Dan Ionuţ Gheonea

Neural network analysis of dynamic sequences of EUS elastography used for the differential diagnosis of chronic pancreatitis and pancreatic cancer

BACKGROUND EUS elastography is a newly developed imaging procedure that characterizes the differences of hardness and strain between diseased and normal tissue. OBJECTIVE To assess the accuracy of real-time EUS elastography in pancreatic lesions. DESIGN Cross-sectional feasibility study. PATIENTS The study group included, in total, 68 patients with normal pancreas (N = 22), chronic pancreatitis (N = 11), pancreatic adenocarcinoma (N = 32), and pancreatic neuroendocrine tumors (N = 3). A subgroup analysis of 43 cases with focal pancreatic masses was also performed. INTERVENTIONS A postprocessing software analysis was used to examine the EUS elastography movies by calculating hue histograms of each individual image, data that were further subjected to an extended neural network analysis to differentiate benign from malignant patterns. MAIN OUTCOME MEASUREMENTS To differentiate normal pancreas, chronic pancreatitis, pancreatic cancer, and neuroendocrine tumors. RESULTS Based on a cutoff of 175 for the mean hue histogram values recorded on the region of interest, the sensitivity, specificity, and accuracy of differentiation of benign and malignant masses were 91.4%, 87.9%, and 89.7%, respectively. The positive and negative predictive values were 88.9% and 90.6%, respectively. Multilayer perceptron neural networks with both one and two hidden layers of neurons (3-layer perceptron and 4-layer perceptron) were trained to learn how to classify cases as benign or malignant, and yielded an excellent testing performance of 95% on average, together with a high training performance that equaled 97% on average. LIMITATION A lack of the surgical standard in all cases. CONCLUSIONS EUS elastography is a promising method that allows characterization and differentiation of normal pancreas, chronic pancreatitis, and pancreatic cancer. The currently developed methodology, based on artificial neural network processing of EUS elastography digitalized movies, enabled an optimal prediction of the types of pancreatic lesions. Future multicentric, randomized studies with adequate power will have to establish the clinical impact of this procedure for the differential diagnosis of focal pancreatic masses.

Accuracy of endoscopic ultrasound elastography used for differential diagnosis of focal pancreatic masses: a multicenter study.

BACKGROUND AND STUDY AIMS Endoscopic ultrasound (EUS) elastography represents a new imaging procedure that might characterize the differences of hardness and strain between diseased tissue and normal tissue. The aim of this study was to assess the efficiency of EUS elastography for the differentiation of focal masses in chronic pancreatitis and pancreatic cancer. PATIENTS AND METHODS The study group comprised 258 patients with focal pancreatic masses included prospectively at 13 participating centers. Qualitative analysis of the diagnoses made by two expert doctors using all recorded video clips was performed in order to test the interobserver variability. A post-processing software analysis was used to examine the EUS elastography videos by calculating average-hue histograms of individual elastography images. The quantitative information was used to calculate intra-observer variability and the accuracy of the method. RESULTS Qualitative analysis of the recorded videos revealed a kappa value of 0.72. Intra-observer variability analysis revealed that the single measure intraclass correlation ranged between 0.86 and 0.94. The average-hue histogram analysis of the data indicated a sensitivity of 93.4 %, a specificity of 66.0 %, a positive predictive value of 92.5 %, a negative predictive value of 68.9 %, and an overall accuracy of 85.4 %, based on a cut-off value of 175. Area under the receiver operating characteristic curve (AUROC) was 0.854 ( P < 0.0001) with a confidence interval of 0.804 - 0.894. CONCLUSION The value of quantitative analysis of EUS elastography recordings was proven by good reproducibility of the videos, as well as good parameters of the AUROC analysis. (Clinical Trials.gov identifier: CT00909103).

Efficacy of an Artificial Neural Network–Based Approach to Endoscopic Ultrasound Elastography in Diagnosis of Focal Pancreatic Masses

BACKGROUND & AIMS By using strain assessment, real-time endoscopic ultrasound (EUS) elastography provides additional information about a lesions characteristics in the pancreas. We assessed the accuracy of real-time EUS elastography in focal pancreatic lesions using computer-aided diagnosis by artificial neural network analysis. METHODS We performed a prospective, blinded, multicentric study at of 258 patients (774 recordings from EUS elastography) who were diagnosed with chronic pancreatitis (n = 47) or pancreatic adenocarcinoma (n = 211) from 13 tertiary academic medical centers in Europe (the European EUS Elastography Multicentric Study Group). We used postprocessing software analysis to compute individual frames of elastography movies recorded by retrieving hue histogram data from a dynamic sequence of EUS elastography into a numeric matrix. The data then were analyzed in an extended neural network analysis, to automatically differentiate benign from malignant patterns. RESULTS The neural computing approach had 91.14% training accuracy (95% confidence interval [CI], 89.87%-92.42%) and 84.27% testing accuracy (95% CI, 83.09%-85.44%). These results were obtained using the 10-fold cross-validation technique. The statistical analysis of the classification process showed a sensitivity of 87.59%, a specificity of 82.94%, a positive predictive value of 96.25%, and a negative predictive value of 57.22%. Moreover, the corresponding area under the receiver operating characteristic curve was 0.94 (95% CI, 0.91%-0.97%), which was significantly higher than the values obtained by simple mean hue histogram analysis, for which the area under the receiver operating characteristic was 0.85. CONCLUSIONS Use of the artificial intelligence methodology via artificial neural networks supports the medical decision process, providing fast and accurate diagnoses.

Quantitative low mechanical index contrast-enhanced endoscopic ultrasound for the differential diagnosis of chronic pseudotumoral pancreatitis and pancreatic cancer

BackgroundSecond-generation intravenous blood-pool ultrasound contrast agents are increasingly used in endoscopic ultrasound (EUS) for characterization of microvascularization, differential diagnosis of benign and malignant focal lesions, as well as improved staging and guidance of therapeutic procedures.MethodsThe aim of our study was to prospectively compare the vascularisation patterns in chronic pseudotumoral pancreatitis and pancreatic cancer using quantitative low mechanical index (MI) contrast-enhanced EUS. We included 51 patients with chronic pseudotumoral pancreatitis (n = 19) and pancreatic cancer (n = 32). Perfusion imaging started with a bolus injection of Sonovue (2.4 ml), followed by analysis in the early arterial (wash-in) and late venous (wash-out) phase. Perfusion analysis was performed by post-processing of the raw data (time intensity curve [TIC] analysis). TIC analysis was performed inside the tumor and the pancreatic parenchyma, with depiction of the dynamic vascular pattern generated by specific software. Statistical analysis was performed on raw data extracted from the TIC analysis. Final diagnosis was based on a combination of EUS-FNA, surgery and follow-up of minimum 6 months in negative cases.ResultsThe sensitivity and specificity of low MI contrast enhanced EUS using TIC were sensitivity and specificity of low MI contrast enhanced EUS using TIC analysis were 93.75% (95% CI = 77.77 - 98.91%) and 89.47% (95% CI = 65.46 - 98.15%), respectively. Pseudotumoral chronic pancreatitis showed in the majority of cases a hypervascular appearance in the early arterial phase of contrast-enhancement, with a dynamic enhancement pattern similar with the rest of the parenchyma. Statistical analysis of the resulting series of individual intensities revealed no statistically relevant differences (p = .78). Pancreatic adenocarcinoma was usually a hypovascular lesion, showing low contrast-enhancement during the early arterial and also during the late venous phase of contrast-enhancement, also lower than the normal surrounding parenchyma. We found statistically significant differences in values during TIC analysis (p < .001).ConclusionsLow MI contrast enhanced EUS technique is expected to improve the differential diagnosis of focal pancreatic lesions. However, further multicentric randomized studies will confirm the exact role of the technique and its place in imaging assessment of focal pancreatic lesions.

Multicenter randomized controlled trial comparing the performance of 22 gauge versus 25 gauge EUS–FNA needles in solid masses

Abstract Background and aims. Few randomized studies have assessed the clinical performance of 25-gauge (25G) needles compared with 22-gauge (22G) needles during endoscopic ultrasound-guided fine needle aspiration (EUS–FNA) biopsy of intra-abdominal lesions. We aimed to compare the diagnostic yield, as well as performance characteristics of 22G versus 25G EUS biopsy needles by determining their diagnostic capabilities, the number of needle passes as well as cellularity of aspirated tissue specimen. Methods. The study is a prospective, randomized, multicenter study. Patients were referred between January 2009 and January 2010 for diagnostic EUS including EUS-guided FNA of different lesions adjacent to the upper GI tract. All patients were randomized to EUS–FNA performed with either a 22G or 25G aspiration needle. Results. EUS–FNA was performed in 135 patients (62 patients with a 22G needle). Sensitivity and specificity of the 22G needle was 94.1% and 95.8%, respectively, and for the 25G needle 94.1% and 100%, respectively. Investigators reported better visualization and performance for the 22G needle compared to the 25G (p < 0.0001). The number of tissue slides obtained was higher for the 22G needle during the second and third needle passes (p < 0.05). We did not observe significant differences between the number and preservation status of obtained cells (p > 0.05). Conclusions. A significant difference was found between the two types of needles in terms of reduced visualization of the 25G needle and suboptimal performance rating. However, this did not impact on overall results since both needles were equally successful in terms of a high diagnostic yield and overall accuracy.

Confocal laser endomicroscopy and immunoendoscopy for real-time assessment of vascularization in gastrointestinal malignancies

Gastrointestinal cancers represent a major cause of morbidity and mortality, with incomplete response to chemotherapy in the advanced stages and poor prognosis. Angiogenesis plays a crucial part in tumor growth and metastasis, with most gastrointestinal cancers depending strictly on the development of a new and devoted capillary network. Confocal laser endomicroscopy is a new technology which allows in vivo microscopic analysis of the gastrointestinal mucosa and its microvascularization during ongoing endoscopy by using topically or systemically administered contrast agents. Targeting markers of angiogenesis in association with confocal laser endomicroscopic examination (immunoendoscopy), as a future challenge, will add functional analysis to the morphological aspect of the neoplastic process. This review describes previous experience in endomicroscopic examination of the upper and lower digestive tract with emphasis on vascularization, resulting in a broad spectrum of potential clinical applications, and also preclinical research that could be translated to human studies.

EUS-assisted rendezvous stenting of the pancreatic duct for chronic calcifying pancreatitis with multiple pseudocysts.

Introduction: The best choice of endoscopic drainage of pancreatic pseudocysts complicating chronic pancreatitis is currently unknown, with EUS-guided transmural drainage competing with ERCP transpapillary techniques. However, recent studies currently recommend the use of both techniques in complex cases. Case Presentation: We present the case of a 60-year-old male patient with chronic calcifying pancreatitis, with severe ductal obstruction and multiple communicating pancreatic pseudocysts. The patient presented in the emergency department with weight loss, jaundice, steatorrhea and diabetes. Initial imaging evaluation (by transabdominal US, EUS and MRCP) depicted a dilated common bile duct, intrahepatic bile ducts and dilated main pancreatic duct (up to 1 cm) with multiple stones, as well as three pseudocysts at the level of the pancreatic head and one pseudocyst at the level of the pancreatic tail. ERCP with direct cannulation and transpapillary drainage of the bile duct or pancreatic duct was unsuccessful. Consequently, a EUS-assisted rendezvous stenting of the pancreatic duct was done, with the transpapillary placement of a 5-cm stent. Biliary cannulation was also possible with the placement of a double pigtail 9-cm stent in the common bile duct. Subsequent evolution was rapidly favorable with the disappearance of the pancreatic pseudocysts on the control CT after 24 h. Conclusion: Our case clearly showed the benefit of combined draining procedures even in cases of chronic pancreatitis with multiple pseudocysts where surgical drainage was previously deemed necessary.

Confocal Laser Endomicroscopy for the Morphometric Evaluation of Microvessels in Human Colorectal Cancer Using Targeted Anti-CD31 Antibodies

Introduction Numerous anti-angiogenic agents are currently developed to limit tumor growth and metastasis. While these drugs offer hope for cancer patients, their transient effect on tumor vasculature is difficult to assess in clinical settings. Confocal laser endomicroscopy (CLE) is a novel endoscopic imaging technology that enables histological examination of the gastrointestinal mucosa. The aim of the present study was to evaluate the feasibility of using CLE to image the vascular network in fresh biopsies of human colorectal tissue. For this purpose we have imaged normal and malignant biopsy tissue samples and compared the vascular network parameters obtained with CLE with established histopathology techniques. Materials and Methods Fresh non-fixed biopsy samples of both normal and malignant colorectal mucosa were stained with fluorescently labeled anti-CD31 antibodies and imaged by CLE using a dedicated endomicroscopy system. Corresponding biopsy samples underwent immunohistochemical staining for CD31, assessing the microvessel density (MVD) and vascular areas for comparison with CLE data, which were measured offline using specific software. Results The vessels were imaged by CLE in both normal and tumor samples. The average diameter of normal vessels was 8.5±0.9 µm whereas in tumor samples it was 13.5±0.7 µm (p = 0.0049). Vascular density was 188.7±24.9 vessels/mm2 in the normal tissue vs. 242.4±16.1 vessels/mm2 in the colorectal cancer samples (p = 0.1201). In the immunohistochemistry samples, the MVD was 211.2±42.9/mm2 and 351.3±39.6/mm2 for normal and malignant mucosa, respectively. The vascular area was 2.9±0.5% of total tissue area for the normal mucosa and 8.5±2.1% for primary colorectal cancer tissue. Conclusion Selective imaging of blood vessels with CLE is feasible in normal and tumor colorectal tissue by using fluorescently labeled antibodies targeted against an endothelial marker. The method could be translated into the clinical setting for monitoring of anti-angiogenic therapy.

Evaluation of new morphometric parameters of neoangiogenesis in human colorectal cancer using confocal laser endomicroscopy (CLE) and targeted panendothelial markers.

The tumor microcirculation is characterized by an abnormal vascular network with dilated, tortuous and saccular vessels. Therefore, imaging the tumor vasculature and determining its morphometric characteristics represent a critical goal for optimizing the cancer treatment that targets the blood vessels (i.e. antiangiogenesis therapy). The aim of this study was to evaluate new vascular morphometric parameters in colorectal cancer, difficult to achieve through conventional immunohistochemistry, by using the confocal laser endomicroscopy method. Fresh biopsies from tumor and normal tissue were collected during colonoscopy from five patients with T3 colorectal carcinoma without metastasis and were marked with fluorescently labeled anti-CD31 antibodies. A series of optical slices spanning 250 µm inside the tissue were immediately collected for each sample using a confocal laser endomicroscope. All measurements were expressed as the mean ± standard error. The mean diameter of tumor vessels was significantly larger than the normal vessels (9.46±0.4 µm vs. 7.60±0.3 µm, p = 0.0166). The vessel density was also significantly higher in the cancer vs. normal tissue samples (5541.05±262.81 vs. 3755.79±194.96 vessels/mm3, p = 0.0006). These results were confirmed by immunohistochemistry. In addition, the tortuosity index and vessel lengths were not significantly different (1.05±0.016 and 28.30±3.27 µm in normal tissue, vs. 1.07±0.008 and 26.49±3.18 µm in tumor tissue respectively, p = 0.5357 and p = 0.7033). The daughter/mother ratio (ratio of the sum of the squares of daughter vessel radii over the square of the mother vessel radius) was 1.15±0.09 in normal tissue, and 1.21±0.08 in tumor tissue (p = 0.6531). The confocal laser endomicroscopy is feasible for measuring more vascular parameters from fresh tumor biopsies than conventional immunohistochemistry alone. Provided new contrast agents will be clinically available, future in vivo use of CLE could lead to identification of novel biomarkers based on the morphometric characteristics of tumor vasculature.

Quantitative RT-PCR analysis of MMR genes on EUS-guided FNA samples from focal pancreatic lesions.

BACKGROUND/AIMS Pancreatic cancer is characterized by a variety of molecular alterations. Mismatch excision repair (MMR) is a DNA repair system that eliminates mismatched base pairs and it plays an important role in the maintaining of genomic integrity. The aim of the study was to assess the role of several MMR genes in the diagnosis of pancreatic cancer in samples collected by EUS-FNA procedure. METHODOLOGY The prospective study included 44 consecutive patients with pancreatic cancer (n=24) and chronic pseudotumoral pancreatitis (n=20). EUS-FNA was performed in all the patients. Gene analysis was performed by extracting the mRNA and by determining the expression of DNA repair genes (MLH1, MLH3, MSH6) using a standard algorithm. RESULTS Total RNA was successfully isolated from all the EUS-FNA pancreatic samples. We analyzed ROC curves to assess the significance of determining the expression of analyzed genes in EUS-FNA samples, obtaining a cutoff value of 476621mRNA copies/mL for MSH6, and sensitivity and specificity of 75.0% and 100%, respectively. For MLH1 and MLH3, sensitivity and specificity were only satisfactory (64.65% and 76.11%, and 75.0% and 63.64%, respectively). CONCLUSIONS The quality and amount of cellular sampling using pancreatic EUS-FNA allow the extraction of sufficient quantities of RNA to perform qRT-PCR analysis. The use of MMR genes for the differentiation between pseudotumoral chronic pancreatitis and pancreatic cancer using a minimally invasive sampling technique could be a promising technique.