Machine learning and medical education.
Ontology highlight
ABSTRACT:
SUBMITTER: Kolachalama VB
PROVIDER: S-EPMC6550167 | biostudies-literature | 2018
REPOSITORIES: biostudies-literature
Publications
Machine learning and medical education.
Kolachalama Vijaya B VB Garg Priya S PS
NPJ digital medicine 20180927
Artificial intelligence (AI) driven by machine learning (ML) algorithms is a branch in computer science that is rapidly gaining popularity within the healthcare sector. Recent regulatory approvals of AI-driven companion diagnostics and other products are glimmers of a future in which these tools could play a key role by defining the way medicine will be practiced. Educating the next generation of medical professionals with the right ML techniques will enable them to become part of this emerging ...[more]
Similar Datasets
Project description:Objective: In order to evaluate the quality of postgraduate medical education in Germany, we examined how the learning of theoretical and practical competencies is conceptualized and how the learning process takes place in real terms. The training conditions, as perceived by medical residents, are compared with the learning objectives, as stated by the Federal Chamber of Physicians in its regulation on postgraduate education. The analysis is based on the data of the "KarMed" study (abbreviation of "career progression and career breaks among physicians during postgraduate education"), a multicentre cohort study of medical school graduates from seven universities who arre followed until they receive their licenses as specialist physicians. The study was conducted by the Institute of Primary Medical Care of the University Medical Center Hamburg-Eppendorf starting in 2008. Methodology: The KarMed study is based on annual, standardized surveys of the population described above. 48% responded in the initial survey (n=1012) and 85% or more in each of the following surveys. Descriptive statistics and logistical regressions were used for analysis. Gender-specific analyses were performed where possible in order to highlight the differences in the professional objectives and workloads between female and male physicians. Results: The study shows that both the practical and the theoretical components of postgraduate medical education in Germany are insufficient. There is a lack of a curriculum with precise learning objectives and descriptions of the corresponding educational settings. In fact, the act of learning is identical to daily clinical work. There is no structuring of the work process with regard to learning; for example, documentation procedures and feedback discussions are prescribed but largely omitted. Evidence-based medicine is not a systematic part of the training, nor is the evaluation of residents' progress. The summative final oral examination pretends that the necessary specialist competencies can be evaluated within 30 minutes. Many factors indicate that female doctors, especially those with children, have fewer learning opportunities than male doctors. Conclusion: The quality of postgraduate medical education in Germany has become inadequate, especially in an international comparison. The deficits are well known. The responsible institutions are called upon to implement sustainable reforms in the sense of postgraduate education that is structured according to educational principles and whose quality is assured.
| S-EPMC5704602 | biostudies-other
Project description:BackgroundResidents receive a numeric performance rating (eg, 1-7 scoring scale) along with a narrative (ie, qualitative) feedback based on their performance in each workplace-based assessment (WBA). Aggregated qualitative data from WBA can be overwhelming to process and fairly adjudicate as part of a global decision about learner competence. Current approaches with qualitative data require a human rater to maintain attention and appropriately weigh various data inputs within the constraints of working memory before rendering a global judgment of performance.ObjectiveThis study explores natural language processing (NLP) and machine learning (ML) applications for identifying trainees at risk using a large WBA narrative comment data set associated with numerical ratings.MethodsNLP was performed retrospectively on a complete data set of narrative comments (ie, text-based feedback to residents based on their performance on a task) derived from WBAs completed by faculty members from multiple hospitals associated with a single, large, residency program at McMaster University, Canada. Narrative comments were vectorized to quantitative ratings using the bag-of-n-grams technique with 3 input types: unigram, bigrams, and trigrams. Supervised ML models using linear regression were trained with the quantitative ratings, performed binary classification, and output a prediction of whether a resident fell into the category of at risk or not at risk. Sensitivity, specificity, and accuracy metrics are reported.ResultsThe database comprised 7199 unique direct observation assessments, containing both narrative comments and a rating between 3 and 7 in imbalanced distribution (scores 3-5: 726 ratings; and scores 6-7: 4871 ratings). A total of 141 unique raters from 5 different hospitals and 45 unique residents participated over the course of 5 academic years. When comparing the 3 different input types for diagnosing if a trainee would be rated low (ie, 1-5) or high (ie, 6 or 7), our accuracy for trigrams was 87%, bigrams 86%, and unigrams 82%. We also found that all 3 input types had better prediction accuracy when using a bimodal cut (eg, lower or higher) compared with predicting performance along the full 7-point rating scale (50%-52%).ConclusionsThe ML models can accurately identify underperforming residents via narrative comments provided for WBAs. The words generated in WBAs can be a worthy data set to augment human decisions for educators tasked with processing large volumes of narrative assessments.
| S-EPMC9187970 | biostudies-literature
Project description:Learning Styles theory promises improved academic performance based on the identification of a personal, sensory preference for informational processing. This promise is not supported by evidence, and is in contrast to our current understanding of the neuroscience of learning. Despite this lack of evidence, prior research shows that that belief in the Learning Styles "neuromyth" remains high amongst educators of all levels, around the world. This perspective article is a follow up on prior research aimed at understanding why belief in the neuromyth of Learning Styles remains so high. We evaluated current research papers from the field of health professions education, to characterize the perspective that an educator would be given, should they search for evidence on Learning Styles. As in earlier research on Higher Education, we found that the use of Learning Style frameworks persist in education research for the health professions; 91% of 112 recent research papers published on Learning Styles are based upon the premise that Learning Styles are a useful approach to education. This is in sharp contrast to the fundamental principle of evidence-based practice within these professions. Thus any educator who sought out the research evidence on Learning Styles would be given a consistent but inaccurate endorsement of the value of a teaching technique that is not evidence based, possibly then propagating the belief in Learning Styles. Here we offer perspectives from both research and student about this apparent mismatch between educational practice and clinical practice, along with recommendations and considerations for the future.
| S-EPMC8385406 | biostudies-literature
Project description:BackgroundResidents need to be trained across the boundaries of their own specialty to prepare them for collaborative practice. Intraprofessional learning (i.e. between individuals of different disciplines within the same profession) has received little attention in the postgraduate medical education literature, in contrast to the extensive literature on interprofessional learning between individuals of different professions. To address this gap, we performed a scoping review to investigate what and how residents learn from workplace-related intraprofessional activities, and what factors influence learning.MethodsThe PRISMA guidelines were used to conduct a scoping review of empirical studies on intraprofessional workplace learning in postgraduate medical education published between 1 January 2000 to 16 April 2020 in Pubmed, Embase, PsycINFO, ERIC and Web of Science. This study applied 'best fit' framework-based synthesis to map the existing evidence, using the presage-process-product (3P) model developed by Tynjälä (2013).ResultsFour thousand three hundred thirty records were screened, and 37 articles were included. This review identified influencing (presage) factors that derived from the sociocultural environment, learner and learning context. Studies described that complexity of care can both facilitate and hinder learning. Furthermore, intraprofessional learning is threatened by professional stereotyping and negative perceptions, and awareness of learning opportunities and explicit reflection are critical in intraprofessional workplace learning. Studies described a range of informal and formal intraprofessional activities (process) under the headings of collaboration in clinical practice, rotations or placements, formal educational sessions and simulated workplace training. In general, learners responded well and their attitudes and perceptions improved, learners reported increased knowledge and skills and positive behavioural changes (product). Learning outcomes were reported in the domains of patient-centred care, collaborative attitudes and respect, mutual knowledge and understanding, collaborative decision making, communication, leadership, teamwork and reflexivity.ConclusionsThis review gives insight into the high learning potential of intraprofessional activities. Many of the included studies relied on self-reported perceptions of change, therefore, future research should focus on generating more robust evidence including objectively examined outcome measures. This review offers a comprehensive overview of the factors that influence intraprofessional workplace learning in postgraduate medical education. Finally, we provide recommendations for enhancing intraprofessional learning in clinical practice.
| S-EPMC8424991 | biostudies-literature
Project description:BackgroundLearning communities (LCs) are intentionally designed groups that are actively engaged in learning with and from each other. While gaining prominence in US medical schools, LCs show significant variability in their characteristics across institutions, creating uncertainty about how best to measure their effects.ObjectiveThe aim of this study is to describe the characteristics of medical school LCs by primary purpose, structures, and processes and lay the groundwork for future outcome studies and benchmarking for best practices.MethodsMedical school LC directors from programs affiliated with the Learning Communities Institute (LCI) were sent an online survey of program demographics and activities, and asked to upload a program description or summary of the LC's purpose, goals, and how it functions. Descriptive statistics were computed for survey responses and a qualitative content analysis was performed on program descriptions by 3 authors to identify and categorize emergent themes.ResultsOf 28 medical school LCs surveyed, 96% (27) responded, and 25 (89%) provided program descriptions for qualitative content analysis. All programs reported longitudinal relationships between students and faculty. Most frequently cited objectives were advising or mentoring (100%), professional development (96%), courses (96%), social activities (85%), and wellness (82%). Primary purpose themes were supporting students' professional development, fostering a sense of community, and creating a sense of wholeness. Structures included a community framework, subdivisions into smaller units, and governance by faculty and students. Process themes included longitudinal relationships, integrating faculty roles, and connecting students across class years.ConclusionsMedical school LCs represent a collection of high-impact educational practices characterized by community and small-group structures, relational continuity, and collaborative learning as a means to guide and holistically support students in their learning and development as physicians. In describing 27 medical school LCs, this study proposes a unifying framework to facilitate future educational outcomes studies across institutions.
| S-EPMC6434432 | biostudies-literature
Project description:BackgroundDespite extensive research data indicating that cooperative learning promotes higher achievement, the creation of positive relationships, and greater psychological health for students at all levels in their education, cooperative learning as a teaching strategy is still underutilized in undergraduate medical education.MethodsA cooperative learning task was introduced as part of the mandatory first Year undergraduate Pathology course. The task was to create an 8.5" x 11" poster summary of pre-assigned content in self-chosen groups of four or five students. On the designated "Poster Day," the posters were displayed and evaluated by the students using a group product evaluation. Students also completed an individual group process reflection survey. An objective evaluation of their understanding was gauged at the midterm examination by specific content-related questions.ResultsMajority (91-96%) of students judged the group products to be relevant, effective, easy-to-understand, and clearly communicated. The majority of the students (90-100%) agreed that their group process skills of time management, task collaboration, decision-making and task execution were effective in completing this exercise. This activity created a dynamic learning environment as was reflected in the students' positive, professional discussion, and evaluation of their posters. The content-related questions on the midterm examination were answered correctly by 70-92% of the students. This was a mutually enriching experience for the instructor and students.ConclusionThese findings demonstrate that cooperative learning as a teaching strategy can be effectively incorporated to address both content and interpersonal skill development in the early years of undergraduate medical education.
| S-EPMC2217551 | biostudies-literature
Project description:A major bottleneck in developing clinically impactful machine learning models is a lack of labeled training data for model supervision. Thus, medical researchers increasingly turn to weaker, noisier sources of supervision, such as leveraging extractions from unstructured text reports to supervise image classification. A key challenge in weak supervision is combining sources of information that may differ in quality and have correlated errors. Recently, a statistical theory of weak supervision called data programming has shown promise in addressing this challenge. Data programming now underpins many deployed machine-learning systems in the technology industry, even for critical applications. We propose a new technique for applying data programming to the problem of cross-modal weak supervision in medicine, wherein weak labels derived from an auxiliary modality (e.g., text) are used to train models over a different target modality (e.g., images). We evaluate our approach on diverse clinical tasks via direct comparison to institution-scale, hand-labeled datasets. We find that our supervision technique increases model performance by up to 6 points area under the receiver operating characteristic curve (ROC-AUC) over baseline methods by improving both coverage and quality of the weak labels. Our approach yields models that on average perform within 1.75 points ROC-AUC of those supervised with physician-years of hand labeling and outperform those supervised with physician-months of hand labeling by 10.25 points ROC-AUC, while using only person-days of developer time and clinician work-a time saving of 96%. Our results suggest that modern weak supervision techniques such as data programming may enable more rapid development and deployment of clinically useful machine-learning models.
| S-EPMC7413132 | biostudies-literature
Project description:Gene expression profiles were generated from 199 primary breast cancer patients. Samples 1-176 were used in another study, GEO Series GSE22820, and form the training data set in this study. Sample numbers 200-222 form a validation set. This data is used to model a machine learning classifier for Estrogen Receptor Status. RNA was isolated from 199 primary breast cancer patients. A machine learning classifier was built to predict ER status using only three gene features.
2013-01-01 | E-GEOD-29210 | biostudies-arrayexpress
Project description:BackgroundBlended learning, which combines face-to-face learning and e-learning, has grown rapidly to be commonly used in education. Nevertheless, the effectiveness of this learning approach has not been completely quantitatively synthesized and evaluated using knowledge outcomes in health education.ObjectiveThe aim of this study was to assess the effectiveness of blended learning compared to that of traditional learning in health education.MethodsWe performed a systematic review of blended learning in health education in MEDLINE from January 1990 to July 2019. We independently selected studies, extracted data, assessed risk of bias, and compared overall blended learning versus traditional learning, offline blended learning versus traditional learning, online blended learning versus traditional learning, digital blended learning versus traditional learning, computer-aided instruction blended learning versus traditional learning, and virtual patient blended learning versus traditional learning. All pooled analyses were based on random-effect models, and the I2 statistic was used to quantify heterogeneity across studies.ResultsA total of 56 studies (N=9943 participants) assessing several types of learning support in blended learning met our inclusion criteria; 3 studies investigated offline support, 7 studies investigated digital support, 34 studies investigated online support, 8 studies investigated computer-assisted instruction support, and 5 studies used virtual patient support for blended learning. The pooled analysis comparing all blended learning to traditional learning showed significantly better knowledge outcomes for blended learning (standardized mean difference 1.07, 95% CI 0.85 to 1.28, I2=94.3%). Similar results were observed for online (standardized mean difference 0.73, 95% CI 0.60 to 0.86, I2=94.9%), computer-assisted instruction (standardized mean difference 1.13, 95% CI 0.47 to 1.79, I2=78.0%), and virtual patient (standardized mean difference 0.62, 95% CI 0.18 to 1.06, I2=78.4%) learning support, but results for offline learning support (standardized mean difference 0.08, 95% CI -0.63 to 0.79, I2=87.9%) and digital learning support (standardized mean difference 0.04, 95% CI -0.45 to 0.52, I2=93.4%) were not significant.ConclusionsFrom this review, blended learning demonstrated consistently better effects on knowledge outcomes when compared with traditional learning in health education. Further studies are needed to confirm these results and to explore the utility of different design variants of blended learning.
| S-EPMC7445617 | biostudies-literature
Project description:PurposeOne of the aims of medical education is to generate lifelong learners, leading to the identification of self-directed learning (SDL) as an essential component of medical education. While SDL is focused on an individual learner, collaboration is critically important in medicine. We developed an online course using the collaborative SDL (CSDL) framework. A goal for the course was for students to gain a better appreciation for the SDL process by exploring the COVID-19 pandemic.MethodsWe utilized CSDL to implement a 2-week elective attended by fifteen M3 and M4 medical students. Students submitted short videos reflecting on their course experience and the relevance of the material to their future training. Qualitative analysis of reflections was conducted to determine the effectiveness of the CSDL framework, and an assessment of the course evaluations was performed to explore student perceptions of the course and its effectiveness at preparing them for practice. A survey regarding student perceptions of SDL was offered to M3 and M4 students in order to explore their experiences with SDL and perceptions of its importance in the context of the school curriculum.ResultsThe CSDL framework was effective in making students aware of the importance of SDL in medical practice. Students gained basic and clinical knowledge about the subject, experienced increased confidence, and appreciated collaborating with their peers. The survey offered to the general student body reflected that all students perceived that they employed SDL in their time as medical students. However, many students indicated that they had not utilized components of SDL beyond synthesizing and assessing their learning needs.ConclusionCSDL is an effective method for promoting self-directed learning. Undergraduate medical course constructs utilizing CSDL will benefit students as they continue their career development.Supplementary informationThe online version contains supplementary material available at 10.1007/s40670-021-01493-7.
| S-EPMC8725963 | biostudies-literature