Related Experiment Video
Updated: Feb 2, 2026
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.3K
Patch-based classification of thyroid nodules in ultrasound images using direction independent features extracted by
Antonin Prochazka1, Sumeet Gulati2, Stepan Holinka3
1Institute of Biophysics and Informatics, 1(st) Faculty of Medicine, Charles University, Salmovska 1, 120 00, Prague, Czech Republic.
Summary
A new computer-aided diagnosis (CAD) system uses direction-independent features from ultrasound images to accurately classify thyroid nodules. This patch-based approach enhances diagnostic accuracy, supporting radiologists in early detection of malignant and benign thyroid conditions.
Area of Science:
- Medical Imaging
- Artificial Intelligence in Medicine
- Oncology
Background:
- Ultrasound imaging is the primary diagnostic tool for early-stage thyroid nodules due to its cost-effectiveness and non-invasive nature.
- Computer-aided diagnosis (CAD) systems can improve radiologist accuracy but are limited by direction-dependent features and static image requirements.
- Current CAD systems often require precise nodule segmentation and are sensitive to image orientation.
Purpose of the Study:
- To design a novel CAD system utilizing direction-independent features for thyroid nodule classification.
- To overcome the limitations of existing CAD systems by developing an orientation-invariant approach.
- To enhance the diagnostic accuracy of ultrasound-based thyroid nodule evaluation.
Main Methods:
- A dataset of 60 thyroid nodules (20 malignant, 40 benign) was analyzed.
- Images were divided into 17x17 pixel patches for feature extraction using Two-Threshold Binary Decomposition.
- Direction-independent features were extracted and classified using Random Forests (RF) and Support Vector Machine (SVM) algorithms.
- A 10-fold cross-validation method was employed for performance evaluation.
Main Results:
- The RF classifier achieved an overall accuracy of 95%, sensitivity of 95%, specificity of 95%, and an Area Under the ROC Curve (AUC) of 0.971.
- The SVM classifier demonstrated an overall accuracy of 91.6%, sensitivity of 95%, specificity of 90%, and an AUC of 0.965.
- The patch-based approach, averaging performance across individual patches, effectively classified whole nodules.
Conclusions:
- The developed patch-based CAD system, using direction-independent features, shows high performance in classifying thyroid nodules.
- This system can serve as a valuable tool for radiologists, augmenting current diagnostic capabilities for thyroid nodules.
- The proposed method offers a robust and accurate solution for computer-aided diagnosis in thyroid ultrasound imaging.
Related Concept Videos
01:02Long-patch Base Excision Repair
8.0K
Since the discovery of the two BER pathways, there has been a debate about how a cell chooses one pathway over the other and the factors determining this selection. Numerous in vitro experiments have pointed out multiple determinants for the sub-pathway selection. These are:
8.0K
02:17Synthesis and Decomposition Reactions
38.2K
Synthesis and decomposition are two types of redox reactions. Synthesis means to make something, whereas decomposition means to break something. The reactions are accompanied by chemical and energy changes.
38.2K
01:26Binary Fission
3.0K
Binary fission is the primary mode of asexual reproduction in prokaryotes, such as bacteria. It results in the production of two genetically identical daughter cells. This highly efficient process ensures the rapid propagation of bacterial populations under favorable conditions and involves coordinated cellular and molecular events.DNA Replication and SeparationThe process begins with the replication of the bacterial chromosome. The circular DNA molecule unwinds at a specific origin of...
3.0K
01:20Binary Fission
63.5K
Fission is the division of a single entity into two or more parts, which regenerate into separate entities that resemble the original. Organisms in the Archaea and Bacteria domains reproduce using binary fission, in which a parent cell splits into two parts that can each grow to the size of the original parent cell. This asexual method of reproduction produces cells that are all genetically identical.
63.5K
01:21Introduction to Test of Independence
3.0K
In statistics, the term independence means that one can directly obtain the probability of any event involving both variables by multiplying their individual probabilities. Tests of independence are chi-square tests involving the use of a contingency table of observed (data) values.
The test statistic for a test of independence is similar to that of a goodness-of-fit test:
The test statistic for a test of independence is similar to that of a goodness-of-fit test:
3.0K
01:16Hypothesis Test for Test of Independence
8.2K
The test of independence is a chi-square-based test used to determine whether two variables or factors are independent or dependent. This hypothesis test is used to examine the independence of the variables. One can construct two qualitative survey questions or experiments based on the variables in a contingency table. The goal is to see if the two variables are unrelated (independent) or related (dependent). The null and alternative hypotheses for this test are:
H0: The two variables (factors)...
H0: The two variables (factors)...
8.2K