Source-free active domain adaptation for diabetic retinopathy grading based on ultra-wide-field fundus images

Domain adaptation (DA) is commonly employed in diabetic retinopathy (DR) grading using unannotated fundus images, allowing knowledge transfer from labeled color fundus images. Existing DAs often struggle with domain disparities, hindering DR grading performance compared to clinical diagnosis. A source-free active domain adaptation method (SFADA), which generates features of color fundus images by noise, selects valuable ultra-wide-field (UWF) fundus images through local representation matching, and adapts models using DR lesion prototypes, is proposed to upgrade DR diagnostic accuracy. Importantly, SFADA enhances data security and patient privacy by excluding source domain data. It reduces image resolution and boosts model training speed by modeling DR grade relationships directly. Experiments show SFADA significantly improves DR grading performance, increasing accuracy by 20.90% and quadratic weighted kappa by 18.63% over baseline, reaching 85.36% and 92.38%, respectively. This suggests SFADA's promise for real clinical applications.

Active learning for left ventricle segmentation in echocardiography

BACKGROUND AND OBJECTIVE

Training deep learning models for medical image segmentation require large annotated datasets, which can be expensive and time-consuming to create. Active learning is a promising approach to reduce this burden by strategically selecting the most informative samples for segmentation. This study investigates the use of active learning for efficient left ventricle segmentation in echocardiography with sparse expert annotations.

METHODS

We adapt and evaluate various sampling techniques, demonstrating their effectiveness in judiciously selecting samples for segmentation. Additionally, we introduce a novel strategy, Optimised Representativeness Sampling, which combines feature-based outliers with the most representative samples to enhance annotation efficiency.

RESULTS

Our findings demonstrate a substantial reduction in annotation costs, achieving a remarkable 99% upper bound performance while utilising only 20% of the labelled data. This equates to a reduction of 1680 images needing annotation within our dataset. When applied to a publicly available dataset, our approach yielded a remarkable 70% reduction in required annotation efforts, representing a significant advancement compared to baseline active learning strategies, which achieved only a 50% reduction. Our experiments highlight the nuanced performance of diverse sampling strategies across datasets within the same domain.

CONCLUSIONS

The study provides a cost-effective approach to tackle the challenges of limited expert annotations in echocardiography. By introducing a distinct dataset, made publicly available for research purposes, our work contributes to the field's understanding of efficient annotation strategies in medical image segmentation.

Teaching Patients About Pain: The Emergence of Pain Science Education, its Learning Frameworks and Delivery Strategies

Since it emerged in the early 2000's, intensive education about 'how pain works', widely known as pain neuroscience education or explaining pain, has evolved into a new educational approach, with new content and new strategies. The substantial differences from the original have led the PETAL collaboration to call the current iteration 'Pain Science Education'. This review presents a brief historical context for Pain Science Education, the clinical trials, consumer perspective, and real-world clinical data that have pushed the field to update both content and method. We describe the key role of educational psychology in driving this change, the central role of constructivism, and the constructivist learning frameworks around which Pain Science Education is now planned and delivered. We integrate terminology and concepts from the learning frameworks currently being used across the PETAL collaboration in both research and practice-the Interactive, Constructive, Active, Passive framework, transformative learning theory, and dynamic model of conceptual change. We then discuss strategies that are being used to enhance learning within clinical encounters, which focus on the skill, will, and thrill of learning. Finally, we provide practical examples of these strategies so as to assist the reader to drive their own patient pain education offerings towards more effective learning. PERSPECTIVE: Rapid progress in several fields and research groups has led to the emergence 'Pain Science Education'. This PETAL review describes challenges that have spurred the field forward, the learning frameworks and educational strategies that are addressing those challenges, and some easy wins to implement and mistakes to avoid.

You'll have to call the attending: Impact of a longitudinal, "real-time" case-based infectious diseases elective on entrustable professional activities to enhance APPE readiness

BACKGROUND AND PURPOSE

Transitioning from the didactic to experiential setting is challenging for student pharmacists, perhaps due to lack of experiences providing "real-time" clinician interaction. We describe findings from a semester-long infectious diseases (ID) didactic elective that utilized a national cohort of preceptors and faculty across the United States to mimic clinician interaction and "real-time" ID management of various disease states. The mechanics of this elective provide a framework for others to implement to enhance advanced pharmacy practice experience (APPE) readiness.

EDUCATION ACTIVITY AND SETTING

Students enrolled in an ID elective course at a school of pharmacy participated in "real-time" acute care scenarios. They assisted in multidisciplinary management of a patient's infection, mimicking "rounds" on an APPE, via interaction with external pharmacist volunteers (playing the roles of other healthcare personnel). Additionally, students formally presented and discussed their cases within the class, further promoting learning while optimizing presentation skills. Pharmacist volunteers were surveyed to assess student performances as measured by four entrustable professional activities (EPAs).

FINDINGS

A total of 48 volunteer opportunities occurred during two course offerings. Results from 43 surveys were analyzed (90% response rate). Of those responses, 22/24 (92%) played the role of attending physician, and 19/24 (79%) played the role of technician. Volunteers agreed that students met the four EPAs evaluated (agreement was 85-100%).

SUMMARY

This semester-long elective provided "real-time" experience and feedback for pre-APPE students to enhance APPE readiness and reinforce EPAs. Students are likely to benefit from mimicked intra-professional interaction and augmented critical thinking skills that could be adapted to various disease states within pharmacy curricula.

Deep learning model for pleural effusion detection via active learning and pseudo-labeling: a multisite study

BACKGROUND

The study aimed to develop and validate a deep learning-based Computer Aided Triage (CADt) algorithm for detecting pleural effusion in chest radiographs using an active learning (AL) framework. This is aimed at addressing the critical need for a clinical grade algorithm that can timely diagnose pleural effusion, which affects approximately 1.5 million people annually in the United States.

METHODS

In this multisite study, 10,599 chest radiographs from 2006 to 2018 were retrospectively collected from an institution in Taiwan to train the deep learning algorithm. The AL framework utilized significantly reduced the need for expert annotations. For external validation, the algorithm was tested on a multisite dataset of 600 chest radiographs from 22 clinical sites in the United States and Taiwan, which were annotated by three U.S. board-certified radiologists.

RESULTS

The CADt algorithm demonstrated high effectiveness in identifying pleural effusion, achieving a sensitivity of 0.95 (95% CI: [0.92, 0.97]) and a specificity of 0.97 (95% CI: [0.95, 0.99]). The area under the receiver operating characteristic curve (AUC) was 0.97 (95% DeLong's CI: [0.95, 0.99]). Subgroup analyses showed that the algorithm maintained robust performance across various demographics and clinical settings.

CONCLUSION

This study presents a novel approach in developing clinical grade CADt solutions for the diagnosis of pleural effusion. The AL-based CADt algorithm not only achieved high accuracy in detecting pleural effusion but also significantly reduced the workload required for clinical experts in annotating medical data. This method enhances the feasibility of employing advanced technological solutions for prompt and accurate diagnosis in medical settings.

Computed tomography-based automated measurement of abdominal aortic aneurysm using semantic segmentation with active learning

Accurate measurement of abdominal aortic aneurysm is essential for selecting suitable stent-grafts to avoid complications of endovascular aneurysm repair. However, the conventional image-based measurements are inaccurate and time-consuming. We introduce the automated workflow including semantic segmentation with active learning (AL) and measurement using an application programming interface of computer-aided design. 300 patients underwent CT scans, and semantic segmentation for aorta, thrombus, calcification, and vessels was performed in 60-300 cases with AL across five stages using UNETR, SwinUNETR, and nnU-Net consisted of 2D, 3D U-Net, 2D-3D U-Net ensemble, and cascaded 3D U-Net. 7 clinical landmarks were automatically measured for 96 patients. In AL stage 5, 3D U-Net achieved the highest dice similarity coefficient (DSC) with statistically significant differences (p < 0.01) except from the 2D-3D U-Net ensemble and cascade 3D U-Net. SwinUNETR excelled in 95% Hausdorff distance (HD95) with significant differences (p < 0.01) except from UNETR and 3D U-Net. DSC of aorta and calcification were saturated at stage 1 and 4, whereas thrombus and vessels were continuously improved at stage 5. The segmentation time between the manual and AL-corrected segmentation using the best model (3D U-Net) was reduced to 9.51 ± 1.02, 2.09 ± 1.06, 1.07 ± 1.10, and 1.07 ± 0.97 min for the aorta, thrombus, calcification, and vessels, respectively (p < 0.001). All measurement and tortuosity ratio measured - 1.71 ± 6.53 mm and - 0.15 ± 0.25. We developed an automated workflow with semantic segmentation and measurement, demonstrating its efficiency compared to conventional methods.

Focused active learning for histopathological image classification

Active Learning (AL) has the potential to solve a major problem of digital pathology: the efficient acquisition of labeled data for machine learning algorithms. However, existing AL methods often struggle in realistic settings with artifacts, ambiguities, and class imbalances, as commonly seen in the medical field. The lack of precise uncertainty estimations leads to the acquisition of images with a low informative value. To address these challenges, we propose Focused Active Learning (FocAL), which combines a Bayesian Neural Network with Out-of-Distribution detection to estimate different uncertainties for the acquisition function. Specifically, the weighted epistemic uncertainty accounts for the class imbalance, aleatoric uncertainty for ambiguous images, and an OoD score for artifacts. We perform extensive experiments to validate our method on MNIST and the real-world Panda dataset for the classification of prostate cancer. The results confirm that other AL methods are 'distracted' by ambiguities and artifacts which harm the performance. FocAL effectively focuses on the most informative images, avoiding ambiguities and artifacts during acquisition. For both experiments, FocAL outperforms existing AL approaches, reaching a Cohen's kappa of 0.764 with only 0.69% of the labeled Panda data.

GANDALF: Graph-based transformer and Data Augmentation Active Learning Framework with interpretable features for multi-label chest Xray classification

Informative sample selection in an active learning (AL) setting helps a machine learning system attain optimum performance with minimum labeled samples, thus reducing annotation costs and boosting performance of computer-aided diagnosis systems in the presence of limited labeled data. Another effective technique to enlarge datasets in a small labeled data regime is data augmentation. An intuitive active learning approach thus consists of combining informative sample selection and data augmentation to leverage their respective advantages and improve the performance of AL systems. In this paper, we propose a novel approach called GANDALF (Graph-based TrANsformer and Data Augmentation Active Learning Framework) to combine sample selection and data augmentation in a multi-label setting. Conventional sample selection approaches in AL have mostly focused on the single-label setting where a sample has only one disease label. These approaches do not perform optimally when a sample can have multiple disease labels (e.g., in chest X-ray images). We improve upon state-of-the-art multi-label active learning techniques by representing disease labels as graph nodes and use graph attention transformers (GAT) to learn more effective inter-label relationships. We identify the most informative samples by aggregating GAT representations. Subsequently, we generate transformations of these informative samples by sampling from a learned latent space. From these generated samples, we identify informative samples via a novel multi-label informativeness score, which beyond the state of the art, ensures that (i) generated samples are not redundant with respect to the training data and (ii) make important contributions to the training stage. We apply our method to two public chest X-ray datasets, as well as breast, dermatology, retina and kidney tissue microscopy MedMNIST datasets, and report improved results over state-of-the-art multi-label AL techniques in terms of model performance, learning rates, and robustness.

Dataset of mathematics learning and assessment of higher education students using the MathE platform

Higher education institutions are promoting the adoption of innovative methodologies and instructional approaches to engage and promote personalized learning paths to their students. Several strategies based on gamification, artificial intelligence, and data mining are adopted to create an interactive educational setting centred around students. Within this personalized learning environment, there is a notable boost in student engagement and enhanced educational outcomes. The MathE platform, an online educational system introduced in 2019, is specifically crafted to support students tackling difficulties in comprehending higher-education-level mathematics or those aspiring to deepen their understanding of diverse mathematical topics - all at their own pace. The MathE platform provides multiple-choice questions, categorized under topics and subtopics, aligning with the content taught in higher education courses. Accessible to students worldwide, the platform enables them to train their mathematical skills through these resources. When the students log in to the training area of the platform, they choose a topic to study and specify whether they prefer basic or advanced questions. The platform then selects a set of seven multiple-choice questions from the available ones under the chosen topic and generates a test for the student. After completing and submitting the test, the answers are recorded and stored on the platform. This paper describes the data stored in the MathE platform, focusing on the 9546 answers to 833 questions, provided by 372 students from 8 countries who use the platform to practice their skills using the questions (and other resources) available on the platform. The information in this paper will help research about active learning tools to support the improvement of future education, especially at higher educational level. Furthermore, these data are valuable for understanding student learning patterns, assessing platform efficacy, gaining a global perspective on mathematics education, and contributing to the advancement of active learning tools for higher education.

The effect of flipped learning on nursing students' Asepsis knowledge and learning skills: A randomized controlled study

AIM

This study was conducted to evaluate the effect of the flipped learning model on nursing students' asepsis knowledge and learning skills.

BACKGROUND

The flipped learning model enables students to pursue their learning with online support whenever and wherever they want. Students have the responsibility for their learning activities. The flipped learning model is an effective method to improve nursing students' knowledge and skills related to the principles of asepsis with online innovative approaches.

DESIGN

This study has a pre-test post-test open-label, randomized controlled design.

METHOD

The study sample consisted of 107 first-year nursing students randomized into experimental (n = 53) and control (n = 54) groups. The experimental group students were trained utilizing the flipped learning model. The data were collected through the "Descriptive Characteristics Form of Nursing Students ", the "Principles of Asepsis Knowledge Test" and the " Self-directed Learning Skills Scale".

RESULTS

It was determined that the post-test knowledge score of the experimental group was statistically significantly higher (p=0.000) than the control group and the median of the retention test knowledge score was statistically significantly higher (p=0.000) than the control group. There was a statistically significant increase (p<0.05) in the median score of the self-directed learning skills scale "self-control" sub-dimension of the experimental group.

CONCLUSION

Flipped learning increased nursing students' knowledge related to the principles of asepsis and enabled them to take responsibility for learning. This model had a positive effect on students' higher order thinking skills such as critical organization and decision making. It is recommended to use the flipped learning within the scope of nursing education and especially in gaining basic skills.

TWEETABLE ABSTRACT

A success in today's education; flipped learning.

Active Learning in Brain Tumor Segmentation with Uncertainty Sampling and Annotation Redundancy Restriction

Deep learning models have demonstrated great potential in medical imaging but are limited by the expensive, large volume of annotations required. To address this, we compared different active learning strategies by training models on subsets of the most informative images using real-world clinical datasets for brain tumor segmentation and proposing a framework that minimizes the data needed while maintaining performance. Then, 638 multi-institutional brain tumor magnetic resonance imaging scans were used to train three-dimensional U-net models and compare active learning strategies. Uncertainty estimation techniques including Bayesian estimation with dropout, bootstrapping, and margins sampling were compared to random query. Strategies to avoid annotating similar images were also considered. We determined the minimum data necessary to achieve performance equivalent to the model trained on the full dataset (α = 0.05). Bayesian approximation with dropout at training and testing showed results equivalent to that of the full data model (target) with around 30% of the training data needed by random query to achieve target performance (p = 0.018). Annotation redundancy restriction techniques can reduce the training data needed by random query to achieve target performance by 20%. We investigated various active learning strategies to minimize the annotation burden for three-dimensional brain tumor segmentation. Dropout uncertainty estimation achieved target performance with the least annotated data.

Debating: effective and satisfactory learning method in dentistry

BACKGROUND

Education in the modern world of health needs diverse methods of learning and teaching. The traditional education model has limited capacity for developing abilities such as critical thinking, problem-solving, and reasoning skills. Therefore, improving the quality of teaching-learning processes requires implementing educational innovations in the classroom and evaluating them. This study aimed to determine the impact of the debate teaching method on improving the abilities of general dentistry doctoral students.

METHODS

The research was a semi-experimental study with pre-tests and post-tests to measure the knowledge and abilities of students. The study included 60 dental students who completed the fall 2022 session of the Community Oral Health (COH) 2 practical course. This course, one of three practical components within the Community Oral Health curriculum, aligns with the educational framework of general dentistry. Challenging topics on which there is no consensus in dentistry were chosen for the debate. The descriptive statistics indicators include an independent t-test and variance analysis test with a significance level of 5%. Were used to analyze the data.

RESULTS

The results of the study showed that the average total knowledge (P < 0.001), 'perception of critical thinking skills (P < 0.001), expression power (P < 0.001), reasoning skills (P = 0.003), interpretation and Information analysis power (P < 0.001), the ability to find and use scientific databases (P < 0.001) and the ability to analyze and evaluate evidence (P < 0.001) increased significantly after intervention in students. 95% of students agreed/strongly agreed that this method enhances their ability to answer people's questions. From an instructor's point of view, students had 93.1% of the ability to reason and analyze information after intervention and 88.5% of the ability to think critically.

CONCLUSION

The results of the study showed that the use of debate in the classroom is an effective way to present content. The process of evaluating data-driven arguments promotes higher-level cognitive skills and teaches students about the knowledge base and the use of scientific databases.

TRIAL REGISTRATION

Registration date: 21/11/2022, Registration number: IRCT20141128020129N3.

Assessment and treatment of visuospatial neglect using active learning with Gaussian processes regression

Visuospatial neglect is a disorder characterised by impaired awareness for visual stimuli located in regions of space and frames of reference. It is often associated with stroke. Patients can struggle with all aspects of daily living and community participation. Assessment methods are limited and show several shortcomings, considering they are mainly performed on paper and do not implement the complexity of daily life. Similarly, treatment options are sparse and often show only small improvements. We present an artificial intelligence solution designed to accurately assess a patient's visuospatial neglect in a three-dimensional setting. We implement an active learning method based on Gaussian process regression to reduce the effort it takes a patient to undergo an assessment. Furthermore, we describe how this model can be utilised in patient oriented treatment and how this opens the way to gamification, tele-rehabilitation and personalised healthcare, providing a promising avenue for improving patient engagement and rehabilitation outcomes. To validate our assessment module, we conducted clinical trials involving patients in a real-world setting. We compared the results obtained using our AI-based assessment with the widely used conventional visuospatial neglect tests currently employed in clinical practice. The validation process serves to establish the accuracy and reliability of our model, confirming its potential as a valuable tool for diagnosing and monitoring visuospatial neglect. Our VR application proves to be more sensitive, while intra-rater reliability remains high.

Engaging Student Pharmacists in Social Determinants of Health and Health Equity Through Photovoice

OBJECTIVE

This study aimed to evaluate the impact of an interactive photovoice activity on the perceptions of social determinants of health (SDOH) and health equity among first professional year student pharmacists.

METHODS

This study used a mixed-methods exploratory approach at 4 institutions. All students completed a standardized intervention using a prerecorded lecture, active learning using photovoice, and an in-depth debriefing session. The photovoice responses and reflections were analyzed through a deductive approach using content analysis with the applied frameworks of Rolfe's reflection model and the social-ecological model. A presurvey/postsurvey assessed the students' perceptions of SDOH and health equity. Paired sample t tests were conducted to assess the prechange and postchange.

RESULTS

A total of 349 students participated; most students reflected at the "what" level (97.7%), whereas 65% reached the "now what" level. Students identified more SDOH factors at the institutional/community level (75.9%) than at the individual/interpersonal level (59.4%) or the society/policy level (28.0%); 191 (55%) students had matchable survey data. A statistically significant improvement was found in the comprehension of health equity concepts (4 items), perceptions of health disparities and system response (4 items), awareness of structural factors impacting equity (3 items), and readiness for inclusivity behavior (3 items).

CONCLUSION

A structured teaching and learning activity allowed deeper reflections among student pharmacists. Student perception of the basic terminologies and the impact of beliefs on health care improved after the photovoice assignment. Although students became aware of the SDOH, they had difficulty identifying the structural or upstream factors when addressing SDOH.

Detecting goals of care conversations in clinical notes with active learning

OBJECTIVE

Goals of care (GOC) discussions are an increasingly used quality metric in serious illness care and research. Wide variation in documentation practices within the Electronic Health Record (EHR) presents challenges for reliable measurement of GOC discussions. Novel natural language processing approaches are needed to capture GOC discussions documented in real-world samples of seriously ill hospitalized patients' EHR notes, a corpus with a very low event prevalence.

METHODS

To automatically detect sentences documenting GOC discussions outside of dedicated GOC note types, we proposed an ensemble of classifiers aggregating the predictions of rule-based, feature-based, and three transformers-based classifiers. We trained our classifier on 600 manually annotated EHR notes among patients with serious illnesses. Our corpus exhibited an extremely imbalanced ratio between sentences discussing GOC and sentences that do not. This ratio challenges standard supervision methods to train a classifier. Therefore, we trained our classifier with active learning.

RESULTS

Using active learning, we reduced the annotation cost to fine-tune our ensemble by 70% while improving its performance in our test set of 176 EHR notes, with 0.557 F1-score for sentence classification and 0.629 for note classification.

CONCLUSION

When classifying notes, with a true positive rate of 72% (13/18) and false positive rate of 8% (13/158), our performance may be sufficient for deploying our classifier in the EHR to facilitate bedside clinicians' access to GOC conversations documented outside of dedicated notes types, without overburdening clinicians with false positives. Improvements are needed before using it to enrich trial populations or as an outcome measure.

Bridging the gap with grad: Integrating active learning into semi-supervised domain generalization

Domain generalization (DG) aims to generalize from a large amount of source data that are fully annotated. However, it is laborious to collect labels for all source data in practice. Some research gets inspiration from semi-supervised learning (SSL) and develops a new task called semi-supervised domain generalization (SSDG). Unlabeled source data is trained jointly with labeled one to significantly improve the performance. Nevertheless, different research adopts different settings, leading to unfair comparisons. Moreover, the initial annotation of unlabeled source data is random, causing unstable and unreliable training. To this end, we first specify the training paradigm, and then leverage active learning (AL) to handle the issues. We further develop a new task called Active Semi-supervised Domain Generalization (ASSDG), which consists of two parts, i.e., SSDG and AL. We delve deep into the commonalities of SSL and AL and propose a unified framework called Gradient-Similarity-based Sample Filtering and Sorting (GSSFS) to iteratively train the SSDG and AL parts. Gradient similarity is utilized to select reliable and informative unlabeled source samples for these two parts respectively. Our methods are simple yet efficient, and extensive experiments demonstrate that our methods can achieve the best results on the DG datasets in the low-data regime without bells and whistles.

The SAFE procedure: a practical stopping heuristic for active learning-based screening in systematic reviews and meta-analyses

Active learning has become an increasingly popular method for screening large amounts of data in systematic reviews and meta-analyses. The active learning process continually improves its predictions on the remaining unlabeled records, with the goal of identifying all relevant records as early as possible. However, determining the optimal point at which to stop the active learning process is a challenge. The cost of additional labeling of records by the reviewer must be balanced against the cost of erroneous exclusions. This paper introduces the SAFE procedure, a practical and conservative set of stopping heuristics that offers a clear guideline for determining when to end the active learning process in screening software like ASReview. The eclectic mix of stopping heuristics helps to minimize the risk of missing relevant papers in the screening process. The proposed stopping heuristic balances the costs of continued screening with the risk of missing relevant records, providing a practical solution for reviewers to make informed decisions on when to stop screening. Although active learning can significantly enhance the quality and efficiency of screening, this method may be more applicable to certain types of datasets and problems. Ultimately, the decision to stop the active learning process depends on careful consideration of the trade-off between the costs of additional record labeling against the potential errors of the current model for the specific dataset and context.

Addressing the challenges of reconstructing systematic reviews datasets: a case study and a noisy label filter procedure

Systematic reviews and meta-analyses typically require significant time and effort. Machine learning models have the potential to enhance screening efficiency in these processes. To effectively evaluate such models, fully labeled datasets-detailing all records screened by humans and their labeling decisions-are imperative. This paper presents the creation of a comprehensive dataset for a systematic review of treatments for Borderline Personality Disorder, as reported by Oud et al. (2018) for running a simulation study. The authors adhered to the PRISMA guidelines and published both the search query and the list of included records, but the complete dataset with all labels was not disclosed. We replicated their search and, facing the absence of initial screening data, introduced a Noisy Label Filter (NLF) procedure using active learning to validate noisy labels. Following the NLF application, no further relevant records were found. A simulation study employing the reconstructed dataset demonstrated that active learning could reduce screening time by 82.30% compared to random reading. The paper discusses potential causes for discrepancies, provides recommendations, and introduces a decision tree to assist in reconstructing datasets for the purpose of running simulation studies.

Preschoolers ask questions about unknown words on video chat and in live interactions at similar rates

The COVID-19 pandemic has led to a major increase in digital interactions in early experience. A crucial question, given expanding virtual platforms, is whether preschoolers' active word learning behaviors extend to their interactions over video chat. When not provided with sufficient information to link new words to meanings, preschoolers drive their word learning by asking questions. In person, 5-year-olds focus their questions on unknown words compared with known words, highlighting their active word learning. Here, we investigated whether preschoolers' question-asking over video chat differs from in-person question-asking. In the study, 5-year-olds were instructed to move toys in response to known and unknown verbs on a video conferencing call (i.e., Zoom). Consistent with in-person results, video chat participants (n = 18) asked more questions about unknown words than about known words. The rate of question-asking about words across video chat and in-person formats did not differ. Differences in the types of questions asked about words indicate, however, that although video chat does not hinder preschoolers' active word learning, the use of video chat may influence how preschoolers request information about words.

Traditional or adaptive design of experiments? A pilot-scale comparison on wood delignification

Traditional design of experiments and response surface methodology are widely used in engineering and process development. Bayesian optimization is an alternative machine learning approach that adaptively selects successive experimental conditions based on a predefined performance measure. Here we compared the two approaches using simulations and empirical experiments on alkaline wood delignification to identify important benefits and drawbacks of Bayesian optimization in the context of design of experiments. The simulations showed that the selection of initial experiments and measurement noise influenced the convergence of the Bayesian optimization algorithm to known optimal conditions. Both methods, however, showed comparable pilot-scale results on optimal digestion conditions, where high cellulose yields were combined with acceptable kappa numbers and pulp viscosities. Bayesian optimization did not enable a decrease in the number of experiments required for reaching these conditions but provided a more accurate model in the vicinity of the optimum based on additional modelling and cross-validation. These results shed light on the practical differences between the two methodologies for process development and are an important contribution to the chemometrics and machine learning communities.

Hybrid representation-enhanced sampling for Bayesian active learning in musculoskeletal segmentation of lower extremities

PURPOSE

Manual annotations for training deep learning models in auto-segmentation are time-intensive. This study introduces a hybrid representation-enhanced sampling strategy that integrates both density and diversity criteria within an uncertainty-based Bayesian active learning (BAL) framework to reduce annotation efforts by selecting the most informative training samples.

METHODS

The experiments are performed on two lower extremity datasets of MRI and CT images, focusing on the segmentation of the femur, pelvis, sacrum, quadriceps femoris, hamstrings, adductors, sartorius, and iliopsoas, utilizing a U-net-based BAL framework. Our method selects uncertain samples with high density and diversity for manual revision, optimizing for maximal similarity to unlabeled instances and minimal similarity to existing training data. We assess the accuracy and efficiency using dice and a proposed metric called reduced annotation cost (RAC), respectively. We further evaluate the impact of various acquisition rules on BAL performance and design an ablation study for effectiveness estimation.

RESULTS

In MRI and CT datasets, our method was superior or comparable to existing ones, achieving a 0.8% dice and 1.0% RAC increase in CT (statistically significant), and a 0.8% dice and 1.1% RAC increase in MRI (not statistically significant) in volume-wise acquisition. Our ablation study indicates that combining density and diversity criteria enhances the efficiency of BAL in musculoskeletal segmentation compared to using either criterion alone.

CONCLUSION

Our sampling method is proven efficient in reducing annotation costs in image segmentation tasks. The combination of the proposed method and our BAL framework provides a semi-automatic way for efficient annotation of medical image datasets.

Twelve tips for designing and implementing peer-led assessment writing schemes in health professions education

Peer-led assessment (PLA) has gained increasing prominence within health professions education as an effective means of engaging learners in the process of assessment writing and practice. Involving students in various stages of the assessment lifecycle, including item writing, quality assurance, and feedback, not only facilitates the creation of high-quality item banks with minimal faculty input but also promotes the development of students' assessment literacy and fosters their growth as teachers. The advantages of involving students in the generation of assessments are evident from a pedagogical standpoint, benefiting both students and faculty. However, faculty members may face uncertainty when it comes to implementing such approaches effectively. To address this concern, this paper presents twelve tips that offer guidance on important considerations for the successful implementation of peer-led assessment schemes in the context of health professions education.

Systematic review using a spiral approach with machine learning

With the accelerating growth of the academic corpus, doubling every 9 years, machine learning is a promising avenue to make systematic review manageable. Though several notable advancements have already been made, the incorporation of machine learning is less than optimal, still relying on a sequential, staged process designed to accommodate a purely human approach, exemplified by PRISMA. Here, we test a spiral, alternating or oscillating approach, where full-text screening is done intermittently with title/abstract screening, which we examine in three datasets by simulation under 360 conditions comprised of different algorithmic classifiers, feature extractions, prioritization rules, data types, and information provided (e.g., title/abstract, full-text included). Overwhelmingly, the results favored a spiral processing approach with logistic regression, TF-IDF for vectorization, and maximum probability for prioritization. Results demonstrate up to a 90% improvement over traditional machine learning methodologies, especially for databases with fewer eligible articles. With these advancements, the screening component of most systematic reviews should remain functionally achievable for another one to two decades.

Automated method development in high-pressure liquid chromatography

Method development in liquid chromatography is a crucial step in the optimization of analytical separations for various applications. However, it is often a challenging endeavour due to its time-consuming, resource intensive and costly nature, which is further hampered by its complexity requiring highly skilled and experienced scientists. This review presents an examination of the methods that are required for a completely automated method development procedure in liquid chromatography, aimed at taking the human out of the decision loop. Some of the presented approaches have recently witnessed an important increase in interest as they offer the promise to facilitate, streamline and speed up the method development process. The review first discusses the mathematical description of the separation problem by means of multi-criteria optimization functions. Two different strategies to resolve this optimization are then presented; an experimental and a model-based approach. Additionally, methods for automated peak detection and peak tracking are reviewed, which, upon integration in an instrument, allow for a completely closed-loop method development process. For each of these approaches, various currently applied methods are presented, recent trends and approaches discussed, short-comings pointed out, and future prospects highlighted.

A new practical approach using TeamSTEPPS strategies and tools: - an educational design

BACKGROUND

Teamwork has played a critical role in ensuring patients' safety and preventing human errors in surgery. With advancements in educational technologies, including virtual reality, it is necessary to develop new teaching methods for interpersonal teamwork based on local needs assessments in countries with indigenous cultures. This study aimed to design and develop a new method of teaching teamwork in cesarean section surgery using virtual reality; we further evaluated the effects of this method on healthcare professionals' knowledge and attitudes about teamwork.

METHODS

This study was designed using the ADDIE instructional design model. The TeamSTEPPS Learning Benchmarks questionnaire was used to assess the educational needs of 85 participants who were members of the cesarean section surgery team. A specialized panel analyzed the extracted needs, and the scenario was compiled during the design stage. Finally, four virtual reality contents were created using 360-video H.265 format, which were prepared from specified scenarios in the development of the educational program. The TeamSTEPPS Learning Benchmarks questionnaire was used to measure knowledge, and the T-TAQ was used to measure the participants' attitudes.

RESULTS

Six micro- skills were identified as training needs, including briefing, debriefing, cross-monitoring, I'M SAFE checklist, call-out and check-back, and two-challenge rule. Intervention results showed that the virtual reality content improved teamwork competencies in an interprofessional team performing cesarean section surgery. A significant increase was observed in the mean score of knowledge and attitude after the intervention.

CONCLUSION

Through addressing the need for teamwork training, utilizing the TeamSTEPPS strategy, and incorporating new educational technologies like virtual reality, the collaboration among surgical team members can be enhanced.

Team-based learning (TBL) in clinical disciplines for undergraduate medical students-a scoping review

BACKGROUND

Team-based learning (TBL) is an evidence-based pedagogical method that has been used in undergraduate medical education since 2001. However, its use in clinical disciplines is rarely reported, and the impact of its implementation is not known. The aim of this study was to explore and map the published literature on the impact of implementing TBL in clinical disciplines in undergraduate medical education.

METHODS

A comprehensive search of Medline, Education Resources Information Center (ERIC), and Web of Science databases was performed on November 24, 2021 and updated April 6, 2023, using relevant Medical Subject Headings (MeSH) and free-text terms. Original research studies reporting on the implementation of TBL in clinical disciplines in undergraduate medical education published in peer-reviewed English language journals were included irrespective of their methodological design.

RESULTS

The initial search identified 2,383 records. Of these, 49 met the inclusion criteria. Most of the studies (n = 44, 90%) described the implementation of a modified version of TBL in which one or more TBL steps were missing, and one study had undefined protocol for the implementation. The most reported outcomes were knowledge acquisition (n = 38, 78%) and students' satisfaction or attitudes toward TBL (n = 34, 69%). Despite some differences in their results, the studies found that implementing TBL is associated with increased knowledge acquisition (n = 19, 39%), student engagement (n = 6, 12%), and student satisfaction (n = 31, 63%).

CONCLUSIONS

Most of the studies reported positive results in students' satisfaction and students' engagement, whilst the results on knowledge acquisition and retention were more contradictory. In most of the studies, TBL was implemented in a modified form and diverse comparators were used. The methodological quality also varied. Thus, no unequivocal conclusions could be drawn regarding the value of implementing TBL in clinical disciplines. More studies with rigorous methodologies are needed in this field.

Bayesian Optimization in Drug Discovery

Drug discovery deals with the search for initial hits and their optimization toward a targeted clinical profile. Throughout the discovery pipeline, the candidate profile will evolve, but the optimization will mainly stay a trial-and-error approach. Tons of in silico methods have been developed to improve and fasten this pipeline. Bayesian optimization (BO) is a well-known method for the determination of the global optimum of a function. In the last decade, BO has gained popularity in the early drug design phase. This chapter starts with the concept of black box optimization applied to drug design and presents some approaches to tackle it. Then it focuses on BO and explains its principle and all the algorithmic building blocks needed to implement it. This explanation aims to be accessible to people involved in drug discovery projects. A strong emphasis is made on the solutions to deal with the specific constraints of drug discovery. Finally, a large set of practical applications of BO is highlighted.

Fine-tuning coreference resolution for different styles of clinical narratives

OBJECTIVE

Coreference resolution (CR) is a natural language processing (NLP) task that is concerned with finding all expressions within a single document that refer to the same entity. This makes it crucial in supporting downstream NLP tasks such as summarization, question answering and information extraction. Despite great progress in CR, our experiments have highlighted a substandard performance of the existing open-source CR tools in the clinical domain. We set out to explore some practical solutions to fine-tune their performance on clinical data.

METHODS

We first explored the possibility of automatically producing silver standards following the success of such an approach in other clinical NLP tasks. We designed an ensemble approach that leverages multiple models to automatically annotate co-referring mentions. Subsequently, we looked into other ways of incorporating human feedback to improve the performance of an existing neural network approach. We proposed a semi-automatic annotation process to facilitate the manual annotation process. We also compared the effectiveness of active learning relative to random sampling in an effort to further reduce the cost of manual annotation.

RESULTS

Our experiments demonstrated that the silver standard approach was ineffective in fine-tuning the CR models. Our results indicated that active learning should also be applied with caution. The semi-automatic annotation approach combined with continued training was found to be well suited for the rapid transfer of CR models under low-resource conditions. The ensemble approach demonstrated a potential to further improve accuracy by leveraging multiple fine-tuned models.

CONCLUSION

Overall, we have effectively transferred a general CR model to a clinical domain. Our findings based on extensive experimentation have been summarized into practical suggestions for rapid transferring of CR models across different styles of clinical narratives.

Machine Learning for Biological Design

We briefly present machine learning approaches for designing better biological experiments. These approaches build on machine learning predictors and provide additional tools to guide scientific discovery. There are two different kinds of objectives when designing better experiments: to improve the predictive model or to improve the experimental outcome. We survey five different approaches for adaptive experimental design that iteratively search the space of possible experiments while adapting to measured data. The approaches are Bayesian optimization, bandits, reinforcement learning, optimal experimental design, and active learning. These machine learning approaches have shown promise in various areas of biology, and we provide broad guidelines to the practitioner and links to further resources.

Active help-seeking and metacognition interact in supporting children's retention of science facts

Determining when to ask for help is a critical self-regulated strategy that can benefit children's learning. Despite its importance, we have a limited understanding about the developmental mechanisms that support adaptive help-seeking. In the current preregistered study, predominately White children aged 8 to 13 years (N = 69, ngirls = 37) had the option to seek help during an online science learning task. Results revealed that children's ability to adaptively seek help improved throughout childhood and early adolescence. Critically, developing metacognitive skills contributed to greater help-related memory benefits (compared with conditions where help was not previously available). Overall, these findings highlight the role of metacognition in children's ability to adaptively seek and benefit from help during online science learning.

How people learn: insights for medical faculty

To increase medical students' and residents' understanding and retention, faculty need to teach from a knowledge standpoint and understanding of how individuals learn. We know from cognitive information processing that learners remember only a small portion of what they read or hear but remember up to 90% of information when strong active learning modalities are included. Faculty also need to be aware of different learning styles-kinesthetic, visual, and auditory-and ensure that they are including methods that can reach all learners. The cognitive and information processing theories of learning provide insights to educators related to building on prior knowledge from learning and limiting the number of points taught so learners can process and retain the information. Strategies such as a flipped classroom model and question clicker technology can assist in reaching learning goals. Fundamental conditions for learning include awareness, interest, motivation, relevance, engagement, reinforcement, and support.

Active learning for medical image segmentation with stochastic batches

The performance of learning-based algorithms improves with the amount of labelled data used for training. Yet, manually annotating data is particularly difficult for medical image segmentation tasks because of the limited expert availability and intensive manual effort required. To reduce manual labelling, active learning (AL) targets the most informative samples from the unlabelled set to annotate and add to the labelled training set. On the one hand, most active learning works have focused on the classification or limited segmentation of natural images, despite active learning being highly desirable in the difficult task of medical image segmentation. On the other hand, uncertainty-based AL approaches notoriously offer sub-optimal batch-query strategies, while diversity-based methods tend to be computationally expensive. Over and above methodological hurdles, random sampling has proven an extremely difficult baseline to outperform when varying learning and sampling conditions. This work aims to take advantage of the diversity and speed offered by random sampling to improve the selection of uncertainty-based AL methods for segmenting medical images. More specifically, we propose to compute uncertainty at the level of batches instead of samples through an original use of stochastic batches (SB) during sampling in AL. Stochastic batch querying is a simple and effective add-on that can be used on top of any uncertainty-based metric. Extensive experiments on two medical image segmentation datasets show that our strategy consistently improves conventional uncertainty-based sampling methods. Our method can hence act as a strong baseline for medical image segmentation. The code is available on: https://github.com/Minimel/StochasticBatchAL.git.

Undergraduate nursing students' perceptions of active learning strategies: A focus group study

BACKGROUND

Active learning strategies have been identified as promoting critical thinking, strengthening clinical reasoning, and supporting the transfer of theoretical knowledge to practice amongst nursing students.

AIM

This study aimed to understand the undergraduate nursing students' perceptions of the active learning strategies being used in the classroom and to identify critical elements within their learning spaces which contribute to their learning.

DESIGN

Qualitative, focus group study.

SETTING

A four-year undergraduate baccalaureate nursing program in the Middle East.

PARTICIPANTS

50 undergraduate nursing students selected through purposive and snowball sampling participated in the study.

METHODS

Five focus group sessions were conducted with 10 participants in each session. Data collected from the discussions were transcribed and thematically analyzed and aligned with the Taxonomy of Significant Learning.

RESULTS

Study results show that undergraduate nursing students affirm that the use of active learning strategies supports the acquisition of foundational understanding, application and integration of knowledge, caring about the learning process, learning to learn, and the human dimension of learning. Participants also identified how best active learning strategies should be utilized and aspects of learning spaces that promote learning.

CONCLUSIONS

Although the use of active learning strategies positively enhances the learning process, it is important to ensure that strategies are intentionally integrated into the classroom and aligned with the expected learning outcomes. Considerations of the learning space used are also of importance.

The positive impact of introducing modified directed self-learning using pre-small group discussion worksheets as an active learning strategy in undergraduate medical education

BACKGROUND

Directed self-learning (DSL) is an active learning approach where the learners are provided with predefined learning objectives and some facilitation through the learning process in the form of guidance and supervision. It can help establish a strong foundation for autonomous and deep learning.

OBJECTIVE

The aim of this study was to introduce a modified form of DSL to second-year undergraduate medical students using pre-small group discussion (pre-SGD) worksheets. The authors intended to evaluate its effectiveness through theme assessment and investigate students' perceptions using a feedback questionnaire.

METHODS

This was an analytical cross-sectional study. Modified DSL (MDSL) was introduced to 96 second-year undergraduate medical students in two themes. Students were divided randomly into two groups. One group was exposed to traditional DSL (TDSL), and the other was introduced to MDSL using pre-SGD worksheets for the first theme. Groups were reversed for the second theme. The activity was followed by a theme assessment, which was scored for research purpose only. The scores of this assessment were compared, and perceptions of the students were gathered using a validated questionnaire. Data were analyzed using IBM's statistical package of social sciences (SPSS) version 22.

RESULTS

The comparison of theme assessment scores revealed statistically significant difference (P = 0.002) in median scores between control TDSL and experimental MDSL groups. The percentage of students scoring ≥80% in theme assessment was significantly higher in the experimental group compared to the control group (P = 0.029). This strategy was well perceived by the students in terms of acceptability and effectiveness as depicted by a high degree of agreement on the Likert-scale.

CONCLUSION

Modified DSL resulted in significant improvement in academic performance of undergraduate medical students. MDSL was also well perceived as an active learning strategy in terms of acceptability, effectiveness, and comparison with TDSL. [Figure: see text].

Evaluating uncertainty-based active learning for accelerating the generalization of molecular property prediction

Deep learning models have proven to be a powerful tool for the prediction of molecular properties for applications including drug design and the development of energy storage materials. However, in order to learn accurate and robust structure-property mappings, these models require large amounts of data which can be a challenge to collect given the time and resource-intensive nature of experimental material characterization efforts. Additionally, such models fail to generalize to new types of molecular structures that were not included in the model training data. The acceleration of material development through uncertainty-guided experimental design has the promise to significantly reduce the data requirements and enable faster generalization to new types of materials. To evaluate the potential of such approaches for electrolyte design applications, we perform comprehensive evaluation of existing uncertainty quantification methods on the prediction of two relevant molecular properties - aqueous solubility and redox potential. We develop novel evaluation methods to probe the utility of the uncertainty estimates for both in-domain and out-of-domain data sets. Finally, we leverage selected uncertainty estimation methods for active learning to evaluate their capacity to support experimental design.

Usefulness of Mobile Devices in Learning Process for Residents of Pediatric Surgical Specialties

INTRODUCTION

In the hospital setting, the use of mobile devices among surgical residents is increasing. To assess the usefulness of mobile devices for residents of pediatric surgical specialties.

MATERIALS AND METHODS

The study used a mixed-method design. First, a self-developed online questionnaire with 23 items was used to obtain quantitative data, which was analyzed via simple discriminant analysis. Qualitative data were obtained using the focus group technique with the subsequent triangulation of quantitative and qualitative data.

RESULTS

The residents used mobile devices for learning and communication. Using quantitative data, the study found that the major functions of mobile devices were communicating with other residents and taking clinical photos, and that for learning, were speaking with attendings, residents, collecting patient information, and searching for unfamiliar terms. Triangulation analysis confirmed that mobile devices aid in agile communication, the search for data on drugs or diseases, and consultation of medical applications. Qualitative data informed the limitations of devices and the inconsistencies between official regulations and their advantages in clinical practice.

CONCLUSIONS

We demonstrate the usefulness of mobile devices among surgical residents in clinical care and recommend that hospitals should regulate policies to maximize the use of mobile devices.

Revolutionizing drug formulation development: The increasing impact of machine learning

Over the past few years, the adoption of machine learning (ML) techniques has rapidly expanded across many fields of research including formulation science. At the same time, the use of lipid nanoparticles to enable the successful delivery of mRNA vaccines in the recent COVID-19 pandemic demonstrated the impact of formulation science. Yet, the design of advanced pharmaceutical formulations is non-trivial and primarily relies on costly and time-consuming wet-lab experimentation. In 2021, our group published a review article focused on the use of ML as a means to accelerate drug formulation development. Since then, the field has witnessed significant growth and progress, reflected by an increasing number of studies published in this area. This updated review summarizes the current state of ML directed drug formulation development, introduces advanced ML techniques that have been implemented in formulation design and shares the progress on making self-driving laboratories a reality. Furthermore, this review highlights several future applications of ML yet to be fully exploited to advance drug formulation research and development.

Factors affecting students' perception of flipped learning over time in a teacher training program

The flipped learning methodology could play a key role in teacher training, as it exposes future teachers to experience this active methodology as students. With the purpose of shedding light on how students' perceptions may vary over time and how they can be related to the improvement of the flipped learning methodology, our study explores different factors in an eight-year period. Specifically, we analyse teaching performance considering data on students' perceptions from the 2015-2016 academic year to 2022-2023 of a course embedded within a master s degree in teacher training in Spain. Once future teachers had experienced flipped learning as students, a sample of 338 completed a survey regarding their perceptions of the flipped classroom approach and the instructor role. In our study, the more experienced the instructor, the better perception the students showed on both the flipped learning methodology and the performance of their teacher. In particular, we found that future teachers had (i) a good or very good opinion about flipped learning, regardless of their gender (ii) a more positive perception about flipped learning, teaching performance and course development in the last five academic years, (iii) no remarkable differences between study specialisations in those last academic years, and (iv) a better opinion about the flipped learning model when they have best grades. We discuss our findings according to six factors that affect the flipped learning experience and, thus, students' perception of flipped learning over time: "student characteristics", "teacher characteristics", "implementation", "task characteristics", "out-of-class activities" and "in-class activities"-factors already unveiled by a recent state-of-the-art review to enhance the effectiveness of flipped classroom. We can conclude that the instructor's teaching experience is a key factor that affects the implementation of flipped learning, influencing students' perception and, consequently, the success of this active methodology.

Encouraging Student Attendance and Engagement in Lectures & Workshops in the Pre- and Post-Covid World

In the post-Covid era, second year chemistry lectures are fully flipped with all content being online. All the live lecture sessions are used for group work and are fully interactive. Students have agency in the lectures by directing what is taught in these student-led sessions. Students find the sessions very engaging and respond positively. In particular they value the agency they are given. In a second study that took place pre-Covid, workshops are changed from 1-hour to 2-hour sessions but with half the number and a much simplified timetable for students. Group work and peer-assessment with marking criteria help make the sessions engaging for students and more useful. The increased level of attendance from less than 20% to more than 70% (in the best case) is evidence of increased value to the students and success of the new format.

Strengthening E-learning strategies for active learning in crisis situations: a mixed-method study in the COVID-19 pandemic

BACKGROUND

Medical universities are responsible for educating and training healthcare workers. One of the fields significantly impacted by the pandemic is medical education. The aim of this study is to identify strategies for enhancing e-learning for active learning and finding solutions for improving its quality.

METHODS

This mixed-method (quantitative-qualitative) research was conducted in 2023 at three selected universities in Mazandaran Province. In the quantitative phase, 507 students participated via stratified random sampling using a standard questionnaire. In the qualitative phase, data were collected through semi-structured interviews with 16 experts until data saturation was achieved. SPSS 21 and MAXQDA 10 software were used for data analysis.

RESULTS

In the multivariate regression analysis, an increase of one point in the dimensions of student-teacher interaction, active time, immediate feedback, and active learning corresponded to an average increase in learning scores of 0.11, 0.17, 0.16, and 1.42 respectively (p≤0.001). After the final analysis in the qualitative phase, four main domains (infrastructure, resources, quantity of education, and quality of education) and 16 sub-domains with 84 items were identified.

CONCLUSIONS

The greatest challenge in e-learning is the interaction and cooperation between students and teachers. The implementation of the identified strategies in this research could provide useful evidence for policymakers and educational administrators to implement interventions aimed at addressing deficiencies and enhancing e-learning.

Active learning based on similarity level histogram and adaptive-scale sampling for very high resolution image classification

In remote sensing image classification, active learning aims to obtain an excellent classification model by selecting informative or representative training samples. However, due to the complexity of remote sensing images, the same class of ground objects usually have different spectral representations. The existing active learning methods may not take into account diverse representations of the same targets, which leads to a possible lack of intra-class diversity in the collected samples. To alleviate this problem, we propose an active learning method based on similarity level histogram (SLH) and adaptive-scale sampling to improve very high resolution remote sensing image classification. Specifically, we construct a SLH for each class of ground objects to effectively consider the intra-class diversity of the same target. To avoid the problem of sample imbalance caused by over-sampling or under-sampling, we design an adaptive-scale sampling strategy. Then, we utilize active learning to mine representative samples from each SLH warehouse according to adaptive-scale sampling strategies until the iteration condition is satisfied. Experiments show that the proposed algorithm can achieve better classification performance with limited training samples and is competitive with other methods based on four sets of publicly available data.

Visual pollution real images benchmark dataset on the public roads

The term quality of life (QoL) refers to a wide range of multifaceted concepts that often involve subjective assessments of both positive and negative aspects of life. It is difficult to quantify QoL as the word has varied meanings in different academic areas and may have different connotations in different circumstances. The five sectors most commonly associated with QoL, however, are Health, Education, Environmental Quality, Personal Security, Civic Engagement, and Work-Life Balance. An emerging issue that falls under environmental quality is visual pollution (VP) which, as detailed in this study, refers to disruptive presences that limit visual ability in public roads with an emphasis on excavation barriers, potholes, and dilapidated sidewalks. Quantifying VP has always been difficult due to its subjective nature and lack of a consistent set of rules for systematic assessment of visual pollution. This emphasizes the need for research and module development that will allow government agencies to automatically predict and detect VP. Our dataset was collected from different regions in the Kingdom of Saudi Arabia (KSA) via the Ministry of Municipal and Rural Affairs and Housing (MOMRAH) as a part of a VP campaign to improve Saudi Arabia's urban landscape. It consists of 34,460 RGB images separated into three distinct classes: excavation barriers, potholes, and dilapidated sidewalks. To annotate all images for detection (i.e., bounding box) and classification (i.e., classification label) tasks, the deep active learning strategy (DAL) is used where an initial 1,200 VP images (i.e., 400 images per class) are manually annotated by four experts. Images with more than one object increase the number of training object ROIs which are recorded to be 8,417 for excavation barriers, 25,975 for potholes, and 7,412 for dilapidated sidewalks. The MOMRAH dataset is publicly published to enrich the research domain with the new VP image dataset.

Equivalent Performance of Exam Items Associated with Case-Based Learning, Flipped Classroom, and Lecture in a Pre-clerkship Medical Curriculum

The purpose of our study was to determine if knowledge acquisition, as measured by exam item performance, differed for active or passive learning activities in our medical curriculum. Additionally, we looked for differences in exam item performance in one second-year course that varies the method of an active learning activity, case-based collaborative learning (CBCL). Finally, we assessed whether item performance was impacted when small group activities were conducted online due to the COVID-19 pandemic. Exam item difficulty values were collected for several years of lectures, flipped classroom, and CBCL. Statistical analysis and modeling of data were performed to identify differences in difficulty of exam items that assess content delivered by different learning activities. Our analysis revealed no differences in difficulty of exam items that assess content delivered by different learning activities. Similarly, we determined that varying the execution of CBCL in one course did not impact exam item performance. Finally, moving CBCL small group sessions online did not impact exam item difficulty. However, we did detect a minor reduction in overall exam scores for the period of online instruction. Our results indicate that knowledge acquisition, as assessed by our multiple-choice summative exams, was equivalent regardless of learning activity modality.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40670-023-01842-8.

The effect of online and face-to-face active learning methods on learning attitudes

AIM

This study aims to compare the effects of online and face-to-face education models using active learning methods on students' learning attitudes.

METHODS

This retrospective and quasi-experimental study included total 203 third-grade nursing students who took the Obstetrics and Gynecological Health Nursing course face-to-face and online in the fall semester of the 2019-2020 and 2020-2021. Active Learning Methods Questionnaire and the Scale Attitude Towards Learning (SATL) were used.

RESULTS

The mean score of active learning methods of the students receiving online education (77.35 ± 18.63) was higher than the face-to-face education group (67.00 ± 20.67). Although there was no difference between the face-to-face and online education groups in terms of the effort to learn and caring for learning, online students had a lower attitude towards learning avoidance (t: 6.540, p: 0.000). There was a negative and low-level significant correlation between the evaluation of active learning methods and the total score of SATL in the online education group (r = -0.200; p = .043), no correlation in face-to-face group (r: 0.004; p: 0.963).

CONCLUSION

Online education, which uses active learning methods, affected positively students' attitudes towards learning and decreased their avoidance of learning more than face-to-face education. This study has shown that active learning methods allow online students to question theoretical knowledge, convey it to patient care, contribute to clinical knowledge, and facilitate their learning and analytical thinking based on the more positively evaluation by online students.

Active learning with label quality control

Training deep neural networks requires a large number of labeled samples, which are typically provided by crowdsourced workers or professionals at a high cost. To obtain qualified labels, samples need to be relabeled for inspection to control the quality of the labels, which further increases the cost. Active learning methods aim to select the most valuable samples for labeling to reduce labeling costs. We designed a practical active learning method that adaptively allocates labeling resources to the most valuable unlabeled samples and the most likely mislabeled labeled samples, thus significantly reducing the overall labeling cost. We prove that the probability of our proposed method labeling more than one sample from any redundant sample set in the same batch is less than 1/k, where k is the number of the k-fold experiment used in the method, thus significantly reducing the labeling resources wasted on redundant samples. Our proposed method achieves the best level of results on benchmark datasets, and it performs well in an industrial application of automatic optical inspection.

Factors associated with utilization of student-centered pedagogy by nurse educators

BACKGROUND

Despite suboptimal level of the utilization of student-centered pedagogy, multidimensional, modifiable factors associated with the utilization have been rarely examined among nurse educators.

PURPOSE

The aim of this study was to examine the utilization status of student-centered pedagogy and factors associated with the utilization by nurse educators.

METHODS

Data on student-centered pedagogy utilization, knowledge, beliefs in effectiveness, stress, coping, support, degree earned, teaching experiences, and other demographic characteristics were analyzed using multiple regression analyses.

RESULTS

The status of the student-centered pedagogy utilization was moderate, and knowledge was consistently associated with the utilization in the total sample (N = 108) and in both subgroups (≤50 vs. >50 years old). Beliefs in effectiveness were associated with the utilization only in the age > 50 years old group.

CONCLUSIONS

Nurse educators need to develop and deliver interventions to facilitate the utilization of student-centered pedagogy through increase in knowledge and beliefs in effectiveness of student-centered pedagogy.

Active inductive inference in children and adults: A constructivist perspective

A defining aspect of being human is an ability to reason about the world by generating and adapting ideas and hypotheses. Here we explore how this ability develops by comparing children's and adults' active search and explicit hypothesis generation patterns in a task that mimics the open-ended process of scientific induction. In our experiment, 54 children (aged 8.97±1.11) and 50 adults performed inductive inferences about a series of causal rules through active testing. Children were more elaborate in their testing behavior and generated substantially more complex guesses about the hidden rules. We take a 'computational constructivist' perspective to explaining these patterns, arguing that these inferences are driven by a combination of thinking (generating and modifying symbolic concepts) and exploring (discovering and investigating patterns in the physical world). We show how this framework and rich new dataset speak to questions about developmental differences in hypothesis generation, active learning and inductive generalization. In particular, we find children's learning is driven by less fine-tuned construction mechanisms than adults', resulting in a greater diversity of ideas but less reliable discovery of simple explanations.

An active learning approach to train a deep learning algorithm for tumor segmentation from brain MR images

PURPOSE

This study focuses on assessing the performance of active learning techniques to train a brain MRI glioma segmentation model.

METHODS

The publicly available training dataset provided for the 2021 RSNA-ASNR-MICCAI Brain Tumor Segmentation (BraTS) Challenge was used in this study, consisting of 1251 multi-institutional, multi-parametric MR images. Post-contrast T1, T2, and T2 FLAIR images as well as ground truth manual segmentation were used as input for the model. The data were split into a training set of 1151 cases and testing set of 100 cases, with the testing set remaining constant throughout. Deep convolutional neural network segmentation models were trained using the NiftyNet platform. To test the viability of active learning in training a segmentation model, an initial reference model was trained using all 1151 training cases followed by two additional models using only 575 cases and 100 cases. The resulting predicted segmentations of these two additional models on the remaining training cases were then addended to the training dataset for additional training.

RESULTS

It was demonstrated that an active learning approach for manual segmentation can lead to comparable model performance for segmentation of brain gliomas (0.906 reference Dice score vs 0.868 active learning Dice score) while only requiring manual annotation for 28.6% of the data.

CONCLUSION

The active learning approach when applied to model training can drastically reduce the time and labor spent on preparation of ground truth training data.

CRITICAL RELEVANCE STATEMENT

Active learning concepts were applied to a deep learning-assisted segmentation of brain gliomas from MR images to assess their viability in reducing the required amount of manually annotated ground truth data in model training.

KEY POINTS

• This study focuses on assessing the performance of active learning techniques to train a brain MRI glioma segmentation model. • The active learning approach for manual segmentation can lead to comparable model performance for segmentation of brain gliomas. • Active learning when applied to model training can drastically reduce the time and labor spent on preparation of ground truth training data.

HAL-IA: A Hybrid Active Learning framework using Interactive Annotation for medical image segmentation

High performance of deep learning models on medical image segmentation greatly relies on large amount of pixel-wise annotated data, yet annotations are costly to collect. How to obtain high accuracy segmentation labels of medical images with limited cost (e.g. time) becomes an urgent problem. Active learning can reduce the annotation cost of image segmentation, but it faces three challenges: the cold start problem, an effective sample selection strategy for segmentation task and the burden of manual annotation. In this work, we propose a Hybrid Active Learning framework using Interactive Annotation (HAL-IA) for medical image segmentation, which reduces the annotation cost both in decreasing the amount of the annotated images and simplifying the annotation process. Specifically, we propose a novel hybrid sample selection strategy to select the most valuable samples for segmentation model performance improvement. This strategy combines pixel entropy, regional consistency and image diversity to ensure that the selected samples have high uncertainty and diversity. In addition, we propose a warm-start initialization strategy to build the initial annotated dataset to avoid the cold-start problem. To simplify the manual annotation process, we propose an interactive annotation module with suggested superpixels to obtain pixel-wise label with several clicks. We validate our proposed framework with extensive segmentation experiments on four medical image datasets. Experimental results showed that the proposed framework achieves high accuracy pixel-wise annotations and models with less labeled data and fewer interactions, outperforming other state-of-the-art methods. Our method can help physicians efficiently obtain accurate medical image segmentation results for clinical analysis and diagnosis.

Peer Instruction as an Alternative Active Learning Pedagogy Across the Pharmacy Curriculum

OBJECTIVE

The objective of this study was to determine if peer instruction (PI) is a useful active learning pedagogy to increase correct responses to pharmacotherapy concepts throughout didactic education in a Doctor of Pharmacy curriculum.

METHODS

Peer instruction was implemented into 3 pharmacy practice courses spanning 3 years of didactic pharmacy education at Cedarville University: Introduction to Self-Care (PHAR 6112) in the first professional year, Respiratory Module (PHAR 6261) in the second professional year, and Special Populations Module (PHAR 7343) in the third professional year. ConcepTests, which are multiple-choice questions written to help students apply previous knowledge to new scenarios, were re-polled based on a PI algorithm after peer discussion. Changes in students paired before and after peer discussion ConcepTest responses were analyzed using a McNemar test and descriptive statistics.

RESULTS

A total of 52 first-year students, 43 second-year students, and 49 third-year students participated in each respective course. Across all courses, an increase in the percentage of correct responses to ConceptTests after peer discussion was observed from the first polling (51.2%) to the second polling (90.4%). This increase in the percentage of correct responses was observed across all years of the curriculum, with greater increases in cohorts with previous participation in PI-based sessions.

CONCLUSION

The use of PI fostered improvement in the percentage of correct responses to ConcepTests focused on pharmacotherapy concepts throughout the first 3 years of didactic education. This pedagogy may be an effective and useful active learning strategy in pharmacy education that does not require significant classroom infrastructure changes.