Table 1 The stages involved in creating the dataset and their respective descriptions.
Stage | Description |
|---|---|
Original Image | The process begins with capturing a raw image of the throat area, which is taken with a mobile phone. |
Manual Check | After initial capture, the image is manually reviewed by specialists to ensure it has adequate clarity and quality for further processing. |
Manual Rotation | If necessary, the image is manually rotated to position the throat area correctly for analysis and diagnosis. |
Crop and Resize | After rotation, the image is cropped and resized to the specified dimensions to prepare it for further processing and AI models. |
Manual Quality Control | The processed image undergoes another manual quality check to ensure it meets high standards, retaining only top-quality images for diagnosis. |
Expert Diagnosis | The final images are reviewed and diagnosed by medical experts, and diagnostic labels are recorded. |
Creation of Labeled Dataset | After expert verification, the labeled images are added to the dataset for use in training machine learning or deep learning models. |
