A Deep Learning Model for Classifying Histological Types of Colorectal Polyps.

In this study a deep learning architecture based on a convolutional neural network has been evaluated for the classification of white light images of colorectal polyps acquired during the process of a colonoscopy, to estimate the accuracy of the optical recognition of histologic types of polyps. Convolutional neural networks (CNNs), a subclass of artificial neural networks that have gained dominance in several computer vision tasks, are gaining popularity in many medical fields, including endoscopy. The TensorFlow framework was used for implementing EfficientNetB7, which was trained with 924 images, drawn from 86 patients. 55% of the polyps were adenomas, 22% were hyperplastic, and 17% were lesions with sessile serrations. The validation loss, accuracy, and AUC ROC were 0.4845, 0.7778, and 0.8881 respectively.

Convolutional Neural Networks for Optical Discrimination Between Histological Types of Colorectal Polyps Based on White Light Endoscopic Images.

The objective of this study was to compare different convolutional neural networks (CNNs), as employed in a Python-produced deep learning process, used on white light images of colorectal polyps acquired during the process of a colonoscopy, in order to estimate the accuracy of the optical recognition of particular histologic types of polyps. The TensorFlow framework was used for Inception V3, ResNet50, DenseNet121, and NasNetLarge, which were trained with 924 images, drawn from 86 patients.

Using Machine Learning for Predicting the Hospitalization of Emergency Department Patients.

Artificial intelligence processes are increasingly being used in emergency medicine, notably for supporting clinical decisions and potentially improving healthcare services. This study investigated demographics, coagulation tests, and biochemical markers routinely used for patients seen in the Emergency Department (ED) concerning hospitalization. This retrospective observational study included 13,991 emergency department visits of patients who had undergone biomarker testing to a tertiary public hospital in Greece during 2020. After applying five well-known classifiers of the caret package for machine learning of the R programming language in the whole data set and to each ED unit separately, the best performance regarding AUC ROC was observed in the Pulmonology ED unit. Furthermore, among the five classification techniques evaluated, a random forest classifier outperformed other models.

Exploratory Clustering for Emergency Department Patients.

Emergency department (ED) overcrowding is an increasing global problem raising safety concerns for the patients. Elaborating an effective triage system that properly separates patients requiring hospital admission remains difficult. The objective of this study was to compare a clustering-related technique assignment of emergency department patients with the admission output using the k-means algorithm. Incorporating such a model into triage practice could theoretically shorten waiting times and reduce ED overcrowding.

Prediction of Hospitalization Using Machine Learning for Emergency Department Patients.

The objective of this study was to evaluate the predictive capability of five machine learning models regarding the admission or discharge of emergency department patients. A Random Forest classifier outperformed other models with respect to the area under the receiver operating characteristic curve (AUC ROC).

New considerations for colorectal cancer screening based on the demographic profile of colorectal cancer in a Greek population.

Colorectal cancer screening has long been recommended for middle age and older individuals. Recent evidence indicates increasing incidence and mortality among young adults. Therefore, the present study re-examined the current recommendations using an asymptomatic average-risk population screened by colonoscopy. A total of 716 participants of a wide age range were prospectively enrolled in an open-access endoscopic screening program based on self-referral. Comparisons between different age, gender and location groups, and receiver operating characteristic curves (ROC) curves for best age selection for detection of lesions were employed. Increased incidence of advanced lesions was observed in adults <50 years old. Although the polyp size was <1 cm in 85% of the cohort, a significant number of participants harbored advanced lesions. A disturbing incidence of lesions in women 30-49 years was located mainly in the left colon. One-third of the important pathology resides exclusively in the right colon. ROC curves demonstrated that with the current starting age of screening at 50 years, 92% of polyps and 95% of adenomas could be detected by colonoscopy, but a number of potential precancerous lesions will appear at an earlier age and therefore will be missed. The present study supported the notion that it is critical to reduce screening initiation below the currently accepted age of 50 years. Colonoscopy is a suitable method for addressing the increased prevalence of proximal lesions and the meticulous resection of smaller polyps.

Alarming endoscopic data in young and older asymptomatic people: Results of an open access, unlimited age colonoscopic screening for colorectal cancer.

There is a lack of a national organized screening program for colorectal cancer in Greece, and asymptomatic detection is usually the result of individual decisions. The collection of epidemiologic endoscopic data from a population of interest would therefore provide valuable information for future treatment guidance, especially during periods of economic austerity. The current cross-sectional study included 380 asymptomatic, average risk individuals undergoing screening colonoscopy for the first time, during the period of one year in a tertiary public hospital in Athens. Descriptive and analytic epidemiologic data were analyzed. The prevalence of adenomas and advanced lesions were compared between the younger and older cohort, and a regression model was applied for risk evaluation. The mean age of participants was 63 years, and 53% were male. A significant proportion of patients presented with polyps (51.5%) and 25% of them had lesions in the proximal colon. The prevalence of adenomas and advanced adenomas was 29.5 and 11.8%, respectively. Similar high prevalence rates of lesions were identified in the cohort of individuals <50 years of age and the older cohort (>50 years of age). Regression models identified age, number and size of polyps as the major risk factors for the detection of adenomas. The increase of advanced lesions in the older and younger cohort requires confirmation by larger studies. Overall, the results of the present study indicate the requirement for a well-organized screening colonoscopy program starting from as early as 40 years of age. This program may confer an additional endoscopic burden with socioeconomic consequences in a country with limited health resources.

Should hot biopsy forceps be abandoned for polypectomy of diminutive colorectal polyps?

Standardized approach to polypectomy of diminutive colorectal polyps (DCPs) is lacking since cold biopsy forceps have been associated with high levels of recurrence, hot biopsy forceps are considered inadequate and risky and cold snaring is currently under investigation for its efficacy and safety. This has led to confusion and a gap in clinical practice. This article discusses the usefulness and contemporary practical applicability of hot biopsy forceps and provides well-intentioned criticism of the new European guidelines for the treatment of DCPs. Diminutive colorectal polyps are a source of frustration for the endoscopist since their small size is accompanied by a considerable risk of premalignant neoplasia and a small but non-negligible risk of advanced neoplasia and even cancer. Since the proportion of diminutive colorectal polyps is substantial and exceeds that of larger polyps, their effective removal poses a considerable workload and a therapeutic challenge. During the last decade, the introduction of cold snaring to routine endoscopy practice has attempted to overcome the use of prior techniques, such as hot biopsy forceps. It is important to recognize that with the exception of endoscopic methods that are obviously unsafe and inadequate to serve their purpose, all other interventional endoscopic methods are operator-dependent in the sense that specific expertise and training are obligatory for the success of any therapeutic intervention. Since relevant publications on hot biopsy forceps are still in favor of its careful use, as it has not yet demonstrated inferiority compared with newer techniques, it would be prudent for any medical practitioner to evaluate the available tools and judge any new proposed technique based on the evidence before it is adopted.

Diagnostic capabilities of high-definition white light endoscopy for the diagnosis of gastric intestinal metaplasia and correlation with histologic and clinical data.

OBJECTIVE: The aim of this study was the evaluation of the diagnostic accuracy of a specific high-definition white light endoscopy (HD-WLE) system for the optical recognition of intestinal metaplasia (IM) and the assessment of its correlation with histologic and clinical data. METHODS: A total of 234 patients undergoing upper gastrointestinal endoscopy in an outpatient endoscopy suite for various indications were prospectively enrolled in this cross-sectional study. Gastric IM was diagnosed on the basis of three mucosal patterns identified using HD-WLE in a per-patient analysis. Histological evaluation was used as the gold standard, and special staining was conducted for subtyping of IM. Main outcome measurements were sensitivity, specificity, and likelihood ratio of HD-WLE and secondary associations with histologic and clinical data. RESULTS: IM was found in 63/234 (27%) patients and low-grade dysplasia in 6/63 patients (9.5%). Sensitivity, specificity, accuracy, and likelihood ratio of all mucosal patterns were 74.6, 94, 88% and 13, respectively. All clinically significant type III IM and dysplasia lesions were endoscopically detected. All nonvisible lesions were of types I and II with mild grade and no dysplasia. Ten patients were considered false positives and the lesions were associated with severe inflammation and antralization. CONCLUSION: The specific HD-WLE system showed satisfactory accuracy and high specificity during real-time, routine endoscopy practice. Specific mucosal patterns were correlated with level and grade of lesions. The sensitivity of the system is even higher when only clinically significant IM lesions are considered.

Gastric Outlet Obstruction due to Cytomegalovirus Infection in an Immunocompromised Patient.

A case of gastric outlet obstruction accompanied by diffuse gastritis and gastric ulcer due to cytomegalovirus infection is presented. The patient, a woman aged 67 years with a long history of rheumatoid arthritis under immunosuppressive treatment (methotrexate), was admitted to our department complaining mostly of abdominal pain located in the epigastrium and the right abdomen. Upper gastrointestinal endoscopy revealed the presence of gastritis accompanied by ulcer in the prepyloric area and gastric outlet obstruction due to cytomegalovirus infection which was confirmed histologically. The patient responded well - although after many weeks - to specific treatment with antiviral treatment (ganciclovir). It is concluded that gastric outlet obstruction caused by cytomegalovirus infection can be observed in immunocompromised patients. The clinician must search for possible cytomegalovirus infection in all immunocompromised patients presenting with a clinical picture of gastric outlet obstruction by obtaining enough biopsies and by asking the histopathologist to specifically stain the specimen for the presence of cytomegalovirus, especially if infection by Helicobacter pylori is not present.