grade in patients with PNETs using the proposed radiomic-signature demonstrated high accuracy. Prospective validation and optimization could overcome the commonly experienced diagnostic uncertainty in the assessment of tumor grade in patients with PNETs and could facilitate clinical decision-making.

Large-scale pancreatic cancer detection via non-contrast CT and deep learning

Pancreatic ductal adenocarcinoma (PDAC), the most deadly solid malignancy, is typically detected late and at an inoperable stage. Early or incidental detection is associated with prolonged survival, but screening asymptomatic individuals for PDAC using a single test remains unfeasible due to the low prevalence and potential harms of false positives. Non-contrast computed tomography (CT), routinely performed for clinical indications, offers the potential for large-scale screening, however, identification of PDAC using non-contrast CT has long been considered impossible. Here, we develop a deep learning approach, pancreatic cancer detection with artificial intelligence (PANDA), that can detect and classify pancreatic lesions with high accuracy via non-contrast CT. PANDA is trained on a dataset of 3,208 patients from a single center. PANDA achieves an area under the receiver operating characteristic curve (AUC) of 0.986-0.996 for lesion detection in a multicenter validation involving 6,239 patients across 10 centers, outperforms the mean radiologist performance by 34.1% in sensitivity and 6.3% in specificity for PDAC identification, and achieves a sensitivity of 92.9% and specificity of 99.9% for lesion detection in a real-world multi-scenario validation consisting of 20,530 consecutive patients. Notably, PANDA utilized with non-contrast CT shows non-inferiority to radiology reports (using contrast-enhanced CT) in the differentiation of common pancreatic lesion subtypes. PANDA could potentially serve as a new tool for large-scale pancreatic cancer screening.

Serous Cystadenoma: A Review on Diagnosis and Management

Incidental pancreatic cysts are highly prevalent, with management dependent on the risk of malignant progression. Serous cystadenomas (SCAs) are the most common benign pancreatic cysts seen on imaging. They have typical morphological patterns but may also show atypical features that mimic precancerous and cancerous cysts. If a confident diagnosis of SCA is made, no further follow-up is warranted. Therefore, a preoperative distinction between SCA and precancerous or cancerous lesions is critically essential. Distinguishing an SCA from other types of pancreatic cysts on imaging remains a challenge, thus leading to misdiagnosis and ramifications. This review summarizes the current evidence on diagnosing and managing SCA.

Improving Pancreatic Cyst Management: Artificial Intelligence-Powered Prediction of Advanced Neoplasms through Endoscopic Ultrasound-Guided Confocal Endomicroscopy

Despite the increasing rate of detection of incidental pancreatic cystic lesions (PCLs), current standard-of-care methods for their diagnosis and risk stratification remain inadequate. Intraductal papillary mucinous neoplasms (IPMNs) are the most prevalent PCLs. The existing modalities, including endoscopic ultrasound and cyst fluid analysis, only achieve accuracy rates of 65-75% in identifying carcinoma or high-grade dysplasia in IPMNs. Furthermore, surgical resection of PCLs reveals that up to half exhibit only low-grade dysplastic changes or benign neoplasms. To reduce unnecessary and high-risk pancreatic surgeries, more precise diagnostic techniques are necessary. A promising approach involves integrating existing data, such as clinical features, cyst morphology, and data from cyst fluid analysis, with confocal endomicroscopy and radiomics to enhance the prediction of advanced neoplasms in PCLs. Artificial intelligence and machine learning modalities can play a crucial role in achieving this goal. In this review, we explore current and future techniques to leverage these advanced technologies to improve diagnostic accuracy in the context of PCLs.

A primer on artificial intelligence in pancreatic imaging

Artificial Intelligence (AI) is set to transform medical imaging by leveraging the vast data contained in medical images. Deep learning and radiomics are the two main AI methods currently being applied within radiology. Deep learning uses a layered set of self-correcting algorithms to develop a mathematical model that best fits the data. Radiomics converts imaging data into mineable features such as signal intensity, shape, texture, and higher-order features. Both methods have the potential to improve disease detection, characterization, and prognostication. This article reviews the current status of artificial intelligence in pancreatic imaging and critically appraises the quality of existing evidence using the radiomics quality score.

Pancreatic Cystic Lesions: Next Generation of Radiologic Assessment

Pancreatic cystic lesions are frequently identified on cross-sectional imaging. As many of these are presumed branch-duct intraductal papillary mucinous neoplasms, these lesions generate much anxiety for the patients and clinicians, often necessitating long-term follow-up imaging and even unnecessary surgical resections. However, the incidence of pancreatic cancer is overall low for patients with incidental pancreatic cystic lesions. Radiomics and deep learning are advanced tools of imaging analysis that have attracted much attention in addressing this unmet need, however, current publications on this topic show limited success and large-scale research is needed.

Artificial Intelligence in the Diagnosis and Treatment of Pancreatic Cystic Lesions and Adenocarcinoma

Pancreatic cancer is projected to become the second leading cause of cancer-related mortality in the United States by 2030. This is in part due to the paucity of reliable screening and diagnostic options for early detection. Amongst known pre-malignant pancreatic lesions, pancreatic intraepithelial neoplasia (PanIN) and intraductal papillary mucinous neoplasms (IPMNs) are the most prevalent. The current standard of care for the diagnosis and classification of pancreatic cystic lesions (PCLs) involves cross-sectional imaging studies and endoscopic ultrasound (EUS) and, when indicated, EUS-guided fine needle aspiration and cyst fluid analysis. However, this is suboptimal for the identification and risk stratification of PCLs, with accuracy of only 65-75% for detecting mucinous PCLs. Artificial intelligence (AI) is a promising tool that has been applied to improve accuracy in screening for solid tumors, including breast, lung, cervical, and colon cancer. More recently, it has shown promise in diagnosing pancreatic cancer by identifying high-risk populations, risk-stratifying premalignant lesions, and predicting the progression of IPMNs to adenocarcinoma. This review summarizes the available literature on artificial intelligence in the screening and prognostication of precancerous lesions in the pancreas, and streamlining the diagnosis of pancreatic cancer.