Downloads & Free Reading Options - Results

The application of deep learning in radiology has markedly improved diagnostic performance; however, widespread clinical adoption is hindered by the opaque, black-box nature of these models, which limits interpretability and undermines trust among healthcare professionals. This study introduces an explainable deep learning framework for brain tumor classification using magnetic resonance imaging (MRI). A convolutional neural network (CNN) was trained and validated on a curated dataset comprising four diagnostic categories: glioma, meningioma, pituitary tumor, and normal brain scans. To address the interpretability challenge, Gradient-weighted Class Activation Mapping (Grad-CAM) was employed to generate visual explanations highlighting the regions most influential to the model’s predictions. The framework achieved high quantitative performance across key metrics, including accuracy, precision, recall, and F1-score. In addition, qualitative assessments by radiologists confirmed that the Grad-CAM visualizations provided clinically meaningful insights, aligning with known diagnostic landmarks and improving trust in the model’s outputs. These findings underscore the value of integrating explainability into deep learning systems for medical imaging, paving the way for safer, more transparent, and clinically acceptable AI assisted diagnostics. 

Education is a dynamic process that involves transferring continuously evolving knowledge and skills to the next generation of future professionals. Professional education requires constant adaptation and updating of curricula to meet the specific needs of new generations of learners. Generation Z has been selected for this study as the majority of current university students belong to this generation. While some members of Generation Z have already completed their degrees and are working in healthcare, the younger portion of the generation is either still in university or about to enter. While there is some data on the specifics of Generation Z among nursing students, there seems to be insufficient knowledge about the potentially different learning preferences of other non-medical healthcare students.

It´s a systematic educational review about virtual reality in interventional radiology. The objective is to compare the traditional learning methods to virtual reality in this area. Eligible studies are gonna be evaluated using the Buckley questionnaire - BEME GUIDE nº11and the Kirkpatrick´s model described by Steinert - BEME GUIDE nº8. It would be included studies that compare traditional learning methods versus virtual reality in interventional radiology, with no restriction over origin, language, publication status of the study, or the population analyzed.

It´s a systematic educational review about virtual reality in interventional radiology. The objective is to compare the traditional learning methods to virtual reality in this area. Eligible studies are gonna be evaluated using the Buckley questionnaire - BEME GUIDE nº11and the Kirkpatrick´s model described by Steinert - BEME GUIDE nº8. It would be included studies that compare traditional learning methods versus virtual reality in interventional radiology, with no restriction over origin, language, publication status of the study, or the population analyzed.

There is considerable interest in the application of artificial intelligence in radiology, primarily related to rapid improvement of deep learning algorithms. Diagnostic accuracy studies that examine the use of deep learning algorithms in radiology should ensure full reporting of essential information, so that the readers can accurately assess their quality. Reporting guidelines have been established for diagnostic accuracy studies in general, such as the Standards for Reporting Diagnostic Accuracy Studies (STARD) 2015. The purpose of this study is to evaluate the adherence of deep learning imaging diagnostic accuracy studies to STARD 2015. This can help establish a baseline of adherence, prompt collaboration among computer scientists and radiologists in improving reporting adherence, and provide direction for development of a deep learning specific STARD extension.

In a hospital, clinical teachers must also care for patients, so there is less time for the teaching of clinical courses, or for discussing clinical cases with interns. However, electronic learning (e-learning) can complement clinical skills education for interns in a blended-learning process. Students discuss and interact with classmates in an e-learning collaborative environment. E-learning can assist clinical training and provides a collaborative environment, but every student has individual learning preferences on the e-learning platform. A typical platform, such as a learning management system (LMS) does not provide individual learning activities for every student. This paper clusters students into two groups: active and inactive groups. In each group, students' learning behavior patterns, i.e., the association rules for activities, are derived from the transaction data for the LMS. The cluster to which a student belongs defines the online learning behaviors, from the activity association rules. The method then provides individual preferred activities. Teachers instruct students in accordance with their aptitude, as derived from the learning behavior pattern. The cluster analysis shows that students in active group often view teaching videos after completing feedback. Students in the inactive group often view teaching materials after adding posts on a forum. [For the complete proceedings, see ED557189.]

Perma.cc archive of https://www.youtube.com/watch?v=BCIlzcVo5bE created on 2022-03-31 16:58:01.593768+00:00.

Perma.cc archive of https://www.youtube.com/watch?v=BCIlzcVo5bE created on 2022-03-31 16:58:01.593768+00:00.