Description Module
The Description Module contains narrative descriptions of the clinical trial, including a brief summary and detailed description. These descriptions provide important information about the study's purpose, methodology, and key details in language accessible to both researchers and the general public.
Description Module path is as follows:
Study -> Protocol Section -> Description Module
Description Module
Ignite Creation Date:
2025-12-24 @ 7:51 PM
Ignite Modification Date:
2025-12-24 @ 7:51 PM
NCT ID:
NCT06636604
Brief Summary:
This project aims to develop an artificial intelligence-based decision support algorithm (DSA) for the detection of the sciatic nerve in safe intramuscular injection applications.
In the first phase of the project,developed an artificial intelligence-based decision support algorithm (DSA) for the detection of the sciatic nerve in safe intramuscular injection applications.
In the second phase of the project, with the integration of the Decision Support Algorithm (DSA) on a computer accompanied by a doctor and an engineer, the presence of the sciatic nerve was tested on 30 volunteers. The test resulted in achieving a 100% sciatic nerve image, confirming the reliability of the DSA.
Subsequently, information about the algorithm was provided to volunteers and nurses in the emergency department of Sakarya Training and Research Hospital who applied for intramuscular injection.
Detailed Description:
This project aims to develop an artificial intelligence-based decision support algorithm (DSA) for the detection of the sciatic nerve in safe intramuscular injection applications.
Method This project was conducted between 01-05-2022 and 01-11-2023.
Data Collection Phase In the first phase of the project, approximately 1500 ultrasound images of the sciatic nerve were obtained from 50 volunteers in right and left positions. Subsequently, the sciatic nerve was labeled on these images. The dimensions of the collected data were normalized, noise reduction was applied, histogram equalization was performed, and class imbalance was addressed by balancing the data. The data were further augmented by applying techniques such as rotation, translation, reflection, and zooming. The YOLOv7 model was chosen for model selection. Parameter optimization was performed for YOLOv7, and preprocessing and data augmentation processes were completed for training the model. The hyperparameters of the YOLOv7 model were manually adjusted, followed by the use of automatic hyperparameter tuning tools, significantly enhancing the model's performance.
Integration and Testing Phase of the DSA In the second phase of the project, with the integration of the Decision Support Algorithm (DSA) on a computer accompanied by a doctor and an engineer, the presence of the sciatic nerve was tested on 30 volunteers subjects based on signals received through a USG probe.
Clinical Trial Phase Subsequently, information about the algorithm was provided to volunteers and nurses in the emergency department of Sakarya Training and Research Hospital who applied for intramuscular injection. Under the supervision of volunteer nurses, project coordinators, and researchers, injections were administered to 30 patients using the DSA and to 30 patients using the traditional method. After completing the injections, nurses filled out and signed the "System Usability Scale" (SUS). Additionally, the "Visual Pain Scale" and "Satisfaction Visual Scale for Injections" were used to assess patients' pain after the injection. Pain levels of volunteer patients were compared immediately after injection and 15 minutes later.
Keywords Nurse, sciatic nerve, decision support algorithm.
Study:
NCT06636604
Study Brief:
Protocol Section:
NCT06636604