Journal of Yeungnam Medical Science

Original article

Large language model usage guidelines in Korean medical journals: a survey using human-artificial intelligence collaboration: Sangzin Ahn; Received July 31, 2024 Accepted November 21, 2024 Published online December 11, 2024; DOI: https://doi.org/10.12701/jyms.2024.00794 [Epub ahead of print]

124 View
9 Download

Abstract PDF Supplementary Material: Background
Large language models (LLMs), the most recent advancements in artificial intelligence (AI), have profoundly affected academic publishing and raised important ethical and practical concerns. This study examined the prevalence and content of AI guidelines in Korean medical journals to assess the current landscape and inform future policy implementation.
Methods
The top 100 Korean medical journals determined by Hirsh index were surveyed. Author guidelines were collected and screened by a human researcher and AI chatbot to identify AI-related content. The key components of LLM policies were extracted and compared across journals. The journal characteristics associated with the adoption of AI guidelines were also analyzed.
Results
Only 18% of the surveyed journals had LLM guidelines, which is much lower than previously reported in international journals. However, the adoption rates increased over time, reaching 57.1% in the first quarter of 2024. High-impact journals were more likely to have AI guidelines. All journals with LLM guidelines required authors to declare LLM tool use and 94.4% prohibited AI authorship. The key policy components included emphasizing human responsibility (72.2%), discouraging AI-generated content (44.4%), and exempting basic AI tools (38.9%).
Conclusion
While the adoption of LLM guidelines among Korean medical journals is lower than the global trend, there has been a clear increase in implementation over time. The key components of these guidelines align with international standards, but greater standardization and collaboration are needed to ensure the responsible and ethical use of LLMs in medical research and writing.

Case report

Development of an automated foot contact area measurement program for podoscopes using ChatGPT-4: a case report: Min Cheol Chang; Received October 28, 2024 Accepted November 18, 2024 Published online December 3, 2024; DOI: https://doi.org/10.12701/jyms.2024.01326 [Epub ahead of print]

188 View
17 Download

Abstract PDF: Accurate measurement of the foot contact area is crucial for diagnosing pes planus (flatfoot) and pes cavus (high arch), which significantly affect pressure distribution across the plantar surface. This study aimed to develop a program using ChatGPT-4 to automate foot contact area measurements using a podoscope, thereby enhancing diagnostic precision. A 53-year-old female volunteer stood on a podoscope to capture images of her feet, which were processed to isolate the foot contours and measure the contact areas. A program developed utilizing ChatCPT-4 was designed to outline the feet, detect contact areas, and calculate their sizes and ratios. The results demonstrated clear visualization of foot contours with automated calculation of the contact area and its ratio to the total foot area. The entire foot area measured 1,091,381.00 pixels, with a contact area of 604,252.50 pixels. The ratio of the ground contact area to the entire foot area was calculated as 55.37%. This method, which employs advanced image-processing techniques powered by ChatGPT-4, demonstrates the potential for integrating artificial intelligence into clinical applications. This approach could improve diagnostic precision and patient outcomes through personalized treatment strategies.

Review articles

Advances and utility of digital twins in critical care and acute care medicine: a narrative review: Gabriele A. Halpern, Marko Nemet, Diksha M. Gowda, Oguz Kilickaya, Amos Lal; Received September 11, 2024 Accepted October 8, 2024 Published online November 25, 2024; DOI: https://doi.org/10.12701/jyms.2024.01053 [Epub ahead of print]

270 View
14 Download

Abstract PDF: Artificial intelligence (AI) has shown promise for revolutionizing healthcare. This narrative review focuses on the evolving discussion of the utility of AI and clinical informatics in critical care and acute care medicine, specifically focusing on digital twin (DT) technology. The improved computational power and iterative validation of these intelligent tools have enhanced medical education, in silico research, and clinical decision support in critical care settings. Integrating DTs into critical care opens vast opportunities, but simultaneously poses complex challenges, from data safety and privacy concerns to potentially increasing healthcare disparities. In medicine, DTs can significantly improve the efficiency of critical care systems. Stakeholder investment is essential for successful implementation and integration of these technologies.

Advances, challenges, and prospects of electroencephalography-based biomarkers for psychiatric disorders: a narrative review: Seokho Yun; J Yeungnam Med Sci. 2024;41(4):261-268. Published online September 9, 2024; DOI: https://doi.org/10.12701/jyms.2024.00668

955 View
55 Download

Abstract PDF: Owing to a lack of appropriate biomarkers for accurate diagnosis and treatment, psychiatric disorders cause significant distress and functional impairment, leading to social and economic losses. Biomarkers are essential for diagnosing, predicting, treating, and monitoring various diseases. However, their absence in psychiatry is linked to the complex structure of the brain and the lack of direct monitoring modalities. This review examines the potential of electroencephalography (EEG) as a neurophysiological tool for identifying psychiatric biomarkers. EEG noninvasively measures brain electrophysiological activity and is used to diagnose neurological disorders, such as depression, bipolar disorder (BD), and schizophrenia, and identify psychiatric biomarkers. Despite extensive research, EEG-based biomarkers have not been clinically utilized owing to measurement and analysis constraints. EEG studies have revealed spectral and complexity measures for depression, brainwave abnormalities in BD, and power spectral abnormalities in schizophrenia. However, no EEG-based biomarkers are currently used clinically for the treatment of psychiatric disorders. The advantages of EEG include real-time data acquisition, noninvasiveness, cost-effectiveness, and high temporal resolution. Challenges such as low spatial resolution, susceptibility to interference, and complexity of data interpretation limit its clinical application. Integrating EEG with other neuroimaging techniques, advanced signal processing, and standardized protocols is essential to overcome these limitations. Artificial intelligence may enhance EEG analysis and biomarker discovery, potentially transforming psychiatric care by providing early diagnosis, personalized treatment, and improved disease progression monitoring.

Original article

Classification of dental implant systems using cloud-based deep learning algorithm: an experimental study: Hyun Jun Kong; J Yeungnam Med Sci. 2023;40(Suppl):S29-S36. Published online July 26, 2023; DOI: https://doi.org/10.12701/jyms.2023.00465

3,277 View
153 Download
4 Web of Science
6 Crossref

Abstract PDF

Background
This study aimed to evaluate the accuracy and clinical usability of implant system classification using automated machine learning on a Google Cloud platform.
Methods
Four dental implant systems were selected: Osstem TSIII, Osstem USII, Biomet 3i Os-seotite External, and Dentsply Sirona Xive. A total of 4,800 periapical radiographs (1,200 for each implant system) were collected and labeled based on electronic medical records. Regions of interest were manually cropped to 400×800 pixels, and all images were uploaded to Google Cloud storage. Approximately 80% of the images were used for training, 10% for validation, and 10% for testing. Google automated machine learning (AutoML) Vision automatically executed a neural architecture search technology to apply an appropriate algorithm to the uploaded data. A single-label image classification model was trained using AutoML. The performance of the mod-el was evaluated in terms of accuracy, precision, recall, specificity, and F1 score.
Results
The accuracy, precision, recall, specificity, and F1 score of the AutoML Vision model were 0.981, 0.963, 0.961, 0.985, and 0.962, respectively. Osstem TSIII had an accuracy of 100%. Osstem USII and 3i Osseotite External were most often confused in the confusion matrix.
Conclusion
Deep learning-based AutoML on a cloud platform showed high accuracy in the classification of dental implant systems as a fine-tuned convolutional neural network. Higher-quality images from various implant systems will be required to improve the performance and clinical usability of the model.

Citations

Citations to this article as recorded by

Accuracy of Artificial Intelligence Models in Dental Implant Fixture Identification and Classification from Radiographs: A Systematic Review
Wael I. Ibraheem
Diagnostics.2024; 14(8): 806. CrossRef
A Comparative Analysis of Deep Learning-Based Approaches for Classifying Dental Implants Decision Support System
Mohammed A. H. Lubbad, Ikbal Leblebicioglu Kurtulus, Dervis Karaboga, Kerem Kilic, Alper Basturk, Bahriye Akay, Ozkan Ufuk Nalbantoglu, Ozden Melis Durmaz Yilmaz, Mustafa Ayata, Serkan Yilmaz, Ishak Pacal
Journal of Imaging Informatics in Medicine.2024; 37(5): 2559. CrossRef
Applications of Machine Learning in Periodontology and Implantology: A Comprehensive Review
Cristiana Adina Șalgău, Anca Morar, Andrei Daniel Zgarta, Diana-Larisa Ancuța, Alexandros Rădulescu, Ioan Liviu Mitrea, Andrei Ovidiu Tănase
Annals of Biomedical Engineering.2024; 52(9): 2348. CrossRef
Artificial neural networks development in prosthodontics - a systematic mapping review
Olivia Bobeică, Denis Iorga
Journal of Dentistry.2024; 151: 105385. CrossRef
Fracture strength of poly ether ether ketone abutment over short implant after fatigue
Mohamed A.E. Elsayed, Radwa A. El-dessouky, Mahmoud A.-A. Shakal
Tanta Dental Journal.2024; 21(3): 288. CrossRef
Advancements in artificial intelligence algorithms for dental implant identification: A systematic review with meta-analysis
Ahmed Yaseen Alqutaibi, Radhwan S. Algabri, Dina Elawady, Wafaa Ibrahim Ibrahim
The Journal of Prosthetic Dentistry.2023;[Epub] CrossRef

First
Prev
Page of 1
Next
Last

Search