An association rule-based method to support medical image diagnosis with efficiency

In this paper, we propose a method based on association rule-mining to enhance the diagnosis of medical images (mammograms). It combines low-level features automatically extracted from images and high-level knowledge from specialists to search for patterns. Our method analyzes medical images and automatically generates suggestions of diagnoses employing mining of association rules. The suggestions of diagnosis are used to accelerate the image analysis performed by specialists as well as to provide them an alternative to work on. The proposed method uses two new algorithms, PreSAGe and HiCARe. The PreSAGe algorithm combines, in a single step, feature selection and discretization, and reduces the mining complexity. Experiments performed on PreSAGe show that this algorithm is highly suitable to perform feature selection and discretization in medical images. HiCARe is a new associative classifier. The HiCARe algorithm has an important property that makes it unique: it assigns multiple keywords per image to suggest a diagnosis with high values of accuracy. Our method was applied to real datasets, and the results show high sensitivity (up to 95%) and accuracy (up to 92%), allowing us to claim that the use of association rules is a powerful means to assist in the diagnosing task.

Two-step tuberculin skin test and booster phenomenon prevalence among Brazilian medical students

SETTING: Five medical schools in three cities in Rio de Janeiro State, Brazil, with different tuberculosis (TB) incidence rates. OBJECTIVE: To evaluate the prevalence of the booster phenomenon and its associated factors in a voting universally BCG-vaccinated TB-exposed population. DESIGN: A two-step tuberculin skin test (TST) was performed among undergraduate medical students. Boosting was defined as an induration >= 10 mm in the second TST (TST2), with an increase of at least 6 mm over the first TST (TST1). The association of boosting with independent variables was evaluated using multivariate analysis. RESULTS: Of the 764 participants (mean age 21.9 +/- 2.7 years), 672 (87.9%) had a BCG scar. The overall booster SUMMARY phenomenon prevalence was 8.4% (95%CI 6.5-10.6). Boosting was associated with TST1 reactions of 1-9 mm (aOR 2.5, 95%CI 1.04-5.9) and with BCG vaccination, mostly after infancy, i.e., after age two years (aOR 9.1, 95%,CI 1.2-70.7). CONCLUSION: The prevalence of the booster phenomenon was high. A two-step TST in young BCG-vaccinated populations, especially in those with TST1 reactions of 1-9 mm, can avoid misdiagnosis as a false conversion and potentially reduce unnecessary treatment for latent TB infection.

Knowledge and practices of medical students to prevent tuberculosis transmission in Rio de Janeiro, Brazil

Objectives. To describe knowledge, practices, and associated factors of medical students to prevent transmission of tuberculosis (TB) in five medical schools. Methods. Cross-sectional survey of undergraduate medical students in preclinical and in early and late clinical years. Information was obtained on sociodemographic profile, previous lectures on TB, knowledge about TB transmission, exposure to patients with active pulmonary TB, and use of respiratory protective masks. Results. Among 1 094 respondents, 575 (52.6%) correctly answered that coughing, speaking, and sneezing can transmit TB. Early [adjusted odds ratio = 4.0 (3.0, 5.5)] and late [adjusted odds ratio = 4.2 (3.1, 5.8)] clinical years were associated with correct answers, but having had previous lectures on TB was not. Among those who had previous lectures on TB, the rate of correct answers increased from 42.1% to 61.6%. Among 332 medical students who reported exposure to TB patients, 194 (58.4%) had not used protective masks. More years of clinical experience was associated with the use of masks [adjusted odds ratio = 2.9 (1.4, 6.1)], while knowledge was inversely associated with the use of masks [adjusted odds ratio = 0.4 (0.2, 0.6)]. Conclusions. Many medical students are not aware of the main routes of TB infection, and lectures on TB are not sufficient to change knowledge and practices. Regardless of knowledge about TB transmission, students engage in risky behaviors: more than two-thirds do not use a protective mask when examining an active TB case. We suggest innovative, effective active learning experiences to change this scenario.

From medical librarian to informacionist: semantic traces of their profiles and areas of performance

This literature review retakes the discussion of the profiles and competences of the information area professional, specifically, in the health field. Therefore, the aim here is to outline the new fields of performance for the informationist and the profiles required in the health context.