An intelligent system for automatic segmentation and grading of thyroid nodules

Abstract

The invention relates to an intelligent system for automatic segmentation and grading of thyroid nodules, which is characterized in that it comprises: a thyroid ultrasound image database; a thyroid ultrasound image preprocessing module; a thyroid nodule feature extraction module: based on U-Net with ResNet34 as the backbone The segmentation model extracts features from the preprocessed thyroid ultrasound images. The U-Net segmentation model with ResNet34 as the backbone includes a downsampling module, a feature fusion module and an upsampling module; a thyroid nodule segmentation module: used to perform feature extraction The extracted image is semantically segmented to form a thyroid nodule segmentation result map.

Description

technical field [0001] The invention relates to image processing and deep learning technology, in particular to an intelligent method for automatic segmentation and grading of thyroid nodules. Background technique [0002] Thyroid cancer is one of the fastest growing cancers in the world. Ultrasound Imaging (Ultrasound Imaging) is widely used in the early diagnosis of thyroid nodules due to its painless, non-destructive, non-radiative, fast, and low-cost advantages. However, due to the principle of ultrasound imaging, thyroid images have disadvantages such as low gray contrast, blurred edges, and many speckle noises. On the other hand, the accuracy of ultrasound diagnosis is affected by the operator's experience, examination skills, and seriousness. With the increase, the labor intensity of doctors will also be greatly increased, which will affect the diagnosis results. In the past decade, deep convolutional neural networks have been widely used in medical image segmentati...

AI Technical Summary

Problems solved by technology

[0003] However, most of the current research only diagnoses benign and malignant nodules, and does not stratify the degree of malignancy of nodules. In clinical practice, patients are mostly diagnosed based on the Kwak TI-RADS grading system. The Kwak TI-RADS grading system summarizes Five malignant features of thyroid nodules were identified, including: 1. Solid nodules; 2. Hypoechoic or very hypoechoic; 3. Small lobes or irregular margins; 4. Gravel-like calcification; 5. Aspect ratio ≥ 1

Images