An accurate diagnosis of solid pancreatic lesions (SPLs) is important because pancreatic cancer cannot be ignored if curative treatment is possible. Prompt and reliable diagnostic procedures are greatly needed for patients presenting with SPLs, particularly where resection is possible for a malignant mass. Several endoscopic ultrasound (EUS)-related technologies including a novel EUS-guided needle-based confocal laser endomicroscopy (EUS-nCLE) can provide real-time images at the cellular level (1,000-fold magnification). A 19-gauge EUS-guided fine needle aspiration (EUS-FNA) needle is recommended because its channel is large enough for the 0.85-mm diameter nCLE miniprobe. The procedure is performed by standard EUS-FNA techniques with either pre- or post-loading technique. Ten percent fluorescein sodium (2.5–5 mL) is used as an enhancing agent and is intravenously injected immediately before puncturing the lesion. Only a few studies have used the technique and reported results. A recent study from 19 malignant and 3 benign SPLs classified EUS-nCLE findings according to 4 signs: dark clumps, and dilated vessels (predominantly seen in malignant SPLs) and fine white fibrous bands and normal acini (predominantly seen in benign SPLs). Using these criteria, researchers correctly diagnosed 18 of the malignant SPLs (94.7%). Another study described 2 lesions as having “dark cells aggregates with pseudo-glandular aspects, and straight hyperdense elements more or less thick corresponding to tumoral fibrosis” in 17 of 18 malignant SPLs. Thus far, no large and systematic study has been performed to evaluate the potential clinical use of EUS-nCLE for diagnosing SPLs. However, based on available information from a few studies and the current limitations of EUS-FNA, EUS-nCLE can potentially provide a complementary role in diagnosing such lesions. Nevertheless, more studies are certainly needed.
Biopsy, Fine-Needle ; Diagnosis* ; Endoscopic Ultrasound-Guided Fine Needle Aspiration ; Fluorescein ; Humans ; Needles ; Pancreatic Neoplasms ; Ultrasonography

BACKGROUND/AIMS: Although flexible spectral imaging color enhancement (FICE) can facilitate the diagnosis of minimal change esophageal reflux disease (MERD), the complicated diagnostic criteria cause suboptimal inter-observer agreement. Confocal laser endomicroscopy (CLE) yields good diagnostic results but its inter-observer agreement has never been explored. This study compares the diagnostic value of magnifying FICE and probe-based CLE (pCLE) for MERD and evaluates the inter-observer agreement of both techniques. METHODS: Thirty-six patients with suspected MERD and 18 asymptomatic controls were recruited. Magnifying FICE was used for evaluation of distal esophagus. pCLE counted the number of intrapapillary capillary loops (IPCLs) using more than five IPCLs in 500×500 micron area as a criterion for MERD diagnosis. The validity scores and interobserever agreement of both FICE and pCLE were assessed. RESULTS: For FICE vs. pCLE, the accuracy was 79% vs. 87%, sensitivity 94% vs. 97%, specificity 50% vs. 66%, positive predictive value 79% vs. 85%, and negative predictive value 82% vs. 92%. Interobserver agreement of FICE was fair to substantial, whereas pCLE had substantial to almost perfect agreement. CONCLUSIONS: Both FICE and pCLE have good operating characteristics and can facilitate the MERD diagnosis. However, among different observers, pCLE is more consistent on MERD diagnosis.
Capillaries ; Diagnosis ; Esophagus ; Gastroesophageal Reflux* ; Humans ; Sensitivity and Specificity

BACKGROUND/AIMS: This study aimed to evaluate the diagnostic accuracy of dual-focus narrow-band imaging (dNBI) and Lugol'schromoendoscopy (LCE) combined with probe-based confocal laser endomicroscopy (pCLE) to screen for esophageal squamous cell neoplasms (ESCNs) in patients with a history of head and neck cancer. METHODS: From March to August 2016, dNBI was performed. Next, LCE was performed, followed by pCLE and biopsy. Histology has historically been the gold standard to diagnose ESCN. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of dNBI and LCE adjunct with pCLE were determined. RESULTS: Twenty-four patients were included. Ten ESCNs were found in 8 patients (33%). Forty percent of high-graded intraepithelial neoplasias and all low-grade intraepithelial neoplasias were overlooked by dNBI. The sensitivity, specificity, PPV, NPV, and accuracy of dNBI vs. LCE combined with pCLE were 50% vs. 80%, 62% vs. 67%, 36% vs. 44%, 75% vs. 91%, and 83% vs. 70%, respectively. CONCLUSIONS: The use of dNBI to detect ESCN was suboptimal. LCE with pCLE following dNBI had additional value for detecting esophageal dysplasia not detected by dNBI. The use of pCLE to detect dNBI-missed lesions yielded a high NPV, while pCLE-guided biopsy could reduce the number of unnecessary biopsies.
Biopsy ; Epithelial Cells ; Head and Neck Neoplasms ; Humans ; Neoplasms, Squamous Cell ; Sensitivity and Specificity

Background/Aims: Previous artificial intelligence (AI) models attempting to segment gastric intestinal metaplasia (GIM) areas have failed to be deployed in real-time endoscopy due to their slow inference speeds. Here, we propose a new GIM segmentation AI model with inference speeds faster than 25 frames per second that maintains a high level of accuracy. Methods: Investigators from Chulalongkorn University obtained 802 histological-proven GIM images for AI model training. Four strategies were proposed to improve the model accuracy. First, transfer learning was employed to the public colon datasets. Second, an image preprocessing technique contrast-limited adaptive histogram equalization was employed to produce clearer GIM areas. Third, data augmentation was applied for a more robust model. Lastly, the bilateral segmentation network model was applied to segment GIM areas in real time. The results were analyzed using different validity values. Results: From the internal test, our AI model achieved an inference speed of 31.53 frames per second. GIM detection showed sensitivity, specificity, positive predictive, negative predictive, accuracy, and mean intersection over union in GIM segmentation values of 93%, 80%, 82%, 92%, 87%, and 57%, respectively. Conclusions The bilateral segmentation network combined with transfer learning, contrast-limited adaptive histogram equalization, and data augmentation can provide high sensitivity and good accuracy for GIM detection and segmentation.

Background/Aims: The Thai Association for Gastrointestinal Endoscopy published recommendations on safe endoscopy during the coronavirus disease 2019 (COVID-19) pandemic. This study aimed to assess the practicality and applicability of the recommendations and the perceptions of endoscopy personnel on them. Methods: A validated questionnaire was sent to 1290 endoscopy personnel globally. Of these, the data of all 330 responders (25.6%) from 15 countries, related to the current recommendations on proper personal protective equipment (PPE), case selection, scope cleaning, and safety perception, were analyzed. Ordinal logistic regression was used to determine the relationships between the variables. Results: Despite an overwhelming agreement with the recommendations on PPE (94.5%) and case selection (95.5%), their practicality and applicability on PPE recommendations and case selection were significantly lower (p=0.001, p=0.047, p<0.001, and p=0.032, respectively). Factors that were associated with lower sense of safety in endoscopy units were younger age (p=0.004), less working experience (p=0.008), in-training status (p=0.04), and higher national prevalence of COVID-19 (p=0.003). High prevalent countries also had more difficulty implementing the guidelines (p<0.001) and they considered the PPE recommendations less practical and showed lower agreement with them (p<0.001 and p=0.008, respectively). A higher number of in-hospital COVID-19 patients was associated with less agreement with PPE recommendations (p=0.039). Conclusions Using appropriate PPE and case selection in endoscopic practice during a pandemic remains a challenge. Resource availability and local prevalence are critical factors influencing the adoption of the current guidelines.