Overview
This document serves as a formal Author Correction for the research article originally published in Scientific Reports in October 2023. The correction is issued to update the "Funding" section of the study, ensuring that all supporting grants and governmental bodies are accurately credited for their contributions to the research.
The Nature of the Correction
The primary purpose of this document is administrative transparency. In academic publishing, the accurate disclosure of financial support is essential for institutional accountability and compliance with the requirements of granting agencies. The original publication contained an incomplete funding acknowledgment, which has now been rectified to include additional support from the National Research Foundation of Korea (NRF).
Context of the Original Research
The underlying study addresses a significant challenge in clinical rehabilitation: the analysis of videofluoroscopic swallowing studies (VFSS). VFSS is a radiographic procedure used to evaluate swallowing function and diagnose dysphagia. Traditionally, "temporal analysis"—the measurement of the precise timing of bolus movement and physiological triggers—is a manual, time-consuming process prone to human error and inter-rater variability.
To solve this, the authors applied deep learning technology to automate these measurements. By utilizing neural network architectures, the research aims to increase diagnostic speed and provide objective accuracy in assessing swallowing disorders.
Updated Funding Acknowledgments
The corrected funding section reveals the extensive public investment required for large-scale AI research in medical imaging. The work was supported by several high-level South Korean initiatives:
- The Ministry of Science and ICT (MSIT): Specifically through the ICT Creative Consilience program.
- The Institute for Information & Communications Technology Planning & Evaluation (IITP).
- The National Research Foundation of Korea (NRF): Including specific grants No. 2022R1A5A1027646 and No. 2021R1A2C1007215.
Conclusion
This correction is limited to the funding record and does not alter the scientific methodology, data, or results of the original study. The authors maintain the integrity of their technical findings regarding the application of deep learning to VFSS.
Note to Readers: To review the technical findings, methodology, and data regarding the deep learning model itself, please refer to the original article: “Application of deep learning technology for temporal analysis of videofluoroscopic swallowing studies,” published in Scientific Reports (2023) 13:17546, available at https://doi.org/10.1038/s41598-023-44802-3.
Summary updated