1
|
AlShehri Y, Sidhu A, Lakshmanan LVS, Lefaivre KA. Applications of Natural Language Processing for Automated Clinical Data Analysis in Orthopaedics. J Am Acad Orthop Surg 2024; 32:439-446. [PMID: 38626429 DOI: 10.5435/jaaos-d-23-00839] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/22/2023] [Accepted: 02/20/2024] [Indexed: 04/18/2024] Open
Abstract
Natural language processing is an exciting and emerging field in health care that can transform the field of orthopaedics. It can aid in the process of automated clinical data analysis, changing the way we extract data for various purposes including research and registry formation, diagnosis, and medical billing. This scoping review will look at the various applications of NLP in orthopaedics. Specific examples of NLP applications include identification of essential data elements from surgical and imaging reports, patient feedback analysis, and use of AI conversational agents for patient engagement. We will demonstrate how NLP has proven itself to be a powerful and valuable tool. Despite these potential advantages, there are drawbacks we must consider. Concerns with data quality, bias, privacy, and accessibility may stand as barriers in the way of widespread implementation of NLP technology. As natural language processing technology continues to develop, it has the potential to revolutionize orthopaedic research and clinical practices and enhance patient outcomes.
Collapse
Affiliation(s)
- Yasir AlShehri
- From the Department of Orthopedics, College of Medicine, Imam Abdulrahman Bin Faisal University, Dammam, Saudi Arabia (AlShehri), the Department of Orthopaedics, Faculty of Medicine, The University of British Columbia, Vancouver, BC, Canada (Sidhu and Lefaivre), and the Department of Computer Science, The University of British Columbia, Vancouver, BC, Canada (Lakshmanan)
| | | | | | | |
Collapse
|
2
|
Ozden I, Gokyar M, Ozden ME, Sazak Ovecoglu H. Assessment of artificial intelligence applications in responding to dental trauma. Dent Traumatol 2024. [PMID: 38742754 DOI: 10.1111/edt.12965] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/26/2024] [Revised: 04/15/2024] [Accepted: 04/16/2024] [Indexed: 05/16/2024]
Abstract
BACKGROUND This study assessed the consistency and accuracy of responses provided by two artificial intelligence (AI) applications, ChatGPT and Google Bard (Gemini), to questions related to dental trauma. MATERIALS AND METHODS Based on the International Association of Dental Traumatology guidelines, 25 dichotomous (yes/no) questions were posed to ChatGPT and Google Bard over 10 days. The responses were recorded and compared with the correct answers. Statistical analyses, including Fleiss kappa, were conducted to determine the agreement and consistency of the responses. RESULTS Analysis of 4500 responses revealed that both applications provided correct answers to 57.5% of the questions. Google Bard demonstrated a moderate level of agreement, with varying rates of incorrect answers and referrals to physicians. CONCLUSIONS Although ChatGPT and Google Bard are potential knowledge resources, their consistency and accuracy in responding to dental trauma queries remain limited. Further research involving specially trained AI models in endodontics is warranted to assess their suitability for clinical use.
Collapse
Affiliation(s)
- Idil Ozden
- Department of Endodontics, Marmara University Faculty of Dentistry, Istanbul, Turkey
| | - Merve Gokyar
- Department of Endodontics, Marmara University Faculty of Dentistry, Istanbul, Turkey
| | - Mustafa Enes Ozden
- Department of Public Health, Hacettepe University Faculty of Medicine, Ankara, Turkey
| | - Hesna Sazak Ovecoglu
- Department of Endodontics, Marmara University Faculty of Dentistry, Istanbul, Turkey
| |
Collapse
|
3
|
Thibaut G, Dabbagh A, Liverneaux P. Does Google's Bard Chatbot perform better than ChatGPT on the European hand surgery exam? INTERNATIONAL ORTHOPAEDICS 2024; 48:151-158. [PMID: 37968408 DOI: 10.1007/s00264-023-06034-y] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/07/2023] [Accepted: 11/01/2023] [Indexed: 11/17/2023]
Abstract
PURPOSE According to a previous research, the chatbot ChatGPT® V3.5 was unable to pass the first part of the European Board of Hand Surgery (EBHS) diploma examination. This study aimed to investigate whether Google's chatbot Bard® would have superior performance compared to ChatGPT on the EBHS diploma examination. METHODS Chatbots were asked to answer 18 EBHS multiple choice questions (MCQs) published in the Journal of Hand Surgery (European Volume) in five trials (A1 to A5). After A3, chatbots received correct answers, and after A4, incorrect answers. Consequently, their ability to modify their response was measured and compared. RESULTS Bard® scored 3/18 (A1), 1/18 (A2), 4/18 (A3) and 2/18 (A4 and A5). The average percentage of correct answers was 61.1% for A1, 62.2% for A2, 64.4% for A3, 65.6% for A4, 63.3% for A5 and 63.3% for all trials combined. Agreement was moderate from A1 to A5 (kappa = 0.62 (IC95% = [0.51; 0.73])) as well as from A1 to A3 (kappa = 0.60 (IC95% = [0.47; 0.74])). The formulation of Bard® responses was homogeneous, but its learning capacity is still developing. CONCLUSIONS The main hypothesis of our study was not proved since Bard did not score significantly higher than ChatGPT when answering the MCQs of the EBHS diploma exam. In conclusion, neither ChatGPT® nor Bard®, in their current versions, can pass the first part of the EBHS diploma exam.
Collapse
Affiliation(s)
- Goetsch Thibaut
- Department of Public Health, Strasbourg University Hospital, FMTS, GMRC, 1 avenue de l'hôpital, 67000, Strasbourg cedex, France
| | | | - Philippe Liverneaux
- ICube CNRS UMR7357, Strasbourg University, 2-4 rue Boussingault, 67000, Strasbourg, France.
- Department of Hand Surgery, Strasbourg University Hospitals, FMTS, 1 avenue Molière, 67200, Strasbourg, France.
| |
Collapse
|
4
|
Almodovar JL, Telhan R. The chatbot left us speechless. Acad Emerg Med 2024; 31:103-104. [PMID: 37607043 DOI: 10.1111/acem.14794] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2023] [Revised: 08/11/2023] [Accepted: 08/15/2023] [Indexed: 08/24/2023]
Affiliation(s)
- Jorge L Almodovar
- University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA
| | - Raj Telhan
- University of North Carolina School of Medicine, Chapel Hill, North Carolina, USA
| |
Collapse
|
5
|
Palermi S, Vittadini F, Vecchiato M, Corsini A, Demeco A, Massa B, Pedret C, Dorigo A, Gallo M, Pasta G, Nanni G, Vascellari A, Marchini A, Lempainen L, Sirico F. Managing Lower Limb Muscle Reinjuries in Athletes: From Risk Factors to Return-to-Play Strategies. J Funct Morphol Kinesiol 2023; 8:155. [PMID: 37987491 PMCID: PMC10660751 DOI: 10.3390/jfmk8040155] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 09/16/2023] [Revised: 10/31/2023] [Accepted: 11/02/2023] [Indexed: 11/22/2023] Open
Abstract
Muscle injuries and subsequent reinjuries significantly impact athletes, especially in football. These injuries lead to time loss, performance impairment, and long-term health concerns. This review aims to provide a comprehensive overview of the current understanding of muscle reinjuries, delving into their epidemiology, risk factors, clinical management, and prevention strategies. Despite advancements in rehabilitation programs and return-to-play criteria, reinjury rates remain alarmingly high. Age and previous muscle injuries are nonmodifiable risk factors contributing to a high reinjury rate. Clinical management, which involves accurate diagnosis, individualized rehabilitation plans, and the establishment of return-to-training and return-to-play criteria, plays a pivotal role during the sports season. Eccentric exercises, optimal loading, and training load monitoring are key elements in preventing reinjuries. The potential of artificial intelligence (AI) in predicting and preventing reinjuries offers a promising avenue, emphasizing the need for a multidisciplinary approach to managing these injuries. While current strategies offer some mitigation, there is a pressing need for innovative solutions, possibly leveraging AI, to reduce the incidence of muscle reinjuries in football players. Future research should focus on this direction, aiming to enhance athletes' well-being and performance.
Collapse
Affiliation(s)
- Stefano Palermi
- Public Health Department, University of Naples Federico II, 80131 Naples, Italy
| | | | - Marco Vecchiato
- Sports and Exercise Medicine Division, Department of Medicine, University of Padova, 35128 Padova, Italy
| | | | - Andrea Demeco
- Department of Medicine and Surgery, University of Parma, 43126 Parma, Italy
| | - Bruno Massa
- Public Health Department, University of Naples Federico II, 80131 Naples, Italy
| | - Carles Pedret
- Sports Medicine and Imaging Department, Clinica Diagonal, 08950 Barcelona, Spain;
| | - Alberto Dorigo
- Radiology Unit, Casa di Cura Giovanni XXIII, 31050 Monastier, Italy
| | - Mauro Gallo
- Radiology Unit, Casa di Cura Giovanni XXIII, 31050 Monastier, Italy
| | | | | | | | | | - Lasse Lempainen
- FinnOrthopaedics, Hospital Pihlajalinna, 20520 Turku, Finland;
| | - Felice Sirico
- Public Health Department, University of Naples Federico II, 80131 Naples, Italy
| |
Collapse
|
6
|
Suárez A, Adanero A, Díaz-Flores García V, Freire Y, Algar J. Using a Virtual Patient via an Artificial Intelligence Chatbot to Develop Dental Students’ Diagnostic Skills. INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH 2022; 19:ijerph19148735. [PMID: 35886584 PMCID: PMC9319956 DOI: 10.3390/ijerph19148735] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 06/09/2022] [Revised: 07/13/2022] [Accepted: 07/15/2022] [Indexed: 11/30/2022]
Abstract
Knowing how to diagnose effectively and efficiently is a fundamental skill that a good dental professional should acquire. If students perform a greater number of clinical cases, they will improve their performance with patients. In this sense, virtual patients with artificial intelligence offer a controlled, stimulating, and safe environment for students. To assess student satisfaction after interaction with an artificially intelligent chatbot that recreates a virtual patient, a descriptive cross-sectional study was carried out in which a virtual patient was created with artificial intelligence in the form of a chatbot and presented to fourth and fifth year dental students. After several weeks interacting with the AI, they were given a survey to find out their assessment. A total of 193 students participated. A large majority of the students were satisfied with the interaction (mean 4.36), the fifth year students rated the interaction better and showed higher satisfaction values. The students who reached a correct diagnosis rated this technology more positively. Our research suggests that the incorporation of this technology in dental curricula would be positively valued by students and would also ensure their training and adaptation to new technological developments.
Collapse
Affiliation(s)
- Ana Suárez
- Department of Preclinical Dentistry, School of Biomedical Sciences, Universidad Europea de Madrid, 28670 Madrid, Spain; (A.S.); (V.D.-F.G.); (Y.F.)
| | - Alberto Adanero
- Department of Clinical Dentistry, School of Biomedical Sciences, Universidad Europea de Madrid, 28670 Madrid, Spain;
- Correspondence:
| | - Víctor Díaz-Flores García
- Department of Preclinical Dentistry, School of Biomedical Sciences, Universidad Europea de Madrid, 28670 Madrid, Spain; (A.S.); (V.D.-F.G.); (Y.F.)
| | - Yolanda Freire
- Department of Preclinical Dentistry, School of Biomedical Sciences, Universidad Europea de Madrid, 28670 Madrid, Spain; (A.S.); (V.D.-F.G.); (Y.F.)
| | - Juan Algar
- Department of Clinical Dentistry, School of Biomedical Sciences, Universidad Europea de Madrid, 28670 Madrid, Spain;
| |
Collapse
|
7
|
Feature Extraction of Athlete’s Post-Match Psychological and Emotional Changes Based on Deep Learning. COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE 2022; 2022:2995205. [PMID: 35774441 PMCID: PMC9239801 DOI: 10.1155/2022/2995205] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 11/02/2021] [Revised: 01/19/2022] [Accepted: 01/26/2022] [Indexed: 11/17/2022]
Abstract
Athletes have had to deal with significant shifts in the way they think about psychology and emotion before and after attending a match in their respective fields. It has become increasingly difficult for players of any sport to overcome these differences due to massive technological advancements that aid in analyzing the difficulties of an athlete. The trainer can use the results of the analysis to help motivate and prepare the athletes for the upcoming competitions. The analysis in this study is based on information about the athletes who competed in the Tokyo Olympics. Deep learning models were used to evaluate the study. Image feature detection can be accomplished through the application of a machine learning technique known as deep learning. It employs a neural network, a computer system that mimics the human brain's multiple layers. One or more unique features can be extracted from each layer. A deep learning model called the behavior recognition algorithm is used for the research. The questionnaire from the dataset was used to generate the results of the analysis.
Collapse
|