Related Experiment Video
Updated: Mar 14, 2026
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.4K
Computer-aided diagnosis of papillary thyroid carcinoma based on deep learning technology
Yingzhang Zhou1, Lei Yao2, Long Jin2
1College of Photonic and Electronic Engineering, Fujian Normal University, Fuzhou 350007, PR China.
Critical Reviews in Oncology/Hematology
|March 12, 2026
Summary
Deep learning (DL) shows promise in diagnosing papillary thyroid carcinoma (PTC), a common thyroid cancer subtype. These AI models improve accuracy in analyzing medical images and pathology slides for better patient outcomes.
Area of Science:
- Oncology
- Radiology
- Artificial Intelligence
Background:
- Global thyroid cancer (TC) incidence is rising, with a high prevalence of papillary thyroid carcinoma (PTC) in China.
- Papillary thyroid carcinoma (PTC) is the most common subtype and is linked to poor prognostic indicators, necessitating early detection.
Purpose of the Study:
- To systematically evaluate the application of deep learning (DL) in diagnosing papillary thyroid carcinoma (PTC).
- To assess DL models' ability to identify pathological subtypes using histopathological features and radiological imaging.
Main Methods:
- Review of deep learning (DL) models applied to diagnostic imaging (ultrasound, CT, MRI) and histopathology (H&E staining) for PTC.
- Analysis of DL model performance in detecting key features like microcalcifications, irregular margins, and nuclear grooves.
Main Results:
- DL models demonstrate enhanced accuracy in analyzing radiological and histopathological data for PTC diagnosis.
- Advancements include radio-genomic correlation and molecular feature prediction using DL.
Conclusions:
- Deep learning (DL) offers significant potential for improving PTC diagnostics, risk stratification, and personalized treatment strategies.
- Future research should focus on multicenter data integration, hybrid models, and explainable AI to overcome current limitations.
