ABSTRACT
Background/Aims@#This study aimed to investigate the possibility of in situ diagnosis of Helicobacter pylori (H. pylori) infection during endoscopic examination. The predictive infection value was estimated using the endoscopic Kyoto scoring system (EKSS), and specific endoscopic findings were evaluated for diagnosing H. pylori infection in H. pylori naïve patients and those with a eradication history. @*Materials and Methods@#A total of 836 patients with H. pylori infection were analyzed. The state of the infection was predicted using the EKSS and specific endoscopic findings. @*Results@#Patients were classified into two groups: the H. pylori naïve group and the group with a the bacterial eradication history. The area under the curve (AUC) on receiver operating characteristics analysis was 0.90 for EKSS in H. pylori naïve patients and 0.83 for the other group patients. For patients with open type atrophy and/or intestinal metaplasia, EKSS (24.4%; 95% CI, 12.4~0.3%) and regular arrangement of collecting venules (RAC) (46.3%; 95% CI, 30.7~62.9%) showed low specificities. Mucosal swelling (66.2%; 95% CI, 62.5~69.7%) and sticky mucus (80.5%; 95% CI, 74.8~85.2%) presented relatively high positive predictive values for H. pylori infection in naïve patients, whereas reflux esophagitis, hematin, red streak, and duodenitis exhibited high negative predictive values in patients with a H. pylori eradication history (98.0%; 95% CI, 96.4~99.1%). @*Conclusions@#EKSS and RAC are excellent tools for predicting H. pylori infection. However, they have a limited role in patients with open type atrophy and/or intestinal metaplasia. Specific endoscopic findings could help predict the infection state.
ABSTRACT
Background@#Vaccine safety surveillance is important because it is related to vaccine hesitancy, which affects vaccination rate. To increase confidence in vaccination, the active monitoring of vaccine adverse events is important. For effective active surveillance, we developed and verified a machine learning-based active surveillance system using national claim data. @*Methods@#We used two databases, one from the Korea Disease Control and Prevention Agency, which contains flu vaccination records for the elderly, and another from the National Health Insurance Service, which contains the claim data of vaccinated people. We developed a casecrossover design based machine learning model to predict the health outcome of interest events (anaphylaxis and agranulocytosis) using a random forest. Feature importance values were evaluated to determine candidate associations with each outcome. We investigated the relationship of the features to each event via a literature review, comparison with the Side Effect Resource, and using the Local Interpretable Model-agnostic Explanation method. @*Results@#The trained model predicted each health outcome of interest with a high accuracy (approximately 70%). We found literature supporting our results, and most of the important drug-related features were listed in the Side Effect Resource database as inducing the health outcome of interest. For anaphylaxis, flu vaccination ranked high in our feature importance analysis and had a positive association in Local Interpretable Model-Agnostic Explanation analysis. Although the feature importance of vaccination was lower for agranulocytosis, it also had a positive relationship in the Local Interpretable Model-Agnostic Explanation analysis. @*Conclusion@#We developed a machine learning-based active surveillance system for detecting possible factors that can induce adverse events using health claim and vaccination databases. The results of the study demonstrated a potentially useful application of two linked national health record databases. Our model can contribute to the establishment of a system for conducting active surveillance on vaccination.
ABSTRACT
Background@#Vaccine safety surveillance is important because it is related to vaccine hesitancy, which affects vaccination rate. To increase confidence in vaccination, the active monitoring of vaccine adverse events is important. For effective active surveillance, we developed and verified a machine learning-based active surveillance system using national claim data. @*Methods@#We used two databases, one from the Korea Disease Control and Prevention Agency, which contains flu vaccination records for the elderly, and another from the National Health Insurance Service, which contains the claim data of vaccinated people. We developed a casecrossover design based machine learning model to predict the health outcome of interest events (anaphylaxis and agranulocytosis) using a random forest. Feature importance values were evaluated to determine candidate associations with each outcome. We investigated the relationship of the features to each event via a literature review, comparison with the Side Effect Resource, and using the Local Interpretable Model-agnostic Explanation method. @*Results@#The trained model predicted each health outcome of interest with a high accuracy (approximately 70%). We found literature supporting our results, and most of the important drug-related features were listed in the Side Effect Resource database as inducing the health outcome of interest. For anaphylaxis, flu vaccination ranked high in our feature importance analysis and had a positive association in Local Interpretable Model-Agnostic Explanation analysis. Although the feature importance of vaccination was lower for agranulocytosis, it also had a positive relationship in the Local Interpretable Model-Agnostic Explanation analysis. @*Conclusion@#We developed a machine learning-based active surveillance system for detecting possible factors that can induce adverse events using health claim and vaccination databases. The results of the study demonstrated a potentially useful application of two linked national health record databases. Our model can contribute to the establishment of a system for conducting active surveillance on vaccination.
ABSTRACT
Background/Aims@#This study examined the clinical features and prognosis of patients with mucinous gastric carcinoma (MGC), non-mucinous gastric carcinoma (NMGC), and signet ring cell gastric carcinoma (SRC). @*Methods@#A retrospective cohort study was performed, enrolling 65 patients with MGC from January 2007 to December 2016.During the same period, 1,814 patients with histologically proven gastric cancers underwent curative or palliative operations. One hundred and ninety-five NMGC patients were selected as the 1:3 age- and sex-matched control groups. In addition, 200 SRC patients were identified. This study evaluated the demographic features of the patients, pathologic features of the tumor, and the predictive factors, such as the recurrence-free survival and overall survival. @*Results@#The recurrence rates were significantly high in MGC than in NMGC or SRC (both p<0.01). The proportion of early gastric cancer was lower in the MGC group than in the other groups (p<0.01). In addition, metastatic lymph nodes were found more frequently in the MGC group (p<0.01), and the proportion of initial pT4, M1 stage, was highest in the MGC group. The recurrence-free survival and overall survival in the MGC group were significantly lower than those in the NMGC or SRC. Subgroup analysis showed that patients with the same American Joint Committee on Cancer (AJCC) stage of each cancer group showed a similar prognosis. @*Conclusions@#MGC frequently presents an advanced stage with an unfavorable prognosis compared to NMGC or SRC. On the other hand, MGC of the same AJCC stage had a similar prognosis to NMGC and SRC.
ABSTRACT
Background/Aims@#This study examined the clinical features and prognosis of patients with mucinous gastric carcinoma (MGC), non-mucinous gastric carcinoma (NMGC), and signet ring cell gastric carcinoma (SRC). @*Methods@#A retrospective cohort study was performed, enrolling 65 patients with MGC from January 2007 to December 2016.During the same period, 1,814 patients with histologically proven gastric cancers underwent curative or palliative operations. One hundred and ninety-five NMGC patients were selected as the 1:3 age- and sex-matched control groups. In addition, 200 SRC patients were identified. This study evaluated the demographic features of the patients, pathologic features of the tumor, and the predictive factors, such as the recurrence-free survival and overall survival. @*Results@#The recurrence rates were significantly high in MGC than in NMGC or SRC (both p<0.01). The proportion of early gastric cancer was lower in the MGC group than in the other groups (p<0.01). In addition, metastatic lymph nodes were found more frequently in the MGC group (p<0.01), and the proportion of initial pT4, M1 stage, was highest in the MGC group. The recurrence-free survival and overall survival in the MGC group were significantly lower than those in the NMGC or SRC. Subgroup analysis showed that patients with the same American Joint Committee on Cancer (AJCC) stage of each cancer group showed a similar prognosis. @*Conclusions@#MGC frequently presents an advanced stage with an unfavorable prognosis compared to NMGC or SRC. On the other hand, MGC of the same AJCC stage had a similar prognosis to NMGC and SRC.