Predicting students' academic progress and related attributes in first-year medical students: an analysis with artificial neural networks and Naïve Bayes.

BACKGROUND: Dropout and poor academic performance are persistent problems in medical schools in emerging economies. Identifying at-risk students early and knowing the factors that contribute to their success would be useful for designing educational interventions. Educational Data Mining (EDM) methods can identify students at risk of poor academic progress and dropping out. The main goal of this study was to use machine learning models, Artificial Neural Networks (ANN) and Naïve Bayes (NB), to identify first year medical students that succeed academically, using sociodemographic data and academic history. METHODS: Data from seven cohorts (2011 to 2017) of admitted medical students to the National Autonomous University of Mexico (UNAM) Faculty of Medicine in Mexico City were analysed. Data from 7,976 students (2011 to 2017 cohorts) of the program were included. Information from admission diagnostic exam results, academic history, sociodemographic characteristics and family environment was used. The main dataset included 48 variables. The study followed the general knowledge discovery process: pre-processing, data analysis, and validation. Artificial Neural Networks (ANN) and Naïve Bayes (NB) models were used for data mining analysis. RESULTS: ANNs models had slightly better performance in accuracy, sensitivity, and specificity. Both models had better sensitivity when classifying regular students and better specificity when classifying irregular students. Of the 25 variables with highest predictive value in the Naïve Bayes model, percentage of correct answers in the diagnostic exam was the best variable. CONCLUSIONS: Both ANN and Naïve Bayes methods can be useful for predicting medical students' academic achievement in an undergraduate program, based on information of their prior knowledge and socio-demographic factors. Although ANN offered slightly superior results, Naïve Bayes made it possible to obtain an in-depth analysis of how the different variables influenced the model. The use of educational data mining techniques and machine learning classification techniques have potential in medical education.

The synthetic parasite-derived peptide GK1 increases survival in a preclinical mouse melanoma model.

PURPOSE: The therapeutic efficacy of a synthetic parasite-derived peptide GK1, an immune response booster, was evaluated in a mouse melanoma model. This melanoma model correlates with human stage IIb melanoma, which is treated with wide surgical excision; a parallel study employing a surgical treatment was carried out as an instructive goal. EXPERIMENTAL DESIGN: C57BL/6 mice were injected subcutaneously in the flank with 2×10(5) B16-F10 murine melanoma cells. When the tumors reached 20 mm3, mice were separated into two different groups; the GK1 group, treated weekly with peritumoral injections of GK1 (10 µg/100 µL of sterile saline solution) and the control group, treated weekly with an antiseptic peritumoral injection of 100 µL of sterile saline solution without further intervention. All mice were monitored daily for clinical appearance, tumor size, and survival. Surgical treatment was performed in parallel when the tumor size was 20 mm3 (group A), 500 mm3 (group B), and >500 mm3 (group C). RESULTS: The GK1 peptide effectively increased the mean survival time by 9.05 days, corresponding to an increase of 42.58%, and significantly delayed tumor growth from day 3 to 12 of treatment. In addition, tumor necrosis was significantly increased (p<0.05) in the treated mice. The overall survival rates obtained with surgical treatment at 6 months were 83.33% for group A, 40% for group B, and 0% for group C, with significant differences (p<0.05) among the groups. CONCLUSIONS: The GK1 peptide demonstrated therapeutic properties in a mouse melanoma model, as treatment resulted in a significant increase in the mean survival time of the treated animals (42.58%). The potential for GK1 to be used as a primary or adjuvant component of chemotherapeutic cocktails for the treatment of experimental and human cancers remains to be determined, and surgical removal remains a challenge for any new experimental treatment of melanoma in mouse models.