- Browse by Subject
Browsing by Subject "Endosonography"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item Improving the diagnostic accuracy of endoscopic ultrasound-guided fine-needle aspiration using microRNAs(ScienceDirect, 2014-10) Sina, Marsela; Cote, Gregory A.; Korc, Murray; Department of Medicine, IU School of MedicineComment on: A microRNA-based test improves endoscopic ultrasound-guided cytologic diagnosis of pancreatic cancer.Item Real-time computer-aided diagnosis of focal pancreatic masses from endoscopic ultrasound imaging based on a hybrid convolutional and long short-term memory neural network model(PLOS, 2021-06-28) Udriştoiu, Anca Loredana; Cazacu, Irina Mihaela; Gruionu, Lucian Gheorghe; Gruionu, Gabriel; Iacob, Andreea Valentina; Burtea, Daniela Elena; Ungureanu, Bogdan Silviu; Costache, Mădălin Ionuţ; Constantin, Alina; Popescu, Carmen Florina; Udriştoiu, Ştefan; Săftoiu, Adrian; Medicine, School of MedicineDifferential diagnosis of focal pancreatic masses is based on endoscopic ultrasound (EUS) guided fine needle aspiration biopsy (EUS-FNA/FNB). Several imaging techniques (i.e. gray-scale, color Doppler, contrast-enhancement and elastography) are used for differential diagnosis. However, diagnosis remains highly operator dependent. To address this problem, machine learning algorithms (MLA) can generate an automatic computer-aided diagnosis (CAD) by analyzing a large number of clinical images in real-time. We aimed to develop a MLA to characterize focal pancreatic masses during the EUS procedure. The study included 65 patients with focal pancreatic masses, with 20 EUS images selected from each patient (grayscale, color Doppler, arterial and venous phase contrast-enhancement and elastography). Images were classified based on cytopathology exam as: chronic pseudotumoral pancreatitis (CPP), neuroendocrine tumor (PNET) and ductal adenocarcinoma (PDAC). The MLA is based on a deep learning method which combines convolutional (CNN) and long short-term memory (LSTM) neural networks. 2688 images were used for training and 672 images for testing the deep learning models. The CNN was developed to identify the discriminative features of images, while a LSTM neural network was used to extract the dependencies between images. The model predicted the clinical diagnosis with an area under curve index of 0.98 and an overall accuracy of 98.26%. The negative (NPV) and positive (PPV) predictive values and the corresponding 95% confidential intervals (CI) are 96.7%, [94.5, 98.9] and 98.1%, [96.81, 99.4] for PDAC, 96.5%, [94.1, 98.8], and 99.7%, [99.3, 100] for CPP, and 98.9%, [97.5, 100] and 98.3%, [97.1, 99.4] for PNET. Following further validation on a independent test cohort, this method could become an efficient CAD tool to differentiate focal pancreatic masses in real-time.Item Role of endoscopic ultrasound fine-needle aspiration evaluating adrenal gland enlargement or mass(Baishideng Publishing Group Inc., 2014-08-06) Martinez, Melissa; LeBlanc, Julia; Al-Haddad, Mohammad; Sherman, Stuart; DeWitt, John; Department of Medicine, IU School of MedicineAIM: To report the clinical impact of adrenal endoscopic ultrasound fine-needle aspiration (EUS-FNA) in the evaluation of patients with adrenal gland enlargement or mass. METHODS: In a retrospective single-center case-series, patients undergoing EUS-FNA of either adrenal gland from 1997-2011 in our tertiary care center were included. Medical records were reviewed and results of EUS, cytology, adrenal size change on follow-up imaging ≥ 6 mo after EUS and any repeat EUS or surgery were abstracted. A lesion was considered benign if: (1) EUS-FNA cytology was benign and the lesion remained < 1 cm from its original size on follow-up computed tomography (CT), magnetic resonance imaging or repeat EUS ≥ 6 mo after EUS-FNA; or (2) subsequent adrenalectomy and surgical pathology was benign. RESULTS: Ninety-four patients had left (n = 90) and/or right (n = 5) adrenal EUS-FNA without adverse events. EUS indications included: cancer staging or suspected recurrence (n = 31), pancreatic (n = 20), mediastinal (n = 10), adrenal (n = 7), lung (n = 7) mass or other indication (n = 19). Diagnoses after adrenal EUS-FNA included metastatic lung (n = 10), esophageal (n= 5), colon (n = 2), or other cancer (n = 8); benign primary adrenal mass or benign tissue (n = 60); or was non-diagnostic (n = 9). Available follow-up confirmed a benign lesion in 5/9 non-diagnostic aspirates and 32/60 benign aspirates. Four of the 60 benign aspirates were later confirmed as malignant by repeat biopsy, follow-up CT, or adrenalectomy. Adrenal EUS-FNA diagnosed metastatic cancer in 24, and ruled out metastasis in 10 patients. For the diagnosis of malignancy, EUS-FNA of either adrenal had sensitivity, specificity, positive predictive value and negative predictive value of 86%, 97%, 96% and 89%, respectively. CONCLUSION: Adrenal gland EUS-FNA is safe, minimally invasive and a sensitive technique with significant impact in the management of adrenal gland mass or enlargement.Item Utility of EUS following endoscopic polypectomy of high-risk rectosigmoid lesions(Wolters Kluwer, 2015-04) Luz, Leticia P.; Cote, Gregory A.; Al-Haddad, Mohammad Ali; McHenry, Lee; LeBlanc, Julia K.; Sherman, Stuart; Moreira, Daniel M.; El Hajj, Ihab I.; McGreevy, Kathleen; DeWitt, John; Department of Medicine, IU School of MedicineBACKGROUND: The utility of endoscopic ultrasound (EUS) compared with standard white light endoscopy (WLE) following recent polypectomy of high-risk colorectal polyps is unknown. OBJECTIVE: To assess the incremental yield of EUS after endoscopic polypectomy of a high-risk rectal lesion. DESIGN: Retrospective cohort. SETTING: Tertiary referral center. MATERIALS AND METHODS: Patients referred for EUS following attempted endoscopic resection of a high-risk rectal neoplasm, defined as a tubulovillous adenoma, tubular adenoma with high-grade dysplasia, carcinoid, carcinoma in-situ or adenocarcinoma (CA). INTERVENTIONS: Sigmoidoscopy ± mucosal biopsy and EUS ± fine-needle aspiration (FNA) to evaluate for: (1) Residual polyp/tumor in the rectal wall or (2) peritumoral adenopathy. MAIN OUTCOME: Sensitivity and specificity for detection of residual neoplasia for WLE ± biopsy (WLE/BX) and EUS ± FNA for cancer (CA group) or benign disease (non-CA group). The incremental yield of EUS defined as: (1) Residual intramural neoplasia not present on WLE ± BX and; (2) abnormal peritumoral adenopathy. RESULTS: A total of 70 patients (mean age 64 ± 11 years, 61% male) with a final diagnosis of CA (n = 38) and non-CA (n = 32) were identified. There was no difference between the sensitivity and specificity of WLE alone (65% and 84%), WLE with biopsy (71% and 95%), and EUS (59% and 84%), for the detection of residual neoplasia (P > 0.05 for all). EUS identified 3 masses missed by WLE, all in the CA group. A malignant (n = 2) or benign (n = 3) node was identified in 5 (13%) CA patients; EUS-FNA in two showed residual malignancy in one and a reactive lymph node (LN) in one. No LNs were identified in the non-CA patients. LIMITATIONS: Retrospective design, incomplete follow-up in some patients. CONCLUSION: Following endoscopic polypectomy of high-risk rectal neoplasia, the incremental yield of EUS compared with WLE/BX for evaluation of residual disease appears limited, especially in patients with benign disease.