Clinical Trial: Efficacy of Mosapride Controlled-release and Nortriptyline in Patients With Functional Dyspepsia: A Multicenter, Double-placebo, Double-blinded, Randomized Controlled, Parallel Clinical Study.

Background/Aims: Prokinetic agents and neuromodulators are among the treatment options for functional dyspepsia (FD), but their comparative efficacy is unclear. We aimed to compare the efficacy of mosapride controlled-release (CR) and nortriptyline in patients with FD after 4 weeks of treatment. Methods: Participants with FD were randomly assigned (1:1) to receive mosapride CR (mosapride CR 15 mg and nortriptyline placebo) or nortriptyline (mosapride CR placebo and nortriptyline 10 mg) in double-placebo, double-blinded, randomized controlled, parallel clinical study. The primary endpoint was defined as the proportion of patients with overall dyspepsia improvement after 4 weeks treatment. The secondary endpoints were changes in individual symptom scores, anxiety, depression, and quality of life. Results: One hundred nine participants were recruited and assessed for eligibility, and 54 in the mosapride CR group and 50 in the nortriptyline group were included in the modified intention-to-treat protocol. The rate of overall dyspepsia improvement was similar between groups (53.7% vs 54.0%, P = 0.976). There was no difference in the efficacy of mosapride CR and nortriptyline in a subgroup analysis by FD subtype (59.3% vs 52.5% in postprandial distress syndrome, P = 0.615; 44.4% vs 40.0% in epigastric pain syndrome, P = > 0.999; 50.0% vs 59.1% in overlap, P = 0.565; respectively). Both treatments significantly improved anxiety, depression, and quality of life from baseline. Conclusion: Mosapride CR and nortriptyline showed similar efficacy in patients with FD regardless of the subtype. Both treatments could be equally helpful for improving quality of life and psychological well-being while also relieving dyspepsia.

Potential risk of proton pump inhibitors for Parkinson's disease: A nationwide nested case-control study.

Proton pump inhibitor (PPI) use is a potential risk factor for neurodegenerative disease development; however, its role in Parkinson's disease (PD) remains unclear. This study aimed to investigate the association between PPI use and PD risk. A total of 31,326 patients with newly diagnosed PD were matches by age, sex, body mass index, diabetes, and hypertension with 125,304 controls at a ratio of 1:4. The data were collected from the Korean National Health Insurance Services Database from January 2010 to December 2019. Cumulative defined daily doses of PPIs were extracted from treatment claims. We examined the association between PPI use and PD risk using conditional logistic regression. To prevent protopathic bias, we excluded patients diagnosed with PD within a 1-year lag period after PPI exposure. We applied 2- and 3-year lag periods for sensitivity analysis. PPI use was associated with an increased risk of PD when a 1-year lag period was applied between PPI exposure and PD development (adjusted odds ratio, 1.10; 95% confidence interval, 1.07-1.13). A significant positive dose-response relationship existed between the cumulative defined daily doses of PPIs and PD development (P<0.001). Similar results were obtained for the 2- or 3-year lag periods. The association did not vary based on gender. Older age, a higher Charlson Comorbidity Index score, no alcohol consumption, and a non-smoking status were associated with a significantly increased PD risk with PPI use. We observed an association between PPI use and PD risk, although long-term follow-up studies are necessary to verify this association.

Comparative effectiveness of explainable machine learning approaches for extrauterine growth restriction classification in preterm infants using longitudinal data.

Introduction: Preterm birth is a leading cause of infant mortality and morbidity. Despite the improvement in the overall mortality in premature infants, the intact survival of these infants remains a significant challenge. Screening the physical growth of infants is fundamental to potentially reducing the escalation of this disorder. Recently, machine learning models have been used to predict the growth restrictions of infants; however, they frequently rely on conventional risk factors and cross-sectional data and do not leverage the longitudinal database associated with medical data from laboratory tests. Methods: This study aimed to present an automated interpretable ML-based approach for the prediction and classification of short-term growth outcomes in preterm infants. We prepared four datasets based on weight and length including weight baseline, length baseline, weight follow-up, and length follow-up. The CHA Bundang Medical Center Neonatal Intensive Care Unit dataset was classified using two well-known supervised machine learning algorithms, namely support vector machine (SVM) and logistic regression (LR). A five-fold cross-validation, and several performance measures, including accuracy, precision, recall and F1-score were used to compare classifier performances. We further illustrated the models' trustworthiness using calibration and cumulative curves. The visualized global interpretations using Shapley additive explanation (SHAP) is provided for analyzing variables' contribution to final prediction. Results: Based on the experimental results with area under the curve, the discrimination ability of the SVM algorithm was found to better than that of the LR model on three of the four datasets with 81%, 76% and 72% in weight follow-up, length baseline and length follow-up dataset respectively. The LR classifier achieved a better ROC score only on the weight baseline dataset with 83%. The global interpretability results revealed that pregnancy-induced hypertension, gestational age, twin birth, birth weight, antenatal corticosteroid use, premature rupture of membranes, sex, and birth length were consistently ranked as important variables in both the baseline and follow-up datasets. Discussion: The application of machine learning models to the early detection and automated classification of short-term growth outcomes in preterm infants achieved high accuracy and may provide an efficient framework for clinical decision systems enabling more effective monitoring and facilitating timely intervention.

Risk of gastric cancer among long-term proton pump inhibitor users: a population-based cohort study.

PURPOSE: To elucidate whether long-term proton pump inhibitor (PPI) users have an increased gastric cancer (GC) risk. METHODS: We searched the 2009-2019 Korean National Health Insurance Services Database for patients aged > 40 years who claimed for Helicobacter pylori eradication (HPE) during 2009-2014. The GC incidence following a PPI exposure of > 180 cumulative defined daily dose (cDDD) and that following an exposure of < 180 cDDD were compared. The outcome was GC development at least 1 year following HPE. A propensity score (PS)-matched dataset was used for analysis within the same quartiles of the follow-up duration. Additionally, dose-response associations were assessed, and the mortality rates were compared between long-term PPI users and non-users. RESULTS: After PS matching, 144,091 pairs of PPI users and non-users were analyzed. During a median follow-up of 8.3 (interquartile range, 6.8-9.6) years, 1053 and 948 GC cases in PPI users and non-users, respectively, were identified, with the GC incidence (95% confidence interval (CI)) being 0.90 (0.85-0.96) and 0.81 (0.76-0.86) per 1000 person-years, respectively. The adjusted hazard ratio (aHR) for GC with PPI use was 1.15 (95% CI, 1.06-1.25). Among PPI users, patients in the highest tertile for annual PPI dose showed higher GC development than those in the lowest tertile (aHR (95% CI): 3.87 (3.25-4.60)). GC-related mortality did not differ significantly between PPI users and non-users. CONCLUSION: In this nationwide analysis in Korea, where the GC prevalence is high, long-term PPI use after HPE showed a significant increase in GC, with a positive dose-response relationship.

Proton pump inhibitor use and the risk of metachronous gastric cancer after H. pylori eradication in patients who underwent endoscopic resection for gastric neoplasms: A population-based cohort study.

BACKGROUND: The association between proton pump inhibitors (PPI) use and gastric cancer remains controversial. AIMS: To investigate the impact of long-term PPI use on metachronous gastric cancer after Helicobacter pylori eradication in high-risk patients who underwent endoscopic resection of gastric neoplasms. METHODS: Using the Korean National Health Insurance Services database, we identified 1836 PPI users and 12,218 non-users among patients who received H. pylori eradication therapy after endoscopic resection for gastric neoplasms between 2009 and 2014. We then compared the incidence of metachronous gastric cancer between the PPI user and non-user groups. We conducted sensitivity analysis using various time lags and propensity score-matched analysis to ensure the robustness of the results. RESULTS: After a median follow-up of 7.3 years, the incidence of metachronous gastric cancer was significantly higher in the PPI user group than in the non-user group, with a crude hazard ratio of 6.20 (95% confidence interval, 5.78-6.65). After adjustment, PPI use was associated with the development of metachronous gastric cancer, with an adjusted hazard ratio of 5.51 (95% confidence interval, 5.12-5.92). The PPI user group was categorised into three subgroups according to the cumulative PPI dose; the increased risk of metachronous gastric cancer remained significant regardless of the PPI dose. Moreover, these results remained robust after applying various time lags and propensity score-matched analyses. CONCLUSIONS: Long-term PPI use is associated with an increased risk of metachronous gastric cancer in patients who undergo H. pylori eradication therapy after endoscopic resection of gastric neoplasms.

The Degree of Hydronephrosis as an Indicator of the Necessity for Ureteric Dilatation during Ureteroscopic Lithotripsy.

During rigid ureteroscopic lithotripsy, it is often encountered that the ureter is difficult to access. Attempts to advance the ureteroscope make the surgery more difficult. This study evaluated the preoperative predictive factors associated with difficult ureteral access (difficult ureter (DU)) during URS and assessed if clinical outcomes differed according to the degree of DU. This study identified 217 patients who underwent rigid ureteroscopic (URS) lithotripsy for the management of ureter stones between June 2017 and July 2021 in a tertiary hospital in Korea. In this group, preoperative factors were identified using univariate and multiple logistic regression analyses that could predict the degree of DU. Additionally, we also evaluated differences in treatment outcomes depending on the degree of DU. In 50 URS cases (22.0%), ureteral access using a ureteroscope was difficult. In the univariate and multivariate analyses, the degree of hydronephrosis was associated with the degree of DU. Treatment outcomes, extended operation times, low stone-free rate, postoperative pain, and secondary treatment were also significantly associated with the degree of DU. Clinicians can counsel patients with a lesser degree of hydronephrosis and approach their management accordingly.

Immunoglobulin G4-related Disease of the Small Bowel: A Case of Long-term Remission Achieved by Surgical Resection without Maintenance Therapy.

Immunoglobulin G4-related disease (IgG4-RD) is an immune-mediated fibroinflammatory disease. IgG4-RD can affect any organ system, including the pancreas, bile ducts, salivary glands, mesentery, and retroperitoneum. On the other hand, small intestine involvement is extremely rare. This paper describes a case of IgG4-RD involving the small bowel, particularly at the distal ileum. An 81-year-old female was admitted to the authors' hospital complaining of abdominal pain, dyspepsia, and hematochezia. The laboratory tests, including tumor markers and IgG4, were within normal limits. A colonoscopy did not show any abnormal findings. Abdominal computed tomography revealed segmental aneurysmal dilatation and wall thickening at the distal ileum, suggesting malignant conditions, such as small bowel lymphoma. The patient underwent an exploratory laparoscopy and ileocecectomy to differentiate a malignancy. A histopathology examination revealed dense lymphoplasmacytic infiltration, storiform fibrosis, and IgG4-positive plasma cells (>50 per high power field). The patient was finally diagnosed with IgG4-RD. The patient was followed up in the outpatient clinic for five years without recurrence. This paper suggests that a radical resection without maintenance therapy can be a treatment option, particularly when the IgG4-RD manifests as a localized gastrointestinal tract lesion.

Effect of Helicobacter pylori Eradication on Body Weight: A Multicenter Propensity Score-matched Analysis in Korea.

Background/Aims: There is growing interest in whether Helicobacter pylori eradication (HPE) can affect body weight. Methods: Data from 5 universities between January 2013 and December 2019 were analyzed retrospectively. H. pylori-positive subjects who had body weight measurements taken at least twice at intervals of 3 months or longer were included. Using propensity score (PS)-matched data, changes in body mass index (BMI) and the lipid profile after HPE were compared with the non-HPE group. Results: Among 363 eligible patients, 131 HPE patients were PS-matched to 131 non-HPE patients. The median intervals between the measurements were 610 (range, 154-1250) days and 606 (range, 154-1648) days in the HPE and non-HPE groups, respectively. In both groups, the mean BMI increased (from 24.5 kg/m2 to 24.7 kg/m2 in the HPE group, and from 24.4 kg/m2 to 24.5 kg/m2 in the non-HPE group). The 2 groups did not show significantly different changes (P = 0.921). In the lowest baseline BMI quartile, the BMI increased after HPE by 1.23 (standard deviation [SD], 3.72) kg/m2 (P = 0.060), and the non-HPE group showed a decreased BMI at the time of follow-up (by -0.24 [SD, 5.25] kg/m2; P = 0.937) (between-group P = 0.214). Triglyceride levels increased after HPE (mean: 135 [SD, 78] to 153 [SD, 100] mg/dL; between-group P = 0.053). Conclusion: The overall BMI change was not significantly different between the HPE and non-HPE groups, but patients with low BMI showed a tendency to gain weight after HPE. Triglyceride levels increased after HPE with marginal significance.

TNM-Based Head-to-Head Comparison of Urachal Carcinoma and Urothelial Bladder Cancer: Stage-Matched Analysis of a Large Multicenter National Cohort.

PURPOSE: Outcome analysis of urachal cancer (UraC) is limited due to the scarcity of cases and different staging methods compared to urothelial bladder cancer (UroBC). We attempted to assess survival outcomes of UraC and compare to UroBC after stage-matched analyses. MATERIALS AND METHODS: Total 203 UraC patients from a multicenter database and 373 UroBC patients in single institution from 2000 to 2018 were enrolled (median follow-up, 32 months). Sheldon stage conversion to corresponding TNM staging for UraC was conducted for head-to-head comparison to UroBC. Perioperative clinical variables and pathological results were recorded. Stage-matched analyses for survival by stage were conducted. RESULTS: UraC patients were younger (mean age, 54 vs. 67 years; p < 0.001), with 163 patients (80.3%) receiving partial cystectomy and 23 patients (11.3%) radical cystectomy. UraC was more likely to harbor ≥ pT3a tumors (78.8% vs. 41.8%). While 5-year recurrence-free survival, cancer-specific survival (CSS) and overall survival were comparable between two groups (63.4%, 67%, and 62.1% in UraC and 61.5%, 75.9%, and 67.8% in UroBC, respectively), generally favorable prognosis for UraC in lower stages (pT1-2) but unfavorable outcomes in higher stages (pT4) compared to UroBC was observed, although only 5-year CSS in ≥ pT4 showed statistical significance (p=0.028). Body mass index (hazard ratio [HR], 0.929), diabetes mellitus (HR, 1.921), pathologic T category (HR, 3.846), and lymphovascular invasion (HR, 1.993) were predictors of CSS for all patients. CONCLUSION: Despite differing histology, UraC has comparable prognosis to UroBC with relatively favorable outcome in low stages but worse prognosis in higher stages. The presented system may be useful for future grading and risk stratification of UraC.

Korean Guidelines for Postpolypectomy Colonoscopic Surveillance: 2022 revised edition.

Colonoscopic polypectomy is effective in decreasing the incidence and mortality of colorectal cancer (CRC). Premalignant polyps discovered during colonoscopy are associated with the risk of metachronous advanced neoplasia. Postpolypectomy surveillance is the most important method for managing advanced metachronous neoplasia. A more efficient and evidence-based guideline for postpolypectomy surveillance is required because of the limited medical resources and concerns regarding colonoscopy complications. In these consensus guidelines, an analytic approach was used to address all reliable evidence to interpret the predictors of CRC or advanced neoplasia during surveillance colonoscopy. The key recommendations state that the high-risk findings for metachronous CRC following polypectomy are as follows: adenoma ≥10 mm in size; 3 to 5 (or more) adenomas; tubulovillous or villous adenoma; adenoma containing high-grade dysplasia; traditional serrated adenoma; sessile serrated lesion containing any grade of dysplasia; serrated polyp of at least 10 mm in size; and 3 to 5 (or more) sessile serrated lesions. More studies are needed to fully comprehend the patients who are most likely to benefit from surveillance colonoscopy and the ideal surveillance interval to prevent metachronous CRC.

Targeting TBK1 to overcome resistance to cancer immunotherapy.

Despite the success of PD-1 blockade in melanoma and other cancers, effective treatment strategies to overcome resistance to cancer immunotherapy are lacking1,2. Here we identify the innate immune kinase TANK-binding kinase 1 (TBK1)3 as a candidate immune-evasion gene in a pooled genetic screen4. Using a suite of genetic and pharmacological tools across multiple experimental model systems, we confirm a role for TBK1 as an immune-evasion gene. Targeting TBK1 enhances responses to PD-1 blockade by decreasing the cytotoxicity threshold to effector cytokines (TNF and IFNγ). TBK1 inhibition in combination with PD-1 blockade also demonstrated efficacy using patient-derived tumour models, with concordant findings in matched patient-derived organotypic tumour spheroids and matched patient-derived organoids. Tumour cells lacking TBK1 are primed to undergo RIPK- and caspase-dependent cell death in response to TNF and IFNγ in a JAK-STAT-dependent manner. Taken together, our results demonstrate that targeting TBK1 is an effective strategy to overcome resistance to cancer immunotherapy.

Long-Term Influenza Outbreak Forecast Using Time-Precedence Correlation of Web Data.

Influenza leads to many deaths every year and is a threat to human health. For effective prevention, traditional national-scale statistical surveillance systems have been developed, and numerous studies have been conducted to predict influenza outbreaks using web data. Most studies have captured the short-term signs of influenza outbreaks, such as one-week prediction using the characteristics of web data uploaded in real time; however, long-term predictions of more than 2-10 weeks are required to effectively cope with influenza outbreaks. In this study, we determined that web data uploaded in real time have a time-precedence relationship with influenza outbreaks. For example, a few weeks before an influenza pandemic, the word "colds" appears frequently in web data. The web data after the appearance of the word "colds" can be used as information for forecasting future influenza outbreaks, which can improve long-term influenza prediction accuracy. In this study, we propose a novel long-term influenza outbreak forecast model utilizing the time precedence between the emergence of web data and an influenza outbreak. Based on the proposed model, we conducted experiments on: 1) selecting suitable web data for long-term influenza prediction; 2) determining whether the proposed model is regionally dependent; and 3) evaluating the accuracy according to the prediction timeframe. The proposed model showed a correlation of 0.87 in the long-term prediction of ten weeks while significantly outperforming other state-of-the-art methods.

What are the risk factors for recurrent UTI with repeated ESBL-producing Enterobacteriaceae? A retrospective cohort study.

INTRODUCTION: A previous study has shown that two-thirds of patients with urinary tract infections (UTIs) caused by extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae experience recurrence with the same bacteria on subsequent UTI episodes. However, little is known about which patients suffer from UTI due to ESBL-producing Enterobacteriaceae repeatedly. This study aimed to investigate the risk factors for recurrent UTI due to repeated ESBL-producing organism infections. METHODS: This retrospective, single-center, observational cohort study screened all patients with UTI caused by ESBL-producing strains between January 2012 and April 2019. Among the patients who were followed up, patients who experienced UTI recurrence were enrolled and divided into two groups: ESBL recurrence group and non-ESBL recurrence group. Multivariable Cox proportional hazards regression analyses were performed to evaluate the association between patient characteristics and the development of recurrent UTI caused by ESBL-producing Enterobacteriaceae. RESULTS: A total of 330 patients were followed up after the diagnosis of UTI caused by ESBL-producing organisms. Among the patients, 115 (34.8%) experienced UTI recurrence, and 71 (61.7%) of them experienced subsequent recurrent UTI due to ESBL-producing organisms. Patient's age (hazard ratio [HR], 1.02; 95% confidence interval [CI], 1.00-1.04; P = 0.046) and recurrent UTI history (HR, 1.69; 95% CI, 1.05-2.72; P = 0.031) were significantly associated with an increased risk of recurrence with ESBL-producing Enterobacteriaceae. CONCLUSION: These findings showed that a history of previous frequent UTI recurrence is the risk factor for recurrence of UTI due to repeated ESBL producing Enterobacteriaceae infections.

Effectively computing transition patterns with privacy-preserved trajectory datasets.

Recent advances in positioning techniques, along with the widespread use of mobile devices, make it easier to monitor and collect user trajectory information during their daily activities. An ever-growing abundance of data about trajectories of individual users paves the way for various applications that utilize user mobility information. One of the most common analysis tasks in these new applications is to extract the sequential transition patterns between two consecutive timestamps from a collection of trajectories. Such patterns have been widely exploited in diverse applications to predict and recommend next user locations based on the current position. Thus, in this paper, we explore the computation of the transition patterns, especially with a trajectory dataset collected using differential privacy, which is a de facto standard for privacy-preserving data collection and processing. Specifically, the proposed scheme relies on geo-indistinguishability, which is a variant of the well-known differential privacy, to collect trajectory data from users in a privacy-preserving manner, and exploits the functionality of the expectation-maximization algorithm to precisely estimate hidden transition patterns based on perturbed trajectory datasets collected under geo-indistinguishability. Experimental results using real trajectory datasets confirm that a good estimation of transition pattern can be achieved with the proposed method.

A Care Robot with Ethical Sensing System for Older Adults at Home.

Many studies have explored emotional and mental services that robots can provide for older adults, such as offering them daily conversation, news, music, or health information. However, the ethical issues raised by using sensors for frail older adults to monitor their daily movements or their medication intake, for instance, are still being discussed. In this study, we develop an older adult-guided, caregiver-monitored robot, Dori, which can detect and recognize movement by sensing human poses in accordance with two factors from the human-centered artificial intelligence (HCAI) framework. To design the care robot's services based on sensing movement during daily activities, we conducted focus group interviews with two groups-caregivers and medical staff-on the topic of care robot services not for patients but for prefrail and frail elderly individuals living at home. Based on their responses, we derived the focal service areas of cognitive support, emotional support, physical activity support, medication management, and caregiver management. We also found the two groups differed in their ethical judgments in the areas of dignity, autonomy, controllability, and privacy for services utilizing sensing by care robots. Therefore, the pose recognition technology adopted in the present work uses only joint coordinate information extracted from camera images and thus is advantageous for protecting human dignity and personal information.

Korean guidelines for postpolypectomy colonoscopic surveillance: 2022 revised edition.

Colonoscopic polypectomy is effective in decreasing the incidence and mortality of colorectal cancer (CRC). Premalignant polyps discovered during colonoscopy are associated with the risk of metachronous advanced neoplasia. Postpolypectomy surveillance is the most important method for the management of advanced metachronous neoplasia. A more efficient and evidence-based guideline for postpolypectomy surveillance is required because of limited medical resources and concerns regarding colonoscopy complications. In these consensus guidelines, an analytic approach was used to address all reliable evidence to interpret the predictors of CRC or advanced neoplasia during surveillance colonoscopy. The key recommendations state that the high-risk findings for metachronous CRC following polypectomy are as follows: (1) adenoma ≥10 mm in size; (2) 3 to 5 (or more) adenomas; (3) tubulovillous or villous adenoma; (4) adenoma containing high-grade dysplasia; (5) traditional serrated adenoma; (6) sessile serrated lesion (SSL) containing any grade of dysplasia; (7) serrated polyp of at least 10 mm in size; and (8) 3 to 5 (or more) SSLs. More studies are needed to fully comprehend the patients most likely to benefit from surveillance colonoscopy and the ideal surveillance interval to prevent metachronous CRC.

[Korean Guidelines for Postpolypectomy Colonoscopic Surveillance: 2022 Revised Edition].

Colonoscopic polypectomy is effective in decreasing the incidence and mortality of colorectal cancer (CRC). Premalignant polyps discovered during colonoscopy are associated with the risk of metachronous advanced neoplasia. Postpolypectomy surveillance is the most important method for managing advanced metachronous neoplasia. A more efficient and evidence-based guideline for postpolypectomy surveillance is required because of the limited medical resources and concerns regarding colonoscopy complications. In these consensus guidelines, an analytic approach was used to address all reliable evidence to interpret the predictors of CRC or advanced neoplasia during surveillance colonoscopy. The key recommendations state that the high-risk findings for metachronous CRC following polypectomy are as follows: 1) adenoma ≥10 mm in size; 2) 3-5 (or more) adenomas; 3) tubulovillous or villous adenoma; 4) adenoma containing high-grade dysplasia; 5) traditional serrated adenoma; 6) sessile serrated lesion (SSL) containing any grade of dysplasia; 7) serrated polyp of at least 10 mm in size; and 8) 3-5 (or more) SSLs. More studies are needed to fully comprehend the patients who are most likely to benefit from surveillance colonoscopy and the ideal surveillance interval to prevent metachronous CRC.

Evaluation of autofluorescence and image of oral pathogens and the tooth.

PURPOSE: To test the applicability of autofluorescence (AF) spectrum and image in the detection and identification of oral pathogens. METHODS: Oral pathogens (Candida albicans, Porphyromonas gingivalis, Streptococcus mutans) and teeth were used. To induce AF, the 405 nm laser was used as a light source, and AF was obtained and observed using a spectrometer, fluorescence camera, and microscope, respectively. RESULTS: The tested oral pathogens had similar spectral distributions, but their peak intensities and peak ratios were different. Their peak positions and spectral patterns were different from those of the tested sound and carious teeth. These differences were also found from the other referenced oral mucosa. Fluorescence image could localize the existence of oral bacteria. Oral pathogens could be imaged by fluorescence, but identification of each pathogen by image was not probable. CLINICAL SIGNIFICANCE: Oral pathogens can be observed and identified from the lesion if autofluorescence spectrum and fluorescence images are combined.

Deep-Learning for the Diagnosis of Esophageal Cancers and Precursor Lesions in Endoscopic Images: A Model Establishment and Nationwide Multicenter Performance Verification Study.

BACKGROUND: Suspicion of lesions and prediction of the histology of esophageal cancers or premalignant lesions in endoscopic images are not yet accurate. The local feature selection and optimization functions of the model enabled an accurate analysis of images in deep learning. OBJECTIVES: To establish a deep-learning model to diagnose esophageal cancers, precursor lesions, and non-neoplasms using endoscopic images. Additionally, a nationwide prospective multicenter performance verification was conducted to confirm the possibility of real-clinic application. METHODS: A total of 5162 white-light endoscopic images were used for the training and internal test of the model classifying esophageal cancers, dysplasias, and non-neoplasms. A no-code deep-learning tool was used for the establishment of the deep-learning model. Prospective multicenter external tests using 836 novel images from five hospitals were conducted. The primary performance metric was the external-test accuracy. An attention map was generated and analyzed to gain the explainability. RESULTS: The established model reached 95.6% (95% confidence interval: 94.2-97.0%) internal-test accuracy (precision: 78.0%, recall: 93.9%, F1 score: 85.2%). Regarding the external tests, the accuracy ranged from 90.0% to 95.8% (overall accuracy: 93.9%). There was no statistical difference in the number of correctly identified the region of interest for the external tests between the expert endoscopist and the established model using attention map analysis (P = 0.11). In terms of the dysplasia subgroup, the number of correctly identified regions of interest was higher in the deep-learning model than in the endoscopist group, although statistically insignificant (P = 0.48). CONCLUSIONS: We established a deep-learning model that accurately classifies esophageal cancers, precursor lesions, and non-neoplasms. This model confirmed the potential for generalizability through multicenter external tests and explainability through the attention map analysis.

No-Code Platform-Based Deep-Learning Models for Prediction of Colorectal Polyp Histology from White-Light Endoscopy Images: Development and Performance Verification.

BACKGROUND: The authors previously developed deep-learning models for the prediction of colorectal polyp histology (advanced colorectal cancer, early cancer/high-grade dysplasia, tubular adenoma with or without low-grade dysplasia, or non-neoplasm) from endoscopic images. While the model achieved 67.3% internal-test accuracy and 79.2% external-test accuracy, model development was labour-intensive and required specialised programming expertise. Moreover, the 240-image external-test dataset included only three advanced and eight early cancers, so it was difficult to generalise model performance. These limitations may be mitigated by deep-learning models developed using no-code platforms. OBJECTIVE: To establish no-code platform-based deep-learning models for the prediction of colorectal polyp histology from white-light endoscopy images and compare their diagnostic performance with traditional models. METHODS: The same 3828 endoscopic images used to establish previous models were used to establish new models based on no-code platforms Neuro-T, VLAD, and Create ML-Image Classifier. A prospective multicentre validation study was then conducted using 3818 novel images. The primary outcome was the accuracy of four-category prediction. RESULTS: The model established using Neuro-T achieved the highest internal-test accuracy (75.3%, 95% confidence interval: 71.0-79.6%) and external-test accuracy (80.2%, 76.9-83.5%) but required the longest training time. In contrast, the model established using Create ML-Image Classifier required only 3 min for training and still achieved 72.7% (70.8-74.6%) external-test accuracy. Attention map analysis revealed that the imaging features used by the no-code deep-learning models were similar to those used by endoscopists during visual inspection. CONCLUSION: No-code deep-learning tools allow for the rapid development of models with high accuracy for predicting colorectal polyp histology.