Data mining procedures identified a predictive model for incident hypertension, which included innovative and easy-to-measure variables. The findings promise great applicability in screening settings and clinical practice.

No abstract

This article has an accompanying continuing medical education activity, also eligible for MOC credit, on page e14. Learning Objective: Upon completion of this CME activity successful learners will be able to describe benefits of using artificial intelligence for quality control in the colonoscopy setting. Scan the quick response (QR) code to the left with your mobile device to watch this article's video abstract and others. Don't have a QR code reader? Get one by searching 'QR Scanner' in your mobile device's app store.

Background: Endoscopic ultrasound (EUS) guided core needle biopsies (CNB) are increasingly being performed to diagnose solid pancreatic lesions. However, studies have been conflicting in terms of CNB improving diagnostic accuracy and procedural efficiency vs fine-needle aspiration (FNA), which this study aims to elucidate.Methods: Data were prospectively collected on consecutive patients with solid pancreatic or peripancreatic lesions at a single tertiary care center from November 2015 to November 2016 that underwent either FNA or CNB. Patient demographics, characteristics of lesions, diagnostic accuracy, final and follow-up pathology, use of rapid on-site evaluation (ROSE), complications, and procedure variables were obtained.Results: A total of 75 FNA and 48 CNB were performed; of these, 13 patients had both. Mean passes were lower with CNB compared to FNA (2.4 vs 2.9, P = .02). Use of ROSE was higher for FNA (97.3% vs 68.1%, P = .001). Mean procedure time was shorter with CNB (34.1 minutes vs 51.2 minutes, P = .02) and diagnostic accuracy was similar (89.2% vs 89.4%, P = .98). There was no difference in diagnostic accuracy when ROSE was performed for CNB vs not performed (93.5% vs 85.7%, P = .58).Additionally, diagnostic accuracy of combined FNA and CNB procedures was 92.3%, which was comparable to FNA (P = .73) or CNB (P = .52) alone.Conclusion: FNA and CNB had comparable safety and diagnostic accuracy. Use of CNB resulted in less number of passes and shorter procedure time as compared to FNA. Moreover, diagnostic accuracy for CNB with or without ROSE was similar.K E Y W O R D S core biopsy, endoscopic ultrasound, fine-needle aspiration, pancreatic cancer, pancreatic mass

A 67-year-old woman with a long-standing history of recurrent dysphagia and esophageal strictures failed to respond to aggressive antireflux management. She required multiple dilations for symptomatic strictures that were discovered throughout the esophagus. Intralesional, topical, and systemic glucocorticoid therapies were utilized without resolution in symptoms. Several years after initial presentation, histopathology ultimately demonstrated lichenoid features and a diagnosis of esophageal lichen planus (ELP) was confirmed. However, as her symptoms had already become significantly disabling with severe strictures that carried an increased risk of endoscopic complications with dilation, she ultimately decided to undergo an esophagectomy for definitive treatment. Moreover, ELP may often go unrecognized for several years. Clinicians should consider ELP in the differential for dysphagia in middle- to elderly-aged women with or without a known history of lichen planus (LP) especially for those with findings of multiple or proximal strictures.

Eleven cases of bilateral extradural haematomas were encountered in a total of 49 cases of extradural haematoma during a period of 4 years among 1000 consecutive cases of acute head injury. Assault was held responsible for 60 per cent of the total head injuries. Most of the patients were seen late; 8 were already unconscious and 5 had dilated fixed pupils. All 11 patients died, 4 soon after admission. Nine cases were found to have associated brain damage at autopsy. In 2 cases bilateral extradural haematomas with compression of the brain were the only intracranial abnormalities. In 4 cases a clost was missed on conventional exploration.

Abstract. Most classification methods assume that the samples are drawn independently and identically from an unknown data generating distribution, yet this assumption is violated in several real life problems. In order to relax this assumption, we consider the case where batches or groups of samples may have internal correlations, whereas the samples from different batches may be considered to be uncorrelated. Two algorithms are developed to classify all the samples in a batch jointly, one based on a probabilistic analysis and another based on a mathematical programming approach. Experiments on three real-life computer aided diagnosis (CAD) problems demonstrate that the proposed algorithms are significantly more accurate than a naive SVM which ignores the correlations among the samples.

Abstract. We propose a new approach to test selection based on the discovery of subgroups of patients sharing the same optimal test, and present its application to breast cancer diagnosis. Subgroups are defined in terms of background information about the patient. We automatically determine the best t subgroups a patient belongs to, and decide for the test proposed by their majority. We introduce the concept of prediction quality to measure how accurate the test outcome is regarding the disease status. The quality of a subgroup is then the best mean prediction quality of its members (choosing the same test for all). Incorporating the quality computation in the search heuristic enables a significant reduction of the search space. In experiments on breast cancer diagnosis data we showed that it is faster than the baseline algorithm APRIORI-SD while preserving its accuracy.