Present status of ultrasound elastography for the diagnosis of pancreatic tumors: review of the literature

Recent developments in the diagnosis of pancreatic neuroendocrine neoplasms

INTRODUCTION

Pancreatic Neuroendocrine Neoplasms (PanNENs) are characterized by a highly heterogeneous clinical and biological behavior, making their diagnosis challenging. PanNENs diagnostic work-up mainly relies on biochemical markers, pathological examination, and imaging evaluation. The latter includes radiological imaging (i.e. computed tomography [CT] and magnetic resonance imaging [MRI]), functional imaging (i.e. 68Gallium [68 Ga]Ga-DOTA-peptide PET/CT and Fluorine-18 fluorodeoxyglucose [18F]FDG PET/CT), and endoscopic ultrasound (EUS) with its associated procedures.

AREAS COVERED

This review provides a comprehensive assessment of the recent advancements in the PanNENs diagnostic field. PubMed and Embase databases were used for the research, performed from inception to October 2023.

EXPERT OPINION

A deeper understanding of PanNENs biology, recent technological improvements in imaging modalities, as well as progresses achieved in molecular and cytological assays, are fundamental players for the achievement of early diagnosis and enhanced preoperative characterization of PanNENs. A multimodal diagnostic approach is required for a thorough disease assessment.

Current status of artificial intelligence analysis for the treatment of pancreaticobiliary diseases using endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography

Pancreatic and biliary diseases encompass a range of conditions requiring accurate diagnosis for appropriate treatment strategies. This diagnosis relies heavily on imaging techniques like endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography. Artificial intelligence (AI), including machine learning and deep learning, is becoming integral in medical imaging and diagnostics, such as the detection of colorectal polyps. AI shows great potential in diagnosing pancreatobiliary diseases. Unlike machine learning, which requires feature extraction and selection, deep learning can utilize images directly as input. Accurate evaluation of AI performance is a complex task due to varied terminologies, evaluation methods, and development stages. Essential aspects of AI evaluation involve defining the AI's purpose, choosing appropriate gold standards, deciding on the validation phase, and selecting reliable validation methods. AI, particularly deep learning, is increasingly employed in endoscopic ultrasonography and endoscopic retrograde cholangiopancreatography diagnostics, achieving high accuracy levels in detecting and classifying various pancreatobiliary diseases. The AI often performs better than doctors, even in tasks like differentiating benign from malignant pancreatic tumors, cysts, and subepithelial lesions, identifying gallbladder lesions, assessing endoscopic retrograde cholangiopancreatography difficulty, and evaluating the biliary strictures. The potential for AI in diagnosing pancreatobiliary diseases, especially where other modalities have limitations, is considerable. However, a crucial constraint is the need for extensive, high-quality annotated data for AI training. Future advances in AI, such as large language models, promise further applications in the medical field.

Artificial intelligence using deep learning analysis of endoscopic ultrasonography images for the differential diagnosis of pancreatic masses

BACKGROUND : There are several types of pancreatic mass, so it is important to distinguish between them before treatment. Artificial intelligence (AI) is a mathematical technique that automates learning and recognition of data patterns. This study aimed to investigate the efficacy of our AI model using endoscopic ultrasonography (EUS) images of multiple types of pancreatic mass (pancreatic ductal adenocarcinoma [PDAC], pancreatic adenosquamous carcinoma [PASC], acinar cell carcinoma [ACC], metastatic pancreatic tumor [MPT], neuroendocrine carcinoma [NEC], neuroendocrine tumor [NET], solid pseudopapillary neoplasm [SPN], chronic pancreatitis, and autoimmune pancreatitis [AIP]). METHODS : Patients who underwent EUS were included in this retrospective study. The included patients were divided into training, validation, and test cohorts. Using these cohorts, an AI model that can distinguish pancreatic carcinomas from noncarcinomatous pancreatic lesions was developed using a deep-learning architecture and the diagnostic performance of the AI model was evaluated. RESULTS : 22 000 images were generated from 933 patients. The area under the curve, sensitivity, specificity, and accuracy (95 %CI) of the AI model for the diagnosis of pancreatic carcinomas in the test cohort were 0.90 (0.84-0.97), 0.94 (0.88-0.98), 0.82 (0.68-0.92), and 0.91 (0.85-0.95), respectively. The per-category sensitivities (95 %CI) of each disease were PDAC 0.96 (0.90-0.99), PASC 1.00 (0.05-1.00), ACC 1.00 (0.22-1.00), MPT 0.33 (0.01-0.91), NEC 1.00 (0.22-1.00), NET 0.93 (0.66-1.00), SPN 1.00 (0.22-1.00), chronic pancreatitis 0.78 (0.52-0.94), and AIP 0.73 (0.39-0.94). CONCLUSIONS : Our developed AI model can distinguish pancreatic carcinomas from noncarcinomatous pancreatic lesions, but external validation is needed.

Diagnostic performance of endoscopic ultrasound elastography for differential diagnosis of solid pancreatic lesions: A propensity score-matched analysis

BACKGROUND

Endoscopic ultrasound-elastography (EUS-EG) is a non-invasive complementary diagnostic method for differential diagnosis of solid pancreatic lesions (SPL). However, the optimal strain ratio (SR) value and diagnostic performance of EUS-EG have not yet been determined in pancreatic neuroendocrine neoplasm (PNEN), mass-forming pancreatitis (MFP), and pancreatic ductal adenocarcinoma (PDAC). We aimed to determine the optimal SR value in EUS-EG for differential diagnosis of SPLs.

METHODS

Patients who underwent EUS-EG for SPL evaluation between July 2016 and June 2019 were retrospectively investigated. Patients were divided into three groups based on the final diagnosis (PNEN, MFP, or PDAC). Patient demographics, characteristics of SPL, and EUS-EG were compared.

RESULTS

The mean (± standard deviation) SR value for each group were 11.85 ± 7.56 (PNEN, n = 10), 11.45 ± 5.97 (MFP, n = 37), and 22.50 ± 13.19 (PDAC, n = 87). Multinomial logistic regression analysis revealed that an increase of SR value was significantly associated with PDAC (PNEN versus PDAC, p = 0.0216; MFP versus PDAC, p = 0.0006). The optimal cut-off value for differential diagnosis was confirmed as 17.14 after propensity score matching.

CONCLUSIONS

We provided the optimal cut-off SR values for differential diagnosis between MFP and PDAC. EUS-EG can be used as a supplementary diagnostic method in the diagnosis of SPLs. (Clinical trial registration number: https://cris.nih.go.kr/cris: KCT0002082).

Transabdominal ultrasonographic diagnosis of relatively rare pancreatic neoplasms

There are various types of pancreatic neoplasms, and their prognosis and treatment methods are different. Therefore, accurate diagnosis is important to determine the best treatment strategy. Transabdominal ultrasonography is frequently used as a screening examination for diagnostic imaging of pancreatic neoplasms. In this review, we have focused on the characteristics of ultrasonic findings for relatively rare pancreatic neoplasms.

Feasibility of endoscopic ultrasonography using a 60-MHz ultrasound miniature probe in the upper gastrointestinal tract

PURPOSE

The use of higher frequencies in ultrasound allows for a more detailed image. This study aimed to investigate the feasibility of delineating the gastrointestinal wall using a 60-MHz miniature ultrasound probe.

METHODS

A phantom study was performed using a multipurpose ultrasonic phantom model, and the depth of imaging was evaluated using 60-MHz and 20-MHz miniature probes and 7.5-MHz conventional convex-type endoscopic ultrasonography. A total of 25 visualized areas from a total of 16 specimens from 16 patients were enrolled. The structures of the layers of the esophagus, stomach, and duodenum were evaluated using a 60-MHz probe and a pathological specimen created from endoscopically or surgically resected specimens.

RESULTS

The 60-MHz probe was able to render to a depth of 2 mm and visualize the esophagus, stomach, and duodenum in five layers, respectively, within the depiction range. The depiction ranges of the 20-MHz probe and 7.5-MHz conventional endoscopic ultrasonography were 5 mm and 60 mm, respectively. The 60-MHz probe visualized the muscularis mucosae as the fourth layer in the esophagus, the fourth layer in the stomach, and the second layer in the duodenum. Muscularis mucosae were delineated in almost all cases, except in two cases where the layered structure disappeared.

CONCLUSION

The 60-MHz probe provided good visualization of the muscularis mucosae and structure of the layers down to the submucosa, which improves the ability to diagnose the depth of early cancer invasion of the upper gastrointestinal tract, leading to more appropriate treatments.

Recent Advances in Endosonography-Elastography: Literature Review

Ultrasonographic elastography is a modality used to visualize the elastic properties of tissues. Technological advances in ultrasound equipment have supported the evaluation of elastography (EG) in endosonography (EUS). Currently, the usefulness of not only EUS-strain elastography (EUS-SE) but also EUS-shear wave elastography (EUS-SWE) has been reported. We reviewed the literature on the usefulness of EUS-EG for various diseases such as chronic pancreatitis, pancreatic solid lesion, autoimmune pancreatitis, lymph node, and gastrointestinal and subepithelial lesions. The importance of this new diagnostic parameter, "tissue elasticity" in clinical practice might be applied not only to the diagnosis of liver fibrosis but also to the elucidation of the pathogeneses of various gastrointestinal diseases, including pancreatic diseases, and to the evaluation of therapeutic effects. The most important feature of EUS-EG is that it is a non-invasive modality. This is an advantage not found in EUS-guided fine needle aspiration (EUS-FNA), which has made remarkable progress in the field of diagnostics in recent years. Further development of artificial intelligence (AI) is expected to improve the diagnostic performance of EUS-EG. Future research on EUS-EG is anticipated.

Objective Assessment of Regional Stiffness in Vastus Lateralis with Different Measurement Methods: A Reliability Study

The objective of this study was to evaluate the reliability of four methods of assessing vastus lateralis (VL) stiffness, and to describe the influence of structural characteristics on them. The stiffness of the dominant lower-limb’s VL was evaluated in 53 healthy participants (28.4 ± 9.1 years) with shear wave elastography (SWE), strain elastography (SE), myotonometry and tensiomyography (TMG). The SWE, SE and myotonometry were performed at 50%, and TMG was assessed at 30%, of the length from the upper pole of the patella to the greater trochanter. The thickness of the VL, adipose tissue and superficial connective tissue was also measured with ultrasound. Three repeated measurements were acquired to assess reliability, using intraclass correlation coefficients (ICC). Pearson’s correlation coefficients were calculated to determine the relationships between methodologic assessments and between structural characteristics and stiffness assessments of the VL. Myotonometry (ICC = 0.93; 95%-CI = 0.89,0.96) and TMG (ICC = 0.89; 95%-CI = 0.82,0.94) showed excellent inter-day reliability whereas with SWE (ICC = 0.62; 95%-CI = 0.41,0.77) and SE (ICC = 0.71; 95%-CI = 0.57,0.81) reliability was moderate. Significant correlations were found between myotonometry and VL thickness (r = 0.361; p = 0.008), adipose tissue thickness (r = −0.459; p = 0.001) and superficial connective tissue thickness (r = 0.340; p = 0.013). Myotonometry and TMG showed the best reliability values, although myotonometry stiffness values were influenced by the structural variables of the supra-adjacent tissue.

Role of Endoscopic Ultrasound in the Diagnosis of Pancreatic Neuroendocrine Neoplasms

Although pancreatic neuroendocrine neoplasms (PNENs) are relatively rare tumors, their number is increasing with advances in diagnostic imaging modalities. Even small lesions that are difficult to detect using computed tomography or magnetic resonance imaging can now be detected with endoscopic ultrasound (EUS). Contrast-enhanced EUS is useful, and not only diagnosis but also malignancy detection has become possible by evaluating the vascularity of tumors. Pathological diagnosis using EUS with fine-needle aspiration (EUS-FNA) is useful when diagnostic imaging is difficult. EUS-FNA can also be used to evaluate the grade of malignancy. Pooling the data of the studies that compared the PNENs grading between EUS-FNA samples and surgical specimens showed a concordance rate of 77.5% (κ-statistic = 0.65, 95% confidence interval = 0.59–0.71, p < 0.01). Stratified analysis for small tumor size (2 cm) showed that the concordance rate was 84.5% and the kappa correlation index was 0.59 (95% confidence interval = 0.43–0.74, p < 0.01). The evolution of ultrasound imaging technologies such as contrast-enhanced and elastography and the artificial intelligence that analyzes them, the evolution of needles, and genetic analysis, will further develop the diagnosis and treatment of PNENs in the future.

Efficacy of Endoscopic Ultrasound Elastography in Differential Diagnosis of Gastrointestinal Stromal Tumor Versus Gastrointestinal Leiomyoma

BACKGROUND The diagnostic efficacy of endoscopic ultrasound (EUS) elastography for alimentary tract diseases remains uncertain. The aim of this study was to evaluate the utility of EUS elastography in differential diagnosis between the 2 most common subepithelium tumors of the digestive tract - gastrointestinal stromal tumors (GISTs) and gastrointestinal leiomyomas (GILs) - which cannot be differentiated by conventional EUS imaging. MATERIAL AND METHODS Electronic records were retrospectively reviewed from Jan 2015 to Jul 2019. Patients accepting EUS elastography with histopathological diagnosis of GISTs or GILs were included. The images of EUS elastography were analyzed by hue histogram in Photoshop. Hue values of RGB, R, G, and B channels of each group were acquired. We used the t test, ROC curve analysis, and binary logistic regression analysis for data post-processing. RESULTS We included 47 patients with GISTs and 14 with GILs. The mean±standard deviations (SD) of hue values were 20.25±0.72, -0.79±0.78, 20.79±1.68, 39.72±1.30 for GISTs and 20.80±0.46, 1.80±1.05, 28.39±2.15, and 31.95±2.60 for GILs of RGB, R, G, and B channels, respectively. The t test showed statistically significant differences in mean hue values between GISTs and GILs in B and G channels, but not in RGB and R channels. The area under the ROC curve combining B and G values was 0.723. Binary logistic regression analysis suggested no statistically significant difference in ability to differentiate between GISTs and GILs with B and G values (P>0.05). CONCLUSIONS There was insufficient evidence to support the application of quantitative EUS elastography for differential diagnosis of GISTs and GILs in this study.

Current status of artificial intelligence analysis for endoscopic ultrasonography

Endoscopic ultrasonography (EUS) is an essential diagnostic tool for various types of pancreatic diseases such as pancreatic tumors and chronic pancreatitis; however, EUS imaging has low specificity for the diagnosis of pancreatic diseases. Artificial intelligence (AI) is a mathematical prediction technique that automates learning and recognizes patterns in data. This review describes the details and principles of AI and deep learning algorithms. The term AI does not have any definite definition; almost all AI systems fall under narrow AI, which can handle single or limited tasks. Deep learning is based on neural networks, which is a machine learning technique that is widely used in the medical field. Deep learning involves three phases: data collection and annotation, building the deep learning architecture, and training and ability validation. For medical image diagnosis, image classification, object detection, and semantic segmentation are performed. In EUS, AI is used for detecting anatomical features, differential pancreatic tumors, and cysts. For this, conventional machine learning architectures are used, and deep learning architecture has been used in only two reports. Although the diagnostic abilities in these reports were about 85-95%, these were exploratory research and very few reports have included substantial evidence. AI is increasingly being used for medical image diagnosis due to its high performance and will soon become an essential technique for medical diagnosis.