Educators have long been trying to motivate students in their introductory programming courses. There are key concepts that are not easy to teach like abstraction, function and reuse. We believe that the visual nature of game-oriented approach to teaching anything is more effective for students. This work describes our experience in the use of Lightbot to introduce students to these concepts. This approach has been implemented in two Universities in CS1: Universidad Nacional de San Agustín and Universidad La Salle, from Arequipa -- Perú. One of them uses Java and the other one uses Python as first programming languages. Lightbot has been used in both of them since 2015 in the first laboratory session

Actualmente las herramientas de Inteligencia Artificial (IA) han surgido de forma disruptiva afectando a campos como el de las tecnologías de la información, específicamente al desarrollo de software, para el cual herramientas como ChatGPT abren nuevas oportunidades y desafíos en la enseñanza y el aprendizaje de conceptos de programación. Actualmente, esta se basa en métodos pedagógicos convencionales y consideramos la adaptación esencial pues estas herramientas de IA cuentan con un potencial que aún está en exploración. Integrar las herramientas de IA en la educación requiere una comprensión profunda de sus capacidades y limitaciones, por eso, en este artículo describimos nuestra experiencia en la adopción de un enfoque de aula invertida para la enseñanza de tópicos básicos de programación aplicada sobre estudiantes de primer año de universidad.

Objective: To help improve radiologists’ efficacy of disease diagnosis in reading computed tomography (CT) images, this study aims to investigate the feasibility of applying a modified deep learning (DL) method as a new strategy to automatically segment disease-infected regions and predict disease severity. Methods: We employed a public dataset acquired from 20 COVID-19 patients, which includes manually annotated lung and infections masks, to train a new ensembled DL model that combines five customized residual attention U-Net models to segment disease infected regions followed by a Feature Pyramid Network model to predict disease severity stage. To test the potential clinical utility of the new DL model, we conducted an observer comparison study. First, we collected another set of CT images acquired from 80 COVID-19 patients and process images using the new DL model. Second, we asked two chest radiologists to read images of each CT scan and report the estimated percentage of the disease-infected lung volume and disease severity level. Third, we also asked radiologists to rate acceptance of DL model-generated segmentation results using a 5-scale rating method. Results: Data analysis results show that agreement of disease severity classification between the DL model and radiologists is >90% in 45 testing cases. Furthermore, >73% of cases received a high rating score (≥4) from two radiologists. Conclusion: This study demonstrates the feasibility of developing a new DL model to automatically segment disease-infected regions and quantitatively predict disease severity, which may help avoid tedious effort and inter-reader variability in subjective assessment of disease severity in future clinical practice.

Assessment of the percentage of disease infected lung volume using computed tomography (CT) images can play an important role to detect lung diseases and predict disease severity. However, manual segmentation of disease infected regions from many CT image slices is tedious and not feasible in clinical practice. To help solve this clinical challenge, this study aims to investigate a new strategy to automatically segment disease infected regions and predict disease severity. We employed a public dataset acquired from 20 COVID-19 patients, which includes manually annotated lung and infections masks, to train a new ensembled deep learning (DL) model that combines the five customized residual attention UNet models to segment disease infected regions followed by a Feature Pyramid Network (FPN) model to classify severity stage of COVID-19 infection. To test potentially clinical utility of new model, we first gathered and processed another set of CT images acquired from 80 Covid-19 patients. Next, we asked two chest radiologists to read CT images of each patient and report the estimated percentage of infected lung volume and disease severity level. Additionally, we asked radiologists to rate acceptance of DL model-generated segmentation results using a 5-scale rating method. Data analysis results show that agreement between disease severity classification is >90% in 45 testing cases. Furthermore, >73% of cases received the high rating score from two radiologists (scored more than 4). This study demonstrates feasibility of developing a new DL-model to efficiently provide quantitative assessment of disease severity based on the automated segmentation of the disease infected regions to support improving efficacy of radiologists in disease diagnosis.