Assessing the Capability of Large Language Models in Naturopathy Consultation.

Background The rapid advancements in natural language processing have brought about the widespread use of large language models (LLMs) across various medical domains. However, their effectiveness in specialized fields, such as naturopathy, remains relatively unexplored. Objective The study aimed to assess the capability of freely available LLM chatbots in providing naturopathy consultations for various types of diseases and disorders. Methods Five free LLMs (viz., Gemini, Copilot, ChatGPT, Claude, and Perplexity) were used to converse with 20 clinical cases (simulation of real-world scenarios). Each case had the case details and questions pertinent to naturopathy. The responses were presented to three naturopathy doctors with > 5 years of practice. The answers were rated by them on a five-point Likert-like scale for language fluency, coherence, accuracy, and relevancy. The average of these four attributes is termed perfection in his study. Results The overall score of the LLMs were Gemini 3.81±0.23, Copilot 4.34±0.28, ChatGPT 4.43±0.2, Claude 3.8±0.26, and Perplexity 3.91±0.28 (ANOVA F [3.034, 57.64] = 33.47, P <0.0001. Together, they showed overall ~80% perfection in consultation. The average measure intraclass correlation coefficient among the LLMs for the overall score was 0.463 (95% CI = -0.028 to 0.76), P = 0.03. Conclusion Although the LLM chatbots could help in providing naturopathy and yoga treatment consultation with approximately an overall fair level of perfection, their solution to the user varies across different chatbots and there was very low reliability among them.

A Brief Review on Good Clinical Practice and its Training Methods.

The field of clinical research continues to advance, and for ethical conduct of research, researchers need to have a strong foundation in good clinical practice (GCP). GCP guidelines are internationally recognized standards that govern the conduct of clinical trials, ensuring the protection of human subjects and the generation of reliable data. However, novice researchers or postgraduate medical students who would delve into research activities may face difficulty getting avenues for GCP training. Hence, in this brief review article, we discuss the significance of GCP in the field of clinical research with a glimpse of its history, development, and key principles. We provide a list of online courses, benefits, and disadvantages of those courses, and a list of organizations that conduct GCP workshops or continued medical education programs in India.

A Cross-Sectional Study on Motor and Sensory Nerve Functions in Women Newly Diagnosed and Untreated for Hypothyroidism in a Tribal Area of Odisha, India.

Introduction: Altered levels of thyroid hormones can impact various body systems, including the nervous system. Hypothyroidism may disrupt nerve conduction due to pathophysiological changes associated with hormone deficiency. The tribal population, characterized by distinct lifestyles and dietary habits, may experience unique influences on their growth and development. Aim: This study aimed to compare nerve conduction in newly diagnosed and untreated tribal women affected by hypothyroidism with euthyroid tribal women. Methods:A cross-sectional study was conducted in southern Odisha, India, spanning from April 2020 to January 2021. Forty-five newly diagnosed hypothyroid tribal women were enlisted from the outpatient department of general medicine as the case group. Additionally, 45 age-matched apparently healthy euthyroid tribal women were included as the control group. The subjects' height and weight were measured by an expert clinician. Nerve conduction (motor and sensory) study on both extremities (left and right side) were conducted for all participants in the human physiology laboratory. Results:The mean age of participants was 48.13±12.12 years in the case group and 47.18±12.2 years in the control group. In hypothyroid tribal women, a significant decrease in conduction velocity was observed in the majority of motor nerves (right median [p = .03], left median [p = .02], left ulnar [p = .04], right posterior tibial [p = .001], left posterior tibial [p = .0001]) and sensory nerves (right median [p =.005], right ulnar [p = .02], right sural [p = .001], and left sural [p = .02]). Conclusion:In newly diagnosed cases of hypothyroidism in tribal women, there is a risk of neuropathy that impacts both motor and sensory neurons. Therefore, it is crucial to initiate early diagnosis and immediate treatment to prevent additional neurological damage.

Predictors of Recovery in Facial Nerve Palsy: Insights From an Observational Study in Odisha, India.

Background Facial nerve paralysis, leading to the loss of facial expression, poses significant discomfort to patients. While most individuals exhibit a favorable response to treatment, a subset experiences enduring facial deformities without clearly defined etiology. This study aimed to identify prognostic factors influencing outcomes and quality of life in facial nerve palsy patients, contributing to enhanced clinical management. Methods A prospective observational study was conducted in the Otorhinolaryngology Department of Maharaja Krishna Chandra Gajapati Medical College and Hospital, a tertiary care hospital. We included patients presenting with any clinical variety of facial nerve palsy, irrespective of age and gender. Only moribund and noncompliant cases were excluded. Patients underwent clinical assessment using the House-Brackmann (HB) grading at presentation and were subsequently monitored at three weeks, three months, and six months post-onset to assess recovery. Results Out of 66 patients, 18 (27.27%) fully recovered at three weeks, 50 (75.76%) recovered at three months, and 54 (81.82%) at six-month follow-up. Incomplete recovery was observed in 13 (19.69%) patients. Factors associated with favorable outcomes included younger age of onset (p = 0.003), lower baseline HB grade (IV or less) (p = 0.001), Electroneurography Degeneration Index (ENoG DI) of <70% (p < 0.0001), early initiation of treatment (within five days of onset) (p = 0.0003), and absence of comorbid conditions (p = 0.03). Gender and affected side (left or right) did not influence the outcome. Conclusion In summary, age, associated comorbid conditions, baseline HB grade, and extent of facial nerve degeneration are crucial predictors of outcomes in facial nerve palsy. This knowledge can guide clinicians in optimizing treatment strategies for improved patient care.

Social Media and Urticaria - A Data Audit of Facebook®, LinkedIn®, and Twitter® Posts.

Introduction: Urticaria is a common debilitating dermatological disorder impairing a patient's quality of life. Such patients are increasingly using socialmedia to manage their health and interact with peers, particularly during the ongoing COVID-19 pandemic. Objectives: To explore and analyse the quality of urticaria related social-media information available to patients. Materials and Methods: An in-depth data audit of the three most commonly used social networks viz. Facebook®, LinkedIn®, and Twitter® were done on a single day, as posts may change or lose relevance over time. The word "urticaria" was searched on three social media, and the first 100 posts in each were further analysed. The post-creator was either categorised as "individual" or "group", and non-English posts were excluded. All types of posts have been analysed, including text, images, video, and website links. We also collected the comments/replies, share/re-tweet, and likes on the posts. Results: Among the total 300 social-media posts, the highest number of "individual" posts was on LinkedIn® followed by Twitter® and Facebook® (χ2 = 82.86, P < 0.0001). Regarding thematic content, most Facebook® posts discussed disease symptoms, followed by the promotion of journal or blog posts, and discussion about causative and triggering agents. LinkedIn® was primarily used for the promotion of journal articles or blog posts, followed by educational webinars and urticaria treatment stories. Twitter® users mostly interacted with peers about their urticaria symptoms and perceived etiologic and triggering factors. Regarding the type of post, images were maximally shared on Facebook®, while video/video links and web links were highest on LinkedIn® (χ2 = 21.59, P < 0.0001). Conclusion: The overall quality of urticaria related information on these 3 social media platforms is satisfactory for patients. Dermatologists may consider utilising social media to further educate such patients and improve the overall treatment outcome. The use of such networking channels will continue to grow, as communication remains crucial for urticaria management.

Efficacy of ChatGPT in solving attitude, ethics, and communication case scenario used for competency-based medical education in India: A case study.

BACKGROUND: Competency-based medical education (CBME) is a method of medical training that focuses on developing learners' competencies rather than simply assessing their knowledge and skills. Attitude, ethics, and communication (AETCOM) are important components of CBME, and the use of artificial intelligence (AI) tools such as ChatGPT for CBME has not been studied. Hence, we aimed to assess the capability of ChatGPT in solving AETCOM case scenarios used for CBME in India. MATERIALS AND METHODS: A total of 11 case scenarios were developed based on the AETCOM competencies. The scenarios were presented to ChatGPT, and the responses generated by ChatGPT were evaluated by three independent experts by awarding score ranging from 0 to 5. The scores were compared with a predefined score of 2.5 (50% accuracy) and 4 (80% accuracy) of a one-sample median test. Scores among the three raters were compared by the Kruskal-Wallis H test. The inter-rater reliability of the evaluations was assessed using the intraclass correlation coefficient (ICC). RESULTS: The mean score of solution provided by ChatGPT was 3.88 ± 0.47 (out of 5), indicating an accuracy of approximately 78%. The responses evaluated by three raters were similar (Kruskal-Wallis H P value 0.51), and the ICC value was 0.796, which indicates a relatively high level of agreement among the raters. CONCLUSION: ChatGPT shows moderate capability in solving AETCOM case scenarios used for CBME in India. The inter-rater reliability of the evaluations suggests that ChatGPT's responses were consistent and reliable. Further studies are needed to explore the potential of ChatGPT and other AI tools in CBME and to determine the optimal use of these tools in medical education.

Radiological Differential Diagnoses Based on Cardiovascular and Thoracic Imaging Patterns: Perspectives of Four Large Language Models.

Background Differential diagnosis in radiology is a critical aspect of clinical decision-making. Radiologists in the early stages may find difficulties in listing the differential diagnosis from image patterns. In this context, the emergence of large language models (LLMs) has introduced new opportunities as these models have the capacity to access and contextualize extensive information from text-based input. Objective The objective of this study was to explore the utility of four LLMs-ChatGPT3.5, Google Bard, Microsoft Bing, and Perplexity-in providing most important differential diagnoses of cardiovascular and thoracic imaging patterns. Methods We selected 15 unique cardiovascular ( n = 5) and thoracic ( n = 10) imaging patterns. We asked each model to generate top 5 most important differential diagnoses for every pattern. Concurrently, a panel of two cardiothoracic radiologists independently identified top 5 differentials for each case and came to consensus when discrepancies occurred. We checked the concordance and acceptance of LLM-generated differentials with the consensus differential diagnosis. Categorical variables were compared by binomial, chi-squared, or Fisher's exact test. Results A total of 15 cases with five differentials generated a total of 75 items to analyze. The highest level of concordance was observed for diagnoses provided by Perplexity (66.67%), followed by ChatGPT (65.33%) and Bing (62.67%). The lowest score was for Bard with 45.33% of concordance with expert consensus. The acceptance rate was highest for Perplexity (90.67%), followed by Bing (89.33%) and ChatGPT (85.33%). The lowest acceptance rate was for Bard (69.33%). Conclusion Four LLMs-ChatGPT3.5, Google Bard, Microsoft Bing, and Perplexity-generated differential diagnoses had high level of acceptance but relatively lower concordance. There were significant differences in acceptance and concordance among the LLMs. Hence, it is important to carefully select the suitable model for usage in patient care or in medical education.

Assessing the Capability of ChatGPT, Google Bard, and Microsoft Bing in Solving Radiology Case Vignettes.

Background The field of radiology relies on accurate interpretation of medical images for effective diagnosis and patient care. Recent advancements in artificial intelligence (AI) and natural language processing have sparked interest in exploring the potential of AI models in assisting radiologists. However, limited research has been conducted to assess the performance of AI models in radiology case interpretation, particularly in comparison to human experts. Objective This study aimed to evaluate the performance of ChatGPT, Google Bard, and Bing in solving radiology case vignettes (Fellowship of the Royal College of Radiologists 2A [FRCR2A] examination style questions) by comparing their responses to those provided by two radiology residents. Methods A total of 120 multiple-choice questions based on radiology case vignettes were formulated according to the pattern of FRCR2A examination. The questions were presented to ChatGPT, Google Bard, and Bing. Two residents wrote the examination with the same questions in 3 hours. The responses generated by the AI models were collected and compared to the answer keys and explanation of the answers was rated by the two radiologists. A cutoff of 60% was set as the passing score. Results The two residents (63.33 and 57.5%) outperformed the three AI models: Bard (44.17%), Bing (53.33%), and ChatGPT (45%), but only one resident passed the examination. The response patterns among the five respondents were significantly different ( p = 0.0117). In addition, the agreement among the generative AI models was significant (intraclass correlation coefficient [ICC] = 0.628), but there was no agreement between the residents (Kappa = -0.376). The explanation of generative AI models in support of answer was 44.72% accurate. Conclusion Humans exhibited superior accuracy compared to the AI models, showcasing a stronger comprehension of the subject matter. All three AI models included in the study could not achieve the minimum percentage needed to pass an FRCR2A examination. However, generative AI models showed significant agreement in their answers where the residents exhibited low agreement, highlighting a lack of consistency in their responses.

Naturopathic Interventions for Reduction of Perceived Pain in Patients Suffering from Arthritis: A Systematic Review and Meta-Analysis.

Arthritis affects millions of lives with its pervasive effects on physical health and quality of life. Addressing the complexities of managing symptoms such as swelling, inflammation, and pain requires prolonged treatment. Naturopathy is a treatment method that enhances the body's innate ability to restore optimal health through a holistic approach including natural products and lifestyle modifications. This systematic review addresses the intersection of naturopathy and arthritis treatment to provide current evidence about its potential benefits. Four databases (PubMed, AYUSH Research Portal, Web of Science, and Google Scholar) were searched with the keywords "Naturopathy" AND "Arthritis". Randomized, non-randomized, and cross-over studies in English were included. Studies reporting perceived pain using a visual analogue scale (VAS) were selected for meta-analysis. A total of 15 studies were included in the systematic review. The studies were from Denmark, Egypt, France, Hungary, Israel, Italy, Spain, and Turkey, and the study periods ranged from 1992 to 2017. They suggested that naturopathic treatment modalities like exercise, mud compress, sand bath, or hydrotherapy may be used in addition to conventional modes of treatment for added benefit. There was a diversity of naturopathic treatment modalities and outcome evaluation methods. Most studies used mud compress or mud baths with reported improvement of symptoms. The meta-analysis of 10 studies (11 sets of data) showed a significant improvement in pain measured by VAS. The studies included in the review have a high level of heterogenicity. There is a need for more studies and uniform assessment methods with standardization of interventions for robust evidence. More clinical trials from countries where naturopathy is approved treatment modalities are needed.

Accuracy of a Non-Invasive Home Glucose Monitor for Measurement of Blood Glucose.

Introduction: Patients with diabetes mellitus monitor their blood glucose at home with monitors that require a drop of blood or use a continuous glucose monitoring device that implants a small needle in the body. However, both cause discomfort to the patients which may inhibit them for regular blood glucose checks. Photoplethysmogram (PPG) sensing technology is an approach for non-invasive blood glucose measurement and PPG sensors can be used to predict hypoglycaemic episodes. InChcek is a PPG-based non-invasive glucose monitor. However, its accuracy has not been checked yet. Hence, this study aimed to evaluate the accuracy of InCheck, a non-invasive glucose monitor for the estimation of blood glucose. Methods: In a tertiary care hospital, patients who came for blood glucose estimation were tested for blood glucose non-invasively on the InCheck device and then by the laboratory method (glucose oxidase-peroxidase). These two readings were compared. We used International Organization for Standardization (ISO) 15197:2013 (95% of values should be within ± 15 mg/dL of reference reading if reference glucose <100 mg/dL or within ± 15% of reference reading if reference glucose ≥100 mg/dL and 99% of the values should be within zones A and B in consensus error grid), and Surveillance Error Grid for analyzing the accuracy. Results: A total of 1223 samples were analyzed. There was a significant difference between the reference method glucose level (135 [Q1-Q3: 97 - 179] mg/dL) and monitor-measured glucose level (188.33 [Q1-Q3: 167.33-209.33] mg/dL) (P < 0.0001). A total of 18.5% of readings were following ISO 15197:2013 criteria and 67.25% of coordinates were within zone A and zone B of the consensus error grid. In the surveillance error grid analysis, about 29.4% of values were in the no-risk zone, 51.8% in slight risk, 18.6% in moderate risk, and 0.2% were in the severe risk zone. Conclusion: The accuracy of the InCheck device for the estimation of blood glucose by PPG signal is not following the recommended guidelines. Hence, further research is necessary for programming or redesigning the hardware and software for a better result from this optical sensor-based non-invasive home glucose monitor.

Relationship of emotional intelligence and capability of answering higher-order knowledge questions in physiology among first-year medical students.

Emotional intelligence (EI) has a positive correlation with the academic performance of medical students. However, why there is a positive correlation needs further exploration. We hypothesized that the capability of answering higher-order knowledge questions (HOQs) is higher in students with higher EI. Hence, we assessed the correlation between EI and the capability of medical students to answer HOQs in physiology. First-year undergraduate medical students (n = 124) from an Indian medical college were recruited as a convenient sample. EI was assessed by the Schutte Self-Report Emotional Intelligence Test (SSEIT), a 33-item self-administered validated questionnaire. A specially designed objective examination with 15 lower-order and 15 higher-order multiple-choice questions was conducted. The correlation between the examination score and the EI score was tested by Pearson's correlation coefficient. Data from 92 students (33 females and 59 males) with a mean age of 20.14 ± 1.87 yr were analyzed. Overall, students got a percentage of 53.37 ± 14.07 in the examination, with 24.46 ± 9.1 in HOQs and 28.91 ± 6.58 in lower-order knowledge questions (LOQs). They had a mean score of 109.58 ± 46.2 in SSEIT. The correlation coefficient of SSEIT score with total marks was r = 0.29 (P = 0.0037), with HOQs was r = 0.41 (P < 0.0001), and with LOQs was r = 0.14 (P = 0.19). Hence, there is a positive correlation between EI and the capability of medical students to answer HOQs in physiology. This study may be the foundation for further exploration of the capability of answering HOQs in other subjects.NEW & NOTEWORTHY This study assessed the correlation between emotional intelligence (EI) and the capability of medical students to answer higher-order knowledge questions (HOQs) in the specific context of physiology. The finding reveals one of the multifaceted dimensions of the relationship between EI and academic performance. This novel perspective opens the door to further investigations to explore the relationship in other subjects and other dimensions to understand why students with higher EI have higher academic performance.

ChatGPT in Answering Queries Related to Lifestyle-Related Diseases and Disorders.

Background Lifestyle-related diseases and disorders have become a significant global health burden. However, the majority of the population ignores or do not consult doctors for such disease or disorders. Artificial intelligence (AI)-based large language model (LLM) like ChatGPT (GPT3.5) is capable of generating customized queries of a user. Hence, it can act as a virtual telehealth agent. Its capability to answer lifestyle-related diseases or disorders has not been explored. Objective This study aimed to evaluate the effectiveness of ChatGPT, an LLM, in providing answers to queries related to lifestyle-related diseases or disorders. Methods A set of 20 lifestyle-related disease or disorder cases covering a wide range of topics such as obesity, diabetes, cardiovascular health, and mental health were prepared with four questions. The case and questions were presented to ChatGPT and asked for the answers to those questions. Two physicians rated the content on a three-point Likert-like scale ranging from accurate (2), partially accurate (1), and inaccurate (0). Further, the content was rated as adequate (2), inadequate (1), and misguiding (0) for testing the applicability of the guides for patients. The readability of the text was analyzed by the Flesch-Kincaid Ease Score (FKES).  Results Among 20 cases, the average score of accuracy was 1.83±0.37 and guidance was 1.9±0.21. Both the scores were higher than the hypothetical median of 1.5 (p=0.004 and p<0.0001, respectively). ChatGPT answered the questions with a natural tone in 11 cases and nine with a positive tone. The text was understandable for college graduates with a mean FKES of 27.8±5.74. Conclusion The analysis of content accuracy revealed that ChatGPT provided reasonably accurate information in the majority of the cases, successfully addressing queries related to lifestyle-related diseases or disorders. Hence, initial guidance can be obtained by patients when they get little time to consult a doctor or wait for an appointment to consult a doctor for suggestions about their condition.

ChatGPT in academic writing: Maximizing its benefits and minimizing the risks.

This review article explores the use of ChatGPT in academic writing and provides insights on how to utilize it judiciously. With the increasing popularity of AI-powered language models, ChatGPT has emerged as a potential tool for assisting writers in the research and writing process. We have provided a list of potential uses of ChatGPT by a novice researcher for getting help during research proposal preparation and manuscript writing. However, there are concerns regarding its reliability and potential risks associated with its use. The review highlights the importance of maintaining human judgment in the writing process and using ChatGPT as a complementary tool rather than a replacement for human effort. The article concludes with recommendations for researchers and writers to ensure responsible and effective use of ChatGPT in academic writing.

Comparing the Efficacy of Individual Approaches and Team-Based Approaches in Solving Clinical Case Vignettes.

Background Clinical case vignettes are a widely adopted pedagogical approach in medical education. The cases may be presented to students with a closed response option for objectivity. While solving clinical cases has demonstrated its effectiveness in enhancing medical students' clinical reasoning, there is an ongoing debate regarding the most effective approach: individual problem-solving or team-based problem-solving. Objective To observe and compare the score obtained from individual clinical problem-solving approaches versus team-based clinical problem-solving approaches. Methods After obtaining consent, a total of 100 students were randomly selected for the study. The participants were divided into two groups: an individual approach group (IAG) (n=25) and a team-based approach group (TAG) comprising 25 groups of three students each. Both groups were presented with a set of 10 clinical problems, each requiring a closed-answer response of "yes", "no", or "don't know". The participants' responses were recorded and analyzed to evaluate their problem-solving efficacy. Results A total of 25 responses were obtained from 25 students from the IAG group and 25 responses from 25 groups from the TAG group. There was no difference between the score in IAG (7.44±1.12) and TAG (7.52 1.66) p-value=0.58. There was no difference between individual scores in 10 questions between IAG and TAG groups. Conclusion The study found no significant score differences between individual and team-based clinical case-solving groups. Hence, for the objective type of case-solving pattern used in this study, a team-based approach may not be necessary. Further research is needed to explore factors for such findings in future studies.

Assessing the Utilization of Large Language Models in Medical Education: Insights From Undergraduate Medical Students.

Background Artificial intelligence (AI) has the potential to be integrated into medical education. Among AI-based technology, large language models (LLMs) such as ChatGPT, Google Bard, Microsoft Bing, and Perplexity have emerged as powerful tools with capabilities in natural language processing. With this background, this study investigates the knowledge, attitude, and practice of undergraduate medical students regarding the utilization of LLMs in medical education in a medical college in Jharkhand, India. Methods A cross-sectional online survey was sent to 370 undergraduate medical students on Google Forms. The questionnaire comprised the following three domains: knowledge, attitude, and practice, each containing six questions. Cronbach's alphas for knowledge, attitude, and practice domains were 0.703, 0.707, and 0.809, respectively. Intraclass correlation coefficients for knowledge, attitude, and practice domains were 0.82, 0.87, and 0.78, respectively. The average scores in the three domains were compared using ANOVA. Results A total of 172 students participated in the study (response rate: 46.49%). The majority of the students (45.93%) rarely used the LLMs for their teaching-learning purposes (chi-square (3) = 41.44, p < 0.0001). The overall score of knowledge (3.21±0.55), attitude (3.47±0.54), and practice (3.26±0.61) were statistically significantly different (ANOVA F (2, 513) = 10.2, p < 0.0001), with the highest score in attitude and lowest in knowledge. Conclusion While there is a generally positive attitude toward the incorporation of LLMs in medical education, concerns about overreliance and potential inaccuracies are evident. LLMs offer the potential to enhance learning resources and provide accessible education, but their integration requires further planning. Further studies are required to explore the long-term impact of LLMs in diverse educational contexts.

Wide Local Excision of Complex or Infected Pilonidal Sinus Followed by Negative Pressure Wound Therapy: Does It Enhance Wound Healing?

Background Pilonidal sinus disease (PSD) is a chronic skin condition caused by hair retention that affects the sacrococcygeal cleft. The purpose of this study is to compare the efficacy of negative pressure wound therapy (NPWT) to routine daily dressings (DDs) in wound healing after complex or infected pilonidal sinus tract excision. Materials and methods The study included 81 individuals who had extensive local excisions for pilonidal sinuses that were complex or infected. Randomly selected participants were given either NPWT or the usual dressing. Between the two groups, the length of hospitalization, the amount of time needed to resume daily activities, and the amount of time needed for full wound closure were compared. Results Forty-two patients received NPWT, while 39 patients received DDs as usual. There was no discernible difference between the two groups in terms of operating time or hospital stay. However, patients who underwent NPWT experienced a quicker final wound closure (59.24 ± 10.21 days compared to routine DD with a mean time of 75.31 ± 14.68 days, P = 0.001) and returned to normal activities earlier (17.36 versus 21.75 days in NPWT and routine DD, respectively). Conclusion Compared to patients who received standard DDs, those who were postoperatively managed with NPWT showed faster wound healing and return to normal activities. Whenever feasible, this strategy may be employed to improve patient recovery.