Medical Interpretation Research Articles

Time-dependent characteristics of analytical measurands.

Biological variation is a relevant component of diagnostic uncertainty. In addition to within-subject andbetween-subject variation, preanalytical variation also includes components that contribute to biological variability. Among these, daily recurring, i.e.,diurnal physiological variation is of particular importance, as it contains both a random and a non-random component ifthe exact time of blood collection is not known. We introduce four time-dependent characteristics (TDC) of diurnal variations for measurands to assess the relevance and extent of time dependence on the evaluation of laboratory results. TDC address (i)a threshold for considering diurnality, (ii) the expected relative changes per time unit, (iii) the permissible time interval between two blood collections at different daytimes within which the expected time dependence does not exceed a defined analytical uncertainty, and (iv) a rhythm-expanded reference change value. TDC and their importance will be exemplified by the measurands aspartate aminotransferase, creatine kinase, glucose, thyroid stimulating hormone, and total bilirubin. TDCs are calculated for four time slots that reflect known blood collection schedules, i.e.,07:00-09:00, 08:00-12:00, 06:00-18:00, and 00:00-24:00. The amplitude and the temporal location of the acrophase are major determinates impacting the diagnostic uncertainty and thus the medical interpretation, especially within the typical blood collection time from 07:00 to 09:00. We propose to check measurands for the existence of diurnal variations and, if applicable, to specify their time-dependent characteristics as outlined in our concept.

Challenges in Assessment and Management of Postoperative Agitation and Delirium in a Stroke Patient with Limited English Proficiency: Case Report.

While Title VI of the Civil Rights Act of 1964 mandates use of interpreters for patients with limited English proficiency, significant disparities persist in intensive postsurgical care. We present the case of a 60-year-old Vietnamese-speaking man with a Type A aortic dissection requiring postoperative mechanical ventilation and stroke care. Despite use of a remote video interpreter, our language-discordant nursing and physician providers faced challenges in managing agitation and delirium and assessing neurological function. This case highlights the need for adequate interpretation equipment, linguistic diversity among providers, and interventions to promote and enable consistent certified and professional medical interpreter use.

PROGRAM PROCEDURES FOR TRAINING A RECOGNITION SYSTEM FOR THE DIFFERENTIAL DIAGNOSIS OF PATIENTS BASED ON HETEROGENEOUS SYMPTOM COMPLEXES

The machine learning recognition system for the differential diagnosis of patients based on heterogeneous nephrology parameter complexes is being considered, transitioning from instrumental means of examination. Training utilizes empirical statistics of clinical cases in a database with reliable diagnoses. The purpose is to expand the capabilities of information extraction from similar databases for training recognition procedures by enriching this toolkit with new features containing characteristic aspects of the extracted information. The research object is the mathematical and software toolkit for training recognition procedures of patient differential diagnosis based on statistics of reliably diagnosed clinical cases. The subject of the study is the software procedures for forming models of parameter complex incidence during training along scales of their values and the procedures for using these models in diagnostics. Model acquisition is perceived as the main content of the training process in ensuring diagnosis differentiation. A criterion for accepting preferential diagnostic decisions using such models is proposed. To simplify the development of mathematical and software procedures, heterogeneous symptom complexes are normalized and transformed to the [0; 1] scale. The introduction states the significant prevalence in medicine and related fields of databases with medical and biomedical data statistics on parameters and characteristics of human organs and systems in different conditions, their medical interpretation, and their use for various purposes, often associated with patient diagnostics. The problems of their formation and use are outlined on real databases, with one complicating factor in the development of diagnostic hardware-software being the substantial heterogeneity of parameters determined by patient examination instruments.

Merlin: A Vision Language Foundation Model for 3D Computed Tomography.

Over 85 million computed tomography (CT) scans are performed annually in the US, of which approximately one quarter focus on the abdomen. Given the current shortage of both general and specialized radiologists, there is a large impetus to use artificial intelligence to alleviate the burden of interpreting these complex imaging studies while simultaneously using the images to extract novel physiological insights. Prior state-of-the-art approaches for automated medical image interpretation leverage vision language models (VLMs) that utilize both the image and the corresponding textual radiology reports. However, current medical VLMs are generally limited to 2D images and short reports. To overcome these shortcomings for abdominal CT interpretation, we introduce Merlin - a 3D VLM that leverages both structured electronic health records (EHR) and unstructured radiology reports for pretraining without requiring additional manual annotations. We train Merlin using a high-quality clinical dataset of paired CT scans (6+ million images from 15,331 CTs), EHR diagnosis codes (1.8+ million codes), and radiology reports (6+ million tokens) for training. We comprehensively evaluate Merlin on 6 task types and 752 individual tasks. The non-adapted (off-the-shelf) tasks include zero-shot findings classification (31 findings), phenotype classification (692 phenotypes), and zero-shot cross-modal retrieval (image to findings and image to impressions), while model adapted tasks include 5-year chronic disease prediction (6 diseases), radiology report generation, and 3D semantic segmentation (20 organs). We perform internal validation on a test set of 5,137 CTs, and external validation on 7,000 clinical CTs and on two public CT datasets (VerSe, TotalSegmentator). Beyond these clinically-relevant evaluations, we assess the efficacy of various network architectures and training strategies to depict that Merlin has favorable performance to existing task-specific baselines. We derive data scaling laws to empirically assess training data needs for requisite downstream task performance. Furthermore, unlike conventional VLMs that require hundreds of GPUs for training, we perform all training on a single GPU. This computationally efficient design can help democratize foundation model training, especially for health systems with compute constraints. We plan to release our trained models, code, and dataset, pending manual removal of all protected health information.

The Development and Evaluation of Alternative Communication Strategies to Facilitate Interactions with Refugees in Obstetric Triage: A Preliminary Study

Purpose: In this pilot study, we investigate the potential of pictographs to facilitate communication and accelerate care for non-English speaking birthing people in triage. The immigrant and refugee populations have been steadily increasing in the United States; reproductive age women are a large percentage of those who seek medical care. Barriers to care include timely access to interpreters, which can delay treatment. Pictographs have been used successfully in medicine in various situations such as pharmaceutical instruction and symptom description. We searched extensively for obstetric triage pictographs with little success. To address this, we created a pictograph sheet specifically for common obstetrical complaints. Methods: The most common reasons for visiting triage were ascertained. From these, key topics were identified for drawings. For each topic, the authors drafted scripts for a list of relevant actions. The actions for each topic were then represented by a professional illustrator in pictographs. Pictographs were developed using the following guidelines: focus on actions people should take, include prompts within pictures such as labels or arrows and exclude nonessential details such as room background and elaborate borders. After the pictographs were developed, its appropriateness, accuracy and relevance were reviewed and validated by two experts with experience in women’ care. The experts perceived that the drawings represented the actions well in a simple and clean manner. The following concerns were depicted: decreased fetal movement, abnormal vaginal discharge, rupture of membranes, vaginal bleeding, vision changes, shortness of breath, nausea, contractions, headaches, and overall assessment of wellbeing. Focus groups were developed collaboratively with Della Lamb, a resettlement agency. Inclusion criteria included women of reproductive age who are non-English speaking and currently pregnant. These focus groups were held at the University Health. Two focus groups were created: one with refugees from Afghanistan and one with refugees from the Congo. Each focus group was guided by a licensed medical interpreter of their native language: Dari and Swahili, respectively, and ranged from 3 to 4 participants. One by one the pictograms were introduced to the group. Each participant was asked what they think the illustration represents. For participants’ comfort, groups were not recorded; instead, notes were taken by observers during and after the discussion. The participants read the text and pictographs and were asked to identify if pictographs were clear, understandable, and acceptable. Results: Congolese Focus Group: Participants’ ages varied in their 30s. All had recently immigrated to the United States and lived in Missouri for less than 6 months. They were in the second trimester of their pregnancy and had at least one prenatal visit in the United States. They were all multiparous women with previous uncomplicated vaginal deliveries in their native country. All could read and write in Swahili but spoke no English. They all had no previous experience with the medical health system in the United States. Afghani Focus Group: Participants’ ages varied in their 20-30s. All had recently immigrated to the United States and lived in Missouri for less than 3 months. 3 women were in their second trimester, 1 was in her third trimester. Aside from 1, the rest (3) were multiparous women with previous uncomplicated vaginal deliveries in their native country. 3 of them could read and write in Dari but spoke no English. They all had no previous experience with the medical health system in the United States. Comprehension was significantly different between the two groups. Overall, the Congolese group of women misunderstood the purpose of the pictographs. No picture was understandable to the patients. It was also apparent that the patients did not realize these were issues for which to come to triage. There was also ambiguity regarding the color, i.e. they did not correlate green with nausea. The general prohibition sign was also absent in meaning for them. However, all of the patients expressed interest in learning what each picture meant and would like to have that option there if they came to triage and could not get an interpreter. The Dari group did not understand shortness of breath or nausea. However, they were familiar with the other symptoms and able to identify those as reasons to come to triage. The patients were also familiar with colors and scales. Conclusion: The pictograph-based approach might be an effective tool in developing health-care instructions for immigrant women with limited literacy skills. However, research is needed regarding limitations due to cultural backgrounds. Future research is also needed to compare the effect of pictograph-enhanced instructions with written text-based instructions on adherence to instructions and health outcomes.

A proposal to teach medical students how to use interpreters

IntroductionMedical interpreters are crucial to ensure fair and high-quality healthcare for patients with limited English proficiency (LEP). Despite the need to use high-quality medical interpreters to communicate with LEP patients, medical schools often do not adequately educate their students on how to work with interpreters.AimsThis study seeks to investigate the efficacy of using peer-assisted learning to teach medical students how to properly use medical interpreters. Moreover, the study strives to elucidate if an interactive peer-led model can be an effective teaching modality to train medical students about the basics of using medical interpreters.MethodsA pre- and post-training design was utilized to investigate the efficacy of peer-assisted learning in teaching medical students how to use interpreters. Second year medical students led a two-part workshop consisting of the following: (1) a didactic training session and (2) a practical session where learners interacted with Spanish-speaking standardized patients through an interpreter. Pre-training and post-training responses to survey questions were analyzed to determine changes in student comfort, confidence, and knowledge of best practices when using a medical interpreter.ResultsThere was a statistically significant increase in comfort and confidence with using interpreters after receiving peer-assisted training.ConclusionsA peer-led didactic training followed by an interactive training session can increase student comfort and confidence with using medical interpreters in clinical settings. Peer-assisted-learning may be an effective way to teach some of the best practices of using medical interpreters.

Comparing Hospital Length of Stay Between Persons With LEP and English-Speaking Patients in a Large Rural Academic Medical Center.

Patients with limited English proficiency that are hospitalized without regular access to professional medical interpreters have a longer length of stay (LOS).1 The authors studied the difference in LOS between English-speaking patients and patients with limited English proficiency in New Hampshire's only academic trauma medical center. The authors also examined race, ethnicity, and distance of residence from hospital. Retrospective data were queried from EPIC, the electronic medical record system used by the authors. Queried data focused on inpatient hospitalizations between January 1, 2019, and December 31, 2021. Patient primary language was grouped into English, Spanish, and other non-English language. Spanish-speaking patients live on average 39.6 miles farther from a hospital than English-speaking patients and have a 0.34 lower case mix index. After English, Spanish is the second-most frequently spoken language. Regression analyses found language to be a significant factor in LOS, LOS variance, and case mix index. A 2.34-day longer LOS for Spanish-speaking patients demonstrates an important health care disparity warranting further attention.

Deep Learning in Medical Image Interpretation: Recent Advances and Impact on Clinical Practice

Deep Learning in Medical Image Interpretation: Recent Advances and Impact on Clinical Practice

Deep Learning Approaches for Medical Image Processing in the Big Data Era

Diagnostics, therapy planning, and patient monitoring are areas where medical image processing has become indispensable in modern healthcare. The advent of deep learning techniques and the broad usage of big data in healthcare have brought a revolutionary shift in medical picture analysis and interpretation. An overview of deep learning techniques for medical image processing based on big data is given in this article. This article covers the pros and cons of adopting big data for medical imaging, from data storage and analysis to data capture. Beyond that, we take a look at medical image analysis using deep learning algorithms such as recurrent neural network (RNN), Convolutional Neural Network (CNN), and generative adversarial network (GAN), and we highlight its advantages and disadvantages. We also examine recent innovations such as transfer learning, multi-modal imaging fusion, and federated learning, which can improve the accuracy and efficiency of medical image processing systems. Finally, we discuss how medical image processing driven by deep learning could improve clinical decision-making, patient outcomes, and the development of personalized medicine in the era of data-driven healthcare.

Validation of article analysis by ChatGPT: Using a cohort of patients with EGFR mutation-positive non-small cell lung cancer treated with first-line osimertinib.

e20542 Background: The rapid advancement of large language models like ChatGPT has shown potential in various fields, including medicine. However, their applicability in clinical practice, especially in complex medical knowledge interpretation, remains unclear. This study developed a framework for article analysis, retrieval of additional information, and knowledge updating using ChatGPT and tested its clinical validity by comparing it to clinical data. Methods: To simplify the prompt to ChatGPT by selecting monotherapy, patients with EGFR mutation-positive non-small cell lung cancer (NSCLC) who had received 1st -line treatment with Osimertinib, were included in this study. ChatGPT analyzed published FLAURA trial data and developed a scoring system for predicting treatment efficacies for comparison with clinical data. Treatment efficacies were evaluated by using progression-free survival (PFS) and overall survival (OS). ChatGPT searched for papers needed to update the scoring system and extracted 5 papers. To reduce the bias of the search, the extraction of papers by ChatGPT was repeated 10 times and the top 5 papers were selected. The revised scoring system, incorporating insights from these 5 papers, was applied to data from Shizuoka Cancer Center (1st-line osimertinib initiated during 2018-2022.3). The clinical validity of analyses by ChatGPT was evaluated in terms of whether the scoring system reflects clinical practice. Analysis with ChatGPT was performed using GPT4 and natural language prompts. Results: Based on the FLAURA study, ChatGPT created a scoring system consisting of the following items: EGFR mutation type, presence of CNS metastasis, smoking history, sex, and performance status (PS). ChatGPT extracted two prospective studies and three retrospective studies. The scoring system, updated by ChatGPT included EGFR mutation type: exon 19 del /L858R (3/1point), sex: female/male (2/1), smoking history: never-smoker/smoker (2/1), CNS metastases: without/with (2/1), PS 0/1/2/3/4 (2/1/0/-1/-2), age: < 75/≥ 75 years (2/1) PD-L1 expression < 50%/≥ 50% (2/0); liver metastasis, without/with (2/0). Group A showing better efficacies to treatment was defined as patients with a cutoff value of 11 or higher. The scoring system was fitted to validate cohort (140 patients without missing scoring system items). Group A (n = 98, ≥11points) outperformed Group B (n = 42, < 11) in both PFS and OS (medianPFS 22.6 months vs. 9.7 months HR 0.32, 95% CI 0.20-0.50, p < 0.0001, medianOS 54.4 months vs. 21.7 months HR 0.28, 95%CI 0.16-0.49, p < 0.0001). Conclusions: The scoring system created by using ChatGPT reflected the efficacies in a cohort of EGFR-positive NSCLC patients treated with 1st-line osimertinib. The framework for article analysis, retrieval of additional information, and knowledge update using ChatGPT is potentially applicable to other cohorts.

DF-GAM: Cross-Domain Ultrasound Image High-Quality Reconstruction Using a Dual Frequency-Domain Guided Adaptation Model

ObjectiveTo enhance the quality of low-resolution (LR) ultrasound images and mitigate artifacts and speckle noise, which can impede accurate medical diagnosis, a novel method called the dual frequency-domain guided adaptation model (DF-GAM) is proposed. The method aims to achieve high-quality image reconstruction across diverse domains, including different ultrasound machines, diseases and phantom images. MethodsDF-GAM utilizes a dual-branch network architecture combined with frequency-domain self-adaptation and self-supervised edge regression. This approach enables cross-domain enhancement by focusing on the reconstruction of clear tissue structures and speckle patterns. The model is designed to adapt to various ultrasound imaging (USI) scenarios, ensuring its applicability in real-world clinical settings. ResultsExperimental evaluations of DF-GAM were conducted using five different datasets. The results demonstrated the method's effectiveness, with DF-GAM outperforming existing enhancement techniques. The average peak signal-to-noise ratio (PSNR) achieved was 34.62, and the structural similarity index (SSIM) was 0.91, indicating a significant improvement in image quality compared to other methods. ConclusionDF-GAM shows great potential in improving medical image diagnosis and interpretation. Its ability to enhance LR ultrasound images across various domains without the need for extensive training data makes it a valuable tool for clinical use. The high PSNR and SSIM scores validate the method's effectiveness, suggesting that DF-GAM could significantly contribute to the field of USI diagnostics.

Alzheimer’s Disease Diagnosis via SVM-Driven Classification of Structural MRI Data

Abstract: In this study we introduce an automated system designed to classify Alzheimers disease (AD) using MRI brain scans. Our technique employs support vector machine (SVM) classification to effectively differentiate between AD patients and older adults without the disease by analyzing brain MRI images. The process includes dividing 2D T1 weighted MR images into regions of interest and extracting gray matter characteristics, from each region. By leveraging SVMs reliability and proven success in classification tasks along with the principle of minimizing risks our research showcases performance in generalizing results. This SVM based approach shows promise in distinguishing AD affected brains from ones presenting a valuable opportunity, for enhancing medical analysis and interpretation of MRI findings.

Survey of perioperative utilization of professional medical interpreters for limited-English proficient patients: Towards a framework for systems-level improvement

IntroductionSurgical patients in United States’ hospitals have grown more linguistically diverse. However, professional medical interpreter (PMI) utilization remains inconsistent at many healthcare institutions for limited-English proficient (LEP) patients. Numerous studies demonstrate that inadequate perioperative access to PMI leads to worse care and outcomes for LEP patients. The perioperative setting presents unique challenges for providers in caring for LEP patients including obtaining informed consent, a transient team of multi-disciplinary providers, and the time pressure of operating room practice. ObjectiveOur study is the first to assess barriers to consistent PMI use in the immediate perioperative setting and to identify system-level approaches to improve PMI use. MethodWe surveyed a multidisciplinary team of perioperative nurses, advanced practice providers, anesthesiology and surgical trainees and faculty on their personal practices and perceived barriers to PMI use. An anonymous online 13-question survey was used. Answer options included rank order, multiple choice, Likert scale, and free text. Survey data was analyzed using univariate statistics and stratified based on providers having received information on the Title VI Civil Rights Act of 1964 and their medical training status. Pearson's chi-squared test was performed and odds ratios calculated to determine if these provider characteristics were associated with increased preference for PMI over bilingual staff and patients’ family members for interpretation and with other LEP evidence-based care practices. ResultsWe received a total of 262 responses with a 28.3 % response rate. Among survey participants, 19.1 % of participants did not know where to find patients’ language preferences in the electronic health record (EHR) and 69.8 % of participants did not know how to update language preferences in the EHR. When a bilingual staff was present, 31.0 % and 24.8 % of providers would forgo using video and phone PMI, respectively. Comparatively, trainees were more likely to use PMI instead of patients’ family or bilingual staff. Providers who received training on the Title VI Civil Rights Act of 1964 were more likely to use PMI for interpretation and provide translated consent forms, and least likely to utilize patients’ families for interpretation. Barriers to PMI use included difficulty obtaining video consoles or phones, long wait times for interpreters, unavailability of language, unfamiliarity with using video consoles or accessing interpreters, poor sound quality and internet connectivity issues. As many as 26 % of respondents cited time pressure for starting a surgical case as the reason for not using an interpreter. DiscussionBased on these survey results, we developed a framework for quality improvement initiatives to effect system-level change in similar high-output perioperative settings. This framework can be categorized into (1) process refinement and harnessing the EHR for identification and documentation of LEP status, (2) interdepartmental collaboration to address workflow constraints, technical issues and supply-demand mismatch, and (3) increasing awareness among leadership and providers about interpretation laws, cost reduction and improved outcomes with PMI use.

Automated translation accurately translates recorded pediatric neurosurgery clinic conversations between Spanish and English.

The purpose of this study is to analyze an automated voice to text translation device by reporting the translation accuracy for recorded pediatric neurosurgery clinic conversations, classifying errors in translation according to their impact on overall understanding, and comparing the incidence of these errors in English to Spanish vs. Spanish to English conversations. English and Spanish speaking patients at a single academic health system's outpatient pediatric neurosurgery clinic had their conversations recorded. These recordings were played back to a Google Pixel handheld smartphone with Live Translate voice to text translation software. A certified medical interpreter evaluated recordings for incidence of minor errors, errors impacting understanding, and catastrophic errors affecting patient-provider relationship or care. Two proportion t-testing was used to compare these outcomes. 50 patient visits were recorded: 40 English recordings translated to Spanish and 10 Spanish recordings translated to English. The mean transcript length was 4244 ± 992 words. The overall accuracy was 98.2% ± 0.5%. On average, 46 words were missed in translation (1.09% error rate), 31 understanding-altering translation errors (0.73% error rate), and 0 catastrophic errors were made. There was no significant difference in English to Spanish or vice versa. Voice to text translation devices using automatic speech recognition accurately translate recorded clinic conversations between Spanish and English with high accuracy and low incidence of errors impacting medical care or understanding. Further study should investigate additional languages, assess patient preferences and potential concerns with respect to device use, and compare these devices directly to medical interpreters in live clinic settings.

Advancements in Deep Learning for Minimally Invasive Surgery: A Journey through Surgical System Evolution

The surge in artificial intelligence (AI) applications across diverse fields owes much to advancements in deep learning and computational processing speed. In medicine, AI's reach extends to medical image analysis and genomic data interpretation. More recently, AI's role in analyzing minimally invasive surgery (MIS) videos has gained traction, with a growing body of research focusing on organ and anatomy identification, instrument recognition, procedure recognition, surgical phase delineation, surgery duration prediction, optimal incision line identification, and surgical education. Concurrently, the development of autonomous surgical robots, exemplified by the Smart Tissue Autonomous Robot (STAR) and RAVEN systems, has shown promising strides. Notably, STAR is currently employed in laparoscopic imaging to discern the surgical site from laparoscopic images and is undergoing trials for an automated suturing system, albeit in animal models. This review contemplates the prospect of fully autonomous surgical robots in the future.

Advancements in Deep Learning for Minimally Invasive Surgery: A Journey through Surgical System Evolution

The surge in artificial intelligence (AI) applications across diverse fields owes much to advancements in deep learning and computational processing speed. In medicine, AI's reach extends to medical image analysis and genomic data interpretation. More recently, AI's role in analyzing minimally invasive surgery (MIS) videos has gained traction, with a growing body of research focusing on organ and anatomy identification, instrument recognition, procedure recognition, surgical phase delineation, surgery duration prediction, optimal incision line identification, and surgical education. Concurrently, the development of autonomous surgical robots, exemplified by the Smart Tissue Autonomous Robot (STAR) and RAVEN systems, has shown promising strides. Notably, STAR is currently employed in laparoscopic imaging to discern the surgical site from laparoscopic images and is undergoing trials for an automated suturing system, albeit in animal models. This review contemplates the prospect of fully autonomous surgical robots in the future.

Relationship between Illness Perception and the need for Professional Medical Interpretation for International Patients in Japan.

This study aimed to conduct a cross-sectional questionnaire survey of foreign patients to analyze the impact of patients' illness perceptions on their need for professional medical interpreters. From February 2022 to May 2023, an online questionnaire was distributed to 4,962 individuals aged 18 years or older who were non-native speakers of Japanese. These individuals were enrolled in organizations such as international exchange associations and Japanese language support classes and had utilized medical institutions in Japan due to their own illness or injury or that of their children. Among the 312 valid responses, international patients with a high score for illness perception were more likely to want to utilize professional medical interpreters than those with a low score for illness perception (odds ratio, 1.968; 95% confidence interval, 1.044-3.709; P = 0.036). Our findings suggest that hospitals should be better prepared to meet the potential language needs of international patients with a higher illness perception.

Teaching foundational language equity concepts in the pre-clinical curriculum

BackgroundDespite the prevalence of non-English languages in the US population, existing medical training to teach communication with linguistically diverse communities is limited to electives or solely focuses on medical interpreting. Language-appropriate communication skills are seldom comprehensively integrated in medical education. This study describes the development and evaluation of an intervention to teach foundational language equity concepts.MethodsThe authors implemented a pre-clinical language equity course at three medical school campuses between August 2020 and March 2022. Sessions focused on the impact of language in health, physician language proficiency standards, and working with medical interpreters. The study sought to (1) understand students’ language skills and prior clinical experiences with patients with non-English language preference and (2) evaluate the curriculum’s impact. Students self-reported their language skills and experiences as part of a voluntary pre-questionnaire. Pre and post-questionnaires evaluated knowledge, attitudes, and intent to apply language equity concepts. Descriptive statistics and chi-squared tests were used to examine trends; themes were identified from free-text responses.ResultsOverall, 301 students completed the course, 252 (83%) completed at least one questionnaire; for each session, between 35% and 46% of learners completed both pre and post-questionnaires. Three quarters (189/252) reported non-English languages. Over half (138/252) reported previous non-English language patient care, and 28% (62/224) had served as ad hoc (untrained) interpreters. Only two students (< 1%) had ever been assessed for medical language abilities. Students demonstrated improved post-course language equity knowledge, strategies for interpreter-mediated encounters, and likelihood to report a plan for language skills assessment (all p < .001). Most plans were multifaceted (61%, 38/62), involving goals like completing a language course, taking a proficiency exam, openly discussing skills and uncertainties with team members, and increasing professional interpreter utilization.ConclusionsA longitudinal language equity curriculum can be feasibly integrated in pre-clinical education, highlight the linguistic diversity of the student body, and serve as a first step in ensuring that all students have a strong language equity foundation prior to clinical rotations. Future steps include evaluating the intervention’s potential long-term effects on professional interpreter utilization, student clinical performance, and institutional culture that promotes multilingualism.

Medical Interpreting Services for Refugees in Canada: Current State of Practice and Considerations in Promoting this Essential Human Right for All.

Language barriers, specifically among refugees, pose significant challenges to delivering quality healthcare in Canada. While the COVID-19 pandemic accelerated the emergence and development of innovative alternatives such as telephone-based and video-conferencing medical interpreting services and AI tools, access remains uneven across Canada. This comprehensive analysis highlights the absence of a cohesive national strategy, reflected in diverse funding models employed across provinces and territories, with gaps and disparities in access to medical interpreting services. Advocating for medical interpreting, both as a moral imperative and a prudent investment, this article draws from human rights principles and ethical considerations, justified in national and international guidelines, charters, codes and regulations. Substantiated by a cost-benefit analysis, it emphasizes that medical interpreting enhances healthcare quality and preserves patient autonomy. Additionally, this article illuminates decision-making processes for utilizing interpreting services; recognizing the pivotal roles of clinicians, interpreters, patients and caregivers within the care circle; appreciating intersectional considerations such as gender, culture and age, underscoring the importance of a collaborative approach. Finally, it provides recommendations at provider, organizational and system levels to ensure equitable access to this right and to promote the health and well-being of refugees and other individuals facing language barriers within Canada's healthcare system.

COMPARATIVE ANALYSIS OF GENERATIVE ARTIFICIAL INTELLIGENCE (AI) VERSUS MANUAL DATA ANALYSIS IN ORTHOPAEDIC PATIENT DATA

As patient data continues to grow, the importance of efficient and precise analysis cannot be overstated. The employment of Generative Artificial Intelligence (AI), specifically Chat GPT-4, in the realm of medical data interpretation has been on the rise. However, its effectiveness in comparison to manual data analysis has been insufficiently investigated.This quality improvement project aimed to evaluate the accuracy and time-efficiency of Generative AI (GPT-4) against manual data interpretation within extensive datasets pertaining to patients with orthopaedic injuries.A dataset, containing details of 6,562 orthopaedic trauma patients admitted to a district general hospital over a span of two years, was reviewed. Two researchers operated independently: one utilised GPT-4 for insights via prompts, while the other manually examined the identical dataset employing Microsoft Excel and IBM® SPSS® software. Both were blinded on each other's procedures and outcomes. Each researcher answered 20 questions based on the dataset including injury details, age groups, injury specifics, activity trends and the duration taken to assess the data.Upon comparison, both GPT-4 and the manual researcher achieved consistent results for 19 out of the 20 questions (95% accuracy). After a subsequent review and refined prompts (prompt engineering) to GPT-4, the answer to the final question aligned with the manual researcher's findings. GPT-4 required just 30 minutes, a stark contrast to the manual researcher's 9-hour analytical duration.This quality improvement project emphasises the transformative potential of Generative AI in the domain of medical data analysis. GPT-4 not only paralleled the accuracy of manual analysis but also achieved this in significantly less time. For optimal accurate results, data analysis by AI can be enhanced through human oversight. Adopting AI-driven approaches, particularly in orthopaedic data interpretation, can enhance efficiency and ultimately improve patient care. We recommend future investigations on large and more varied datasets to reaffirm these outcomes.