Feasibility and efficacy of endoscopy with blue laser imaging for the detection and diagnosis of cardia polyps: A single-center randomized controlled study.

OBJECTIVE: To compare the detection rate and diagnostic accuracy of cardia polyps using endoscopy with blue laser imaging (BLI) and white-light imaging (WLI). METHODS: Patients were randomly divided into the BLI group and WLI group according to the endoscopic procedures. BLI followed by WLI was conducted in the BLI group, whereas WLI followed by BLI examination was conducted in the WLI group. The number, size, microstructure, and microvascular patterns of cardia polyps detected were recorded. Biopsy of the polyps was then performed. RESULTS: The detection rate of cardia polyps in the BLI group was higher than that in the WLI group (7.87% vs 4.22%, P = 0.018). The rate of overlooked lesions in the BLI group was lower than in the WLI group (0.64% vs 3.38%, P = 0.003). The diagnostic coincidence rate between magnifying BLI and histopathology was 88.16%. The sensitivity, specificity, positive predictive value and negative predictive value for the diagnosis of neoplastic lesions by magnifying endoscopy with BLI were 90.91%, 87.69%, 55.56%, and 98.28%, respectively. The most remarkable patterns for predicting inflammatory polyps were the prolonged and fine network patterns (sensitivity 71.43%, specificity 93.75%). Small round combined with honeycomb patterns were the most common among fundic gland polyps (sensitivity 80.00%, specificity 98.48%). Neoplastic lesions presented as villous or ridge-like combined with core vascular or unclear pattern for both microvascular and microstructure patterns. CONCLUSION: BLI is more effective than WLI in the detection and diagnosis of cardia polyps, and magnifying endoscopy with BLI may help diagnose such lesions.

Analysis of the detection rate and clinical characteristics of early gastric cancer by painless gastroscopy and ordinary gastroscopy.

OBJECTIVE: To investigate the difference of early gastric cancer (EGC) detection rate and endoscopic characteristics between painless and ordinary electronic gastroscopy, and summarize the clinical data of gastric cancer (GC) patients. METHODS: Clinical data of 72,000 patients who underwent gastroscopy in the First People Hospital of Huzhou (Zhejiang, China) from January 2016 to December 2021 were retrospectively analyzed. The patients were divided into painless gastroscopy group (observation group, 36,000 cases) and ordinary gastroscopy group (control group, 36,000 cases) according to the examination methods. The detection rate of EGC between the 2 groups and the endoscopic characteristics of EGC lesions between the 2 groups were compared, and the clinical data of GC were summarized. RESULTS: Painless gastroscopy is safer than ordinary gastroscopy. The detection rate of GC and EGC in the observation group was significantly higher than that in the control group (P < .05); the difference between the 2 groups in the detection rate of advanced GC was not statistically significant. The average length of EGC lesions in the observation group was significantly shorter than that in the control group (P < .05). The proportion of EGC with lesion length <2.0 cm in the observation group was significantly higher than that in the control group (P < .05). The proportion of EGC lesions with type II morphology, normal or pallor mucosal color, and no rupture in mucosa in the control group were significantly lower than that in the observation group, respectively (P < .05). The proportion of EGC distributed in the cardia, fundus and corpus was higher in the observation group than in the control group (P < .05). The incidence of helicobacter pylori (HP) infection, precancerous diseases, first-degree relatives of GC patients, and risk factors in patients with GC was significantly higher than that in non-GC patients (P < .05), multivariate logistic regression analysis showed that these were independent influencing factors for the occurrence of GC. CONCLUSION: Painless gastroscopy can effectively improve the screening and diagnostic efficiency of EGC, especially for EGC lesions that are not easy to expose the site, small in size, superficial, without obvious mucosal color change or without mucosal breakage. Therefore, the value of painless gastroscopy in EGC screening is worth further promotion and research.

The principles of enhanced recovery after percutaneous endoscopic gastrostomy (ERaPEG): a UK tertiary center experience.

PURPOSE: Enhanced recovery after surgery (ERAS) pathways have been shown to improve surgical outcomes and patient satisfaction. The aim of the study was to assess whether the implementation of a perioperative enhanced recovery after percutaneous endoscopic gastrostomy (ERaPEG) pathway based on ERAS principles was safe, satisfactory to parents and improved outcomes. METHODS: Following a quality improvement project, a multimodal ERaPEG pathway was introduced as standard practice within the department and children undergoing elective same-day admission percutaneous endoscopic gastrostomy (PEG) at a single UK tertiary center were prospectively enrolled. Exclusion criteria were patients undergoing other concurrent procedures and those who underwent a laparoscopic assisted/open procedure. Data included patient demographics, underlying diagnosis, indication, length of stay (LOS) and 30-day readmission. Parental experience and satisfaction were determined using a questionnaire including 5-point Likert scales. A retrospective cohort was used for comparison. Data were analyzed using Chi-Square test and Mann-Whitney U tests. RESULTS: Ninety-five patients met the inclusion criteria: 50 pre and 45 post the implementation of ERaPEG. Median age was 3 and 2 years, respectively. Neurodisability was the underlying diagnosis in most patients (84%-pre-ERaPEG; 76%-post-ERaPEG). Most common PEG indication was medication/nutritional supplementation (52%-pre-ERaPEG; 51%-post-ERaPEG). The LOS significantly decreased from a median of 51.5 h (pre-ERaPEG) to 32 h (post-ERaPEG) (p < 0.001). Thirty-day readmission rates were similar (6% vs 11%). Most parents felt that the educational material was easy to access and understand. Post-operatively the majority of parents (≥ 80%) were confident in managing the gastrostomy device, setting up/giving the feeds and also felt that the LOS was appropriate. CONCLUSION: This study shows that the implementation of an ERaPEG pathway significantly reduced LOS following PEG. In addition, the pathway was satisfactory to parents and offered the benefit of improved resource utilization.

Safe and effective hybrid endoscopic submucosal dissection with ALL IN ONE snare in porcine gastric model (with video).

This study aimed to evaluate the safety and efficiency of hybrid endoscopic submucosal dissection (H-ESD) using a newly developed ALL IN ONE (AIO) snare. This was a matched control study in a porcine model. Five paired simulated stomach lesions 2-2.5 cm in size were removed by H-ESD using an AIO snare or conventional ESD (C-ESD) using an endoscopic knife. The outcomes of the two procedures were compared, including en-bloc resection rates, procedure times, intraprocedural bleeding volumes, muscular injuries, perforations, thicknesses of the submucosal layer in resected specimens, and stomach defects. All simulated lesions were resected en-bloc. Specimens resected by H-ESD and C-ESD were similar in size (7.68 ± 2.92 vs. 8.42 ± 2.42 cm2; P = 0.676). H-ESD required a significantly shorter procedure time (13.39 ± 3.78 vs. 25.99 ± 4.52 min; P = 0.031) and submucosal dissection time (3.99 ± 1.73 vs. 13.1 ± 4.58 min; P = 0.003) versus C-ESD; H-ESD also yielded a faster dissection speed (241.37 ± 156.84 vs. 68.56 ± 28.53 mm2/min; P = 0.042) and caused fewer intraprocedural bleeding events (0.40 ± 0.55 vs. 3.40 ± 1.95 times/per lesion; P = 0.016) than C-ESD. The thicknesses of the submucosal layer of the resected specimen (1190.98 ± 134.07 vs. 1055.90 ± 151.76 µm; P = 0.174) and the residual submucosal layer of the stomach defect (1607.94 ± 1026.74 vs. 985.98 ± 445.58 µm; P = 0.249) were similar with both procedures. The AIO snare is a safe and effective device for H-ESD and improves the treatment outcomes of gastric lesions by shortening the procedure time.

Machine learning models to predict submucosal invasion in early gastric cancer based on endoscopy features and standardized color metrics.

Conventional endoscopy is widely used in the diagnosis of early gastric cancers (EGCs), but the graphical features were loosely defined and dependent on endoscopists' experience. We aim to establish a more accurate predictive model for infiltration depth of early gastric cancer including a standardized colorimetric system, which demonstrates promising clinical implication. A retrospective study of 718 EGC cases was performed. Clinical and pathological characteristics were included, and Commission Internationale de l'Eclariage (CIE) standard colorimetric system was used to evaluate the chromaticity of lesions. The predicting models were established in the derivation set using multivariate backward stepwise logistic regression, decision tree model, and random forest model. Logistic regression shows location, macroscopic type, length, marked margin elevation, WLI color difference and histological type are factors significantly independently associated with infiltration depth. In the decision tree model, margin elevation, lesion located in the lower 1/3 part, WLI a*color value, b*color value, and abnormal thickness in enhanced CT were selected, which achieved an AUROC of 0.810. A random forest model was established presenting the importance of each feature with an accuracy of 0.80, and an AUROC of 0.844. Quantified color metrics can improve the diagnostic precision in the invasion depth of EGC. We have developed a nomogram model using logistic regression and machine learning algorithms were also explored, which turned out to be helpful in decision-making progress.

Optimization of tracheoesophageal fistula model established with T-shaped magnet system based on magnetic compression technique.

BACKGROUND: The magnetic compression technique has been used to establish an animal model of tracheoesophageal fistula (TEF), but the commonly shaped magnets present limitations of poor homogeneity of TEF and poor model control. We designed a T-shaped magnet system to overcome these problems and verified its effectiveness via animal experiments. AIM: To investigate the effectiveness of a T-shaped magnet system for establishing a TEF model in beagle dogs. METHODS: Twelve beagles were randomly assigned to groups in which magnets of the T-shaped scheme (study group, n = 6) or normal magnets (control group, n = 6) were implanted into the trachea and esophagus separately under gastroscopy. Operation time, operation success rate, and accidental injury were recorded. After operation, the presence and timing of cough and the time of magnet shedding were observed. Dogs in the control group were euthanized after X-ray and gastroscopy to confirm establishment of TEFs after coughing, and gross specimens of TEFs were obtained. Dogs in the study group were euthanized after X-ray and gastroscopy 2 wk after surgery, and gross specimens were obtained. Fistula size was measured in all animals, and then harvested fistula specimens were examined by hematoxylin and eosin (HE) and Masson trichrome staining. RESULTS: The operation success rate was 100% for both groups. Operation time did not differ between the study group (5.25 min ± 1.29 min) and the control group (4.75 min ± 1.70 min; P = 0.331). No bleeding, perforation, or unplanned magnet attraction occurred in any animal during the operation. In the early postoperative period, all dogs ate freely and were generally in good condition. Dogs in the control group had severe cough after drinking water at 6-9 d after surgery. X-ray indicated that the magnets had entered the stomach, and gastroscopy showed TEF formation. Gross specimens of TEFs from the control group showed the formation of fistulas with a diameter of 4.94 mm ± 1.29 mm (range, 3.52-6.56 mm). HE and Masson trichrome staining showed scar tissue formation and hierarchical structural disorder at the fistulas. Dogs in the study group did not exhibit obvious coughing after surgery. X-ray examination 2 wk after surgery indicated fixed magnet positioning, and gastroscopy showed no change in magnet positioning. The magnets were removed using a snare under endoscopy, and TEF was observed. Gross specimens showed well-formed fistulas with a diameter of 6.11 mm ± 0.16 mm (range, 5.92-6.36 mm), which exceeded that in the control group (P < 0.001). Scar formation was observed on the internal surface of fistulas by HE and Masson trichrome staining, and the structure was more regular than that in the control group. CONCLUSION: Use of the modified T-shaped magnet scheme is safe and feasible for establishing TEF and can achieve a more stable and uniform fistula size compared with ordinary magnets. Most importantly, this model offers better controllability, which improves the flexibility of follow-up studies.

[Therapeutic efficacy analysis of endoscopic combined with serological diagnosis strategy and endoscopic in G1 and G2 gastric neuroendocrine neoplasms].

Objective: To investigate the endoscopic combined serological diagnosis strategy for G1 and G2 gastric neuroendocrine neoplasms (G-NENs), and to evaluate the safety, short-term, and long-term efficacy of two endoscopic treatment procedures: endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD). Methods: This study retrospectively analyzed the clinical data of 100 consecutive patients with G-NENs who were hospitalized at the Cancer Hospital of the Chinese Academy of Medical Sciences from January 2011 to October 2023. These patients underwent endoscopic treatment, and propensity score matching (PSM) was used to compare clinicopathological characteristics, as well as short-term and long-term efficacy of lesions in the EMR group and ESD group before and after treatment. Results: Among the 100 patients with G-NENs, the median age was 54 years old. Before surgery, 29 cases underwent endoscopic combined serological examination, and 24 of them (82.2%) had abnormally elevated plasma chromogranin A. The combined diagnostic strategy for autoimmune atrophic gastritis (AIG) achieved a diagnostic accuracy of 100%(22/22). A total of 235 G-NEN lesions were included, with 84 in the ESD group and 151 in the EMR group. The median size of the lesions in the ESD group (5.0 mm) was significantly larger than that in the EMR group (2.0 mm, P＜0.001). Additionally, the ESD group had significantly more lesions with pathological grade G2[23.8%(20/84) vs. 1.3%(2/151), P＜0.001], infiltration depth reaching the submucosal layer [78.6%(66/84) vs. 51.0%(77/151), P＜0.001], and more T2 stage compared to the EMR group[15.5%(13/84) vs. 0.7%(1/151), P＜0.001]. After PSM, 49 pairs of lesions were successfully matched between the two groups. Following PSM, there were no significant differences in the en bloc resection rate [100.0%(49/49) vs. 100.0%(49/49)], complete resection rate [93.9%(46/49) vs. 100.0%(49/49)], and complication rate [0(0/49) vs. 4.1%(2/49)] between the two groups. During the follow-up period, no recurrence or distant metastasis was observed in any of the lesions in both groups. Conclusions: The combination of endoscopy and serology diagnostic strategy has the potential to enhance the accuracy of diagnosing G1 and G2 stage G-NENs and their background mucosa. Endoscopic resection surgery (EMR, ESD) is a proven and safe treatment approach for G1 and G2 stage G-NENs.

Characteristics of early gastric tumors with different differentiation and predictors of long-term outcomes after endoscopic submucosal dissection.

BACKGROUND: Gastric cancer is a common malignant tumor of the digestive tract, and endoscopic submucosal dissection (ESD) is the preferred treatment for early-stage gastric cancer. The analysis of the epidemiological characteristics of gastric mucosal tumors with different differentiation degrees and the influencing factors of long-term ESD efficacy may have certain significance for revealing the development of gastric cancer and ESD. AIM: To analyze the features of gastric mucosal tumors at different differentiation levels, and to explore the prognostic factors of ESD. METHODS: We retrospectively studied 301 lesions in 285 patients at The Second Affiliated Hospital of Xi'an Jiaotong University from 2014 to 2021, according to the latest Japanese guidelines (sixth edition), and divided them into low-grade intraepithelial neoplasia (LGIN), high-grade intraepithelial neoplasia (HGIN), and differentiated and undifferentiated early carcinoma. They are followed up by endoscopy, chest and abdominal computed tomography at 3, 6 and 12 months after ESD. We compared clinicopathologic characteristics, ESD efficacy, and complications with different degrees of differentiation, and analyzed the related factors associated with ESD. RESULTS: HGIN and differentiated carcinoma patients were significantly older compared with LGIN patients (P < 0.001) and accounted for more 0-IIc (P < 0.001), atrophic gastritis was common (P < 0.001), and irregular microvascular patterns (IMVPs) and demarcation lines (DLs) were more obvious (P < 0.001). There was more infiltration in the undifferentiated carcinoma tissue (P < 0.001), more abnormal folds and poorer mucosal peristalsis (P < 0.001), and more obvious IMVPs, irregular microsurface patterns and DLs (P < 0.05) than in the LGIN and HGIN tissues. The disease-free survival rates at 2, 5, and 8 years after ESD were 95.0%, 90.1%, and 86.9%, respectively. Undifferentiated lesions (HR 5.066), white moss (HR 7.187), incomplete resection (HR 3.658), and multiple primary cancers (HR 2.462) were significantly associated with poor prognosis. CONCLUSION: Differentiations of gastric mucosal tumors have different epidemiological and endoscopic characteristics, which are closely related to the safety and efficacy of ESD.

Comparative Analysis of Enbloc or Piecemeal Removal After Enbloc Resection of Gastrointestinal Stromal Tumors.

OBJECTIVE: To investigate the safety and prognosis of enbloc or piecemeal removal after enbloc resection of a gastric GIST by comparing the clinical data of endoscopic en block resection and piecemeal removal (EP) and en block resection and complete removal (EC) of gastric GISTs. METHODS: A total of 111 (43 endoscopic piecemeal, and 68 complete removal) patients with gastric GIST's ≥ 2 cm in diameter who underwent endoscopic therapy from January 2016 to June 2020 at the First Affiliated Hospital of Zhengzhou University were retrospectively analyzed. In all cases, it was ensured that the tumor was intact during the resection, however, it was divided into EP group and EC group based on whether the tumor was completely removed or was cut into pieces which were then removed. The patients' recurrence-free survival rate and recurrence-free survival (RFS) were recorded. RESULTS: There was no statistically significant difference in RFS rates between the two groups (P = 0.197). The EP group had relatively high patient age, tumor diameter, risk classification, and operation time. However, there was no statistically significant difference in the number of nuclear fission images, postoperative hospitalization time, postoperative fasting time, complication rate and complication grading between the two groups (P > 0.05). CONCLUSION: Endoscopic piecemeal removal after en block resection of gastric GIST is safe and effective and achieves similar clinical outcomes as complete removal after en block resection.

Precutting Endoscopic Band Ligation-Assisted Resection Is Safe and Effective for Treating Gastric Submucosal Tumors from the Muscularis Propria.

BACKGROUND: We previously treated small gastric submucosal tumors originating from the muscularis propria layer by precutting endoscopic band ligation but lacked precise pathological results. Then, precutting endoscopic band ligation was modified by additional snare resection after ligation to obtain tumor specimens, termed precutting endoscopic band ligation-assisted resection. AIMS: In this study, we aimed to explore the safety, feasibility, and efficacy of precutting endoscopic band ligation-assisted resection. METHODS: From 2021 to 2022, a total of 16 consecutive patients underwent precutting endoscopic band ligation-assisted resection to treat small gastric submucosal tumors originating from the muscularis propria. The clinical demography, perioperative data, and follow-up outcomes were retrospectively collected. RESULTS: With a mean operative time of 21.3 min, all lesions were successfully and completely resected, and no severe adverse events or local recurrences occurred postoperatively. More importantly, en bloc and R0 resection were achieved in all 16 patients. CONCLUSION: Precutting endoscopic band ligation-assisted resection is a safe, effective, and time-saving endoscopic technique for managing gastric small gastric submucosal tumors originating from the muscularis propria for both diagnosis and eradication.

Endoscopic resection and laparoscopic lymph node dissection for early gastric cancer beyond conventional endoscopic treatment indications: a 10-year outcome study.

BACKGROUND: Endoscopic full-thickness gastric resection (EFTGR) with laparoscopic regional lymph node dissection (LLND) and endoscopic submucosal dissection (ESD) with LLND have been investigated as treatment options for early gastric cancer beyond the absolute indications for ESD. However, comparative studies on the long-term outcomes of these procedures are lacking. This study aimed to analyze and compare the 10-year outcomes of both procedures in a real clinical setting. METHODS: Between January 2009 and December 2013, 28 and 37 patients diagnosed with EGC beyond the absolute indications for ESD were treated with EFTGR with LLND and ESD with LLND, respectively. In both procedures, the dye was injected into the tumor. However, after injection and LLND, EFTGR was performed immediately in the EFTGR with LLND group, whereas LLND was followed by ESD in the ESD with LLND group. The primary endpoint was the 10-year survival rate. RESULTS: The EFTGR with LLND group had one case of local recurrence (3.6%) and mortality (3.6%) each, while the ESD with LLND group had none (0.0% for both); however, the differences were not statistically significant (P = 0.247 for each). Furthermore, there was no significant difference in complications such as ischemia and anastomosis leakage between the groups (P = 0.247). CONCLUSIONS: When the procedures were properly applied, EFTGR with LLND and ESD with LLND did not increase the 10-year mortality in patients with EGC beyond the absolute ESD indications compared with conventional radical gastrectomy.

Endoscopic features and treatments of gastric cystica profunda.

BACKGROUND: Gastric cystica profunda (GCP) represents a rare condition characterized by cystic dilation of gastric glands within the mucosal and/or submucosal layers. GCP is often linked to, or may progress into, early gastric cancer (EGC). AIM: To provide a comprehensive evaluation of the endoscopic features of GCP while assessing the efficacy of endoscopic treatment, thereby offering guidance for diagnosis and treatment. METHODS: This retrospective study involved 104 patients with GCP who underwent endoscopic resection. Alongside demographic and clinical data, regular patient follow-ups were conducted to assess local recurrence. RESULTS: Among the 104 patients diagnosed with GCP who underwent endoscopic resection, 12.5% had a history of previous gastric procedures. The primary site predominantly affected was the cardia (38.5%, n = 40). GCP commonly exhibited intraluminal growth (99%), regular presentation (74.0%), and ulcerative mucosa (61.5%). The leading endoscopic feature was the mucosal lesion type (59.6%, n = 62). The average maximum diameter was 20.9 ± 15.3 mm, with mucosal involvement in 60.6% (n = 63). Procedures lasted 73.9 ± 57.5 min, achieving complete resection in 91.3% (n = 95). Recurrence (4.8%) was managed via either surgical intervention (n = 1) or through endoscopic resection (n = 4). Final pathology confirmed that 59.6% of GCP cases were associated with EGC. Univariate analysis indicated that elderly males were more susceptible to GCP associated with EGC. Conversely, multivariate analysis identified lesion morphology and endoscopic features as significant risk factors. Survival analysis demonstrated no statistically significant difference in recurrence between GCP with and without EGC (P = 0.72). CONCLUSION: The findings suggested that endoscopic resection might serve as an effective and minimally invasive treatment for GCP with or without EGC.

Computer-aided diagnosis in real-time endoscopy for all stages of gastric carcinogenesis: Development and validation study.

OBJECTIVE: Using endoscopic images, we have previously developed computer-aided diagnosis models to predict the histopathology of gastric neoplasms. However, no model that categorizes every stage of gastric carcinogenesis has been published. In this study, a deep-learning-based diagnosis model was developed and validated to automatically classify all stages of gastric carcinogenesis, including atrophy and intestinal metaplasia, in endoscopy images. DESIGN: A total of 18,701 endoscopic images were collected retrospectively and randomly divided into train, validation, and internal-test datasets in an 8:1:1 ratio. The primary outcome was lesion-classification accuracy in six categories: normal/atrophy/intestinal metaplasia/dysplasia/early /advanced gastric cancer. External-validation of performance in the established model used 1427 novel images from other institutions that were not used in training, validation, or internal-tests. RESULTS: The internal-test lesion-classification accuracy was 91.2% (95% confidence interval: 89.9%-92.5%). For performance validation, the established model achieved an accuracy of 82.3% (80.3%-84.3%). The external-test per-class receiver operating characteristic in the diagnosis of atrophy and intestinal metaplasia was 93.4 ± 0% and 91.3 ± 0%, respectively. CONCLUSIONS: The established model demonstrated high performance in the diagnosis of preneoplastic lesions (atrophy and intestinal metaplasia) as well as gastric neoplasms.

Assessing wound complications in gastroscopy with Streptomyces protease enzyme combined with Shutai.

Current gastroscopy practices necessitate a balance between procedural efficiency and patient safety. It has been hypothesized that increasing procedure outcomes through the use of Streptomyces protease enzyme and Shutai is possible; however, precise nature of any potential adverse reactions and complications remains unknown. In Zhanjiang, China, 213 patients undergoing gastroscopy participated in this controlled trial. The subjects were allocated at random into two groups: control and treatment. The treatment group was administered topical Streptomyces protease enzyme and intravenous Shutai. Using chi-square and t-tests, information regarding patient demographics, adverse reactions, wound healing, procedure duration, distress levels, and satisfaction was gathered and analysed. The demographic and medical history characteristics of the groups were comparable. There was a greater prevalence of modest immediate reactions in the treatment group (p < 0.05), whereas there were no significant variations observed in delayed reactions and long-term complications (p > 0.05). The treatment group exhibited superior efficiency metrics, including shorter durations for diagnosis, procedure completion and recuperation (p < 0.05). The treatment group exhibited significantly higher patient satisfaction scores (p < 0.05). The incorporation of Streptomyces protease enzyme and Shutai into gastroscopy procedures resulted in significantly enhanced level of procedural efficacy and patient contentment while not introducing an additional risk of long-term complications. The increase in moderate immediate reactions that have been observed requires additional research in order to determine their clinical significance. Although these agents present a possible progression in the field of gastroscopy, their application should be tempered by the immediate adverse reactions that have been documented.

Artificial intelligence-assisted system for the assessment of Forrest classification of peptic ulcer bleeding: a multicenter diagnostic study.

BACKGROUND: Inaccurate Forrest classification may significantly affect clinical outcomes, especially in high risk patients. Therefore, this study aimed to develop a real-time deep convolutional neural network (DCNN) system to assess the Forrest classification of peptic ulcer bleeding (PUB). METHODS: A training dataset (3868 endoscopic images) and an internal validation dataset (834 images) were retrospectively collected from the 900th Hospital, Fuzhou, China. In addition, 521 images collected from four other hospitals were used for external validation. Finally, 46 endoscopic videos were prospectively collected to assess the real-time diagnostic performance of the DCNN system, whose diagnostic performance was also prospectively compared with that of three senior and three junior endoscopists. RESULTS: The DCNN system had a satisfactory diagnostic performance in the assessment of Forrest classification, with an accuracy of 91.2% (95%CI 89.5%-92.6%) and a macro-average area under the receiver operating characteristic curve of 0.80 in the validation dataset. Moreover, the DCNN system could judge suspicious regions automatically using Forrest classification in real-time videos, with an accuracy of 92.0% (95%CI 80.8%-97.8%). The DCNN system showed more accurate and stable diagnostic performance than endoscopists in the prospective clinical comparison test. This system helped to slightly improve the diagnostic performance of senior endoscopists and considerably enhance that of junior endoscopists. CONCLUSION: The DCNN system for the assessment of the Forrest classification of PUB showed satisfactory diagnostic performance, which was slightly superior to that of senior endoscopists. It could therefore effectively assist junior endoscopists in making such diagnoses during gastroscopy.

Transumbilical Endoscopic Resection of Intra-abdominal Mesenchymal Tumor.

Gastrointestinal stromal tumors (GISTs) typically occur in the stomach and proximal small intestine but can also be found in any other part of the digestive tract, including the abdominal cavity, albeit rarely. In the present case, the tumor was resected endoscopically through the anterior gastric wall. Computed tomography (CT) scan and gastroscopy of a 60-year-old woman revealed submucosal lesions in the gastric body. The possibility of a stromal tumor was considered more likely. The endoscopic surgery was performed under endotracheal anesthesia. After a solution had been injected at the lesion site in the stomach, the entire gastric wall was dissected to expose the tumor. As the lesion was in the abdominal cavity and its base was attached to the abdominal wall, it was accessed using a sterilized PCF colonoscope. A sodium chloride injection was administered at the base. The tumor was then peeled along its boundaries using the hooking and excision knife combined with the precutting knife. Subsequently, the tumor was pulled into the stomach through the incision made in the stomach and then extracted externally through the upper digestive tract using the ERCP spiral mesh basket. After confirming the absence of bleeding at the incision site, the endoscope was returned to the stomach, and the stomach opening was closed using purse-string sutures. The patient recovered satisfactorily following the surgery and was discharged on day 4. Histological examination revealed a low-risk stromal tumor (spindle cell type, <5 mitosis/50 high-power fields [HPF]). Immunohistochemistry revealed positive staining for CD34 and CD117, negative staining for SMA, positive staining for DOG1, and negative staining for S100. Additionally, the expression of ki67 was 3%.