Multi‐objectives optimisation of features selection for the classification of thyroid nodules in ultrasound images

Aboudi, Noura; Guetari, Ramzi; Khlifa, Nawrès

doi:10.1049/iet-ipr.2019.1540

Cited by 26 publications

(19 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Work [2,3] collected large-scaled datasets and trained deep learning models on them; however, these datasets are not publicly available. Because of these limitations and challenges, many recent works [4,5,6] used shallow models like SVM, KNN, and logistic regression rather than deep learning based models. Some other work [7] manually pre-processed the images to reduce noises before a neural network model.…”

Section: Input Imagementioning

confidence: 99%

Improving Ultrasound Image Classification with Local Texture Quantisation

Liang

Nagala

et al. 2022

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Ultrasound image classification is important for disease diagnosis. It is more challenging than usual image classification tasks since ultrasound images are difficult to collect and usually contain lots of noise. This paper proposes a novel image classification framework for small-scaled and noisy ultrasound image datasets. The framework first transforms images into discrete index grids. The index grids use discrete indices encoding the local texture patterns of the images. Then, it will conduct classification based on index grids. The proposed framework can significantly reduce the impact of noise as well as the amount of training data that needed. Comparing with existing models, the proposed framework is a lite model and has better explainability. We evaluated the proposed approach on two public ultrasound image datasets for thyroid nodule classification and breast nodule classification. The experiment results show that the proposed approach achieves the new state-of-the-art.

show abstract

Section: Input Imagementioning

confidence: 99%

Improving Ultrasound Image Classification with Local Texture Quantisation

Liang

Nagala

et al. 2022

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…However, ultrasonic images have disadvantages, such as low contrast, low resolution and ease of being polluted by noise, and the rates of missed diagnosis and misdiagnosis by doctors are higher 1 . Therefore, some scholars use traditional machine learning algorithms 2 – 4 to analyze and process ultrasonic images, but this requires the artificial design of feature extraction algorithms, so they are weak and difficult to deploy on large-scale medical data. In contrast, deep learning carries out big data training through the construction of a deep convolutional neural network, and the network learns autonomously and is robust.…”

Section: Introductionmentioning

confidence: 99%

A novel ultrasound image diagnostic method for thyroid nodules

Zheng

Wang

et al. 2023

Sci Rep

View full text Add to dashboard Cite

The incidence of thyroid nodules is increasing year by year. Accurate determination of benign and malignant nodules is an important basis for formulating treatment plans. Ultrasonography is the most widely used methodology in the diagnosis of benign and malignant nodules, but diagnosis by doctors is highly subjective, and the rates of missed diagnosis and misdiagnosis are high. To improve the accuracy of clinical diagnosis, this paper proposes a new diagnostic model based on deep learning. The diagnostic model adopts the diagnostic strategy of localization-classification. First, the distribution laws of the nodule size and nodule aspect ratio are obtained through data statistics, a multiscale localization network structure is a priori designed, and the nodule aspect ratio is obtained from the positioning results. Then, uncropped ultrasound images and nodule area image are correspondingly input into a two-way classification network, and an improved attention mechanism is used to enhance the feature extraction performance. Finally, the deep features, the shallow features, and the nodule aspect ratio are fused, and a fully connected layer is used to complete the classification of benign and malignant nodules. The experimental dataset consists of 4021 ultrasound images, where each image has been labeled under the guidance of doctors, and the ratio of the training set, validation set, and test set sizes is close to 3:1:1. The experimental results show that the accuracy of the multiscale localization network reaches 93.74%, and that the accuracy, specificity, and sensitivity of the classification network reach 86.34%, 81.29%, and 90.48%, respectively. Compared with the champion model of the TNSCUI 2020 classification competition, the accuracy rate is 1.52 points higher. Therefore, the network model proposed in this paper can effectively diagnose benign and malignant thyroid nodules.

show abstract

“…The area under the ROC curve (AUC) reached 97.5% in 75 patients. In 2020, Aboudi et al (2020) employed SVM and random forests (RFs) (Breiman 2001) in the judgement of benign and malignant thyroid nodules. The maximum accuracy of SVM and RFs were 94.28% and 96.13%, respectively.…”

Section: Introductionmentioning

confidence: 99%

Automatic ultrasound diagnosis of thyroid nodules: a combination of deep learning and KWAK TI-RADS

Zhang,

Wang,

Zhao

et al. 2023

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective:There has been a considerable amount of Computer-Aided Diagnosis (CAD) methods highlighted in the field of ultrasonic examination (USE) of thyroid nodules. However, few researches focused on the automatic risk classification, which was the basis for determining whether Fine Needle Aspiration (FNA) was needed. The aim of this work was to implement automatic risk level assessment of thyroid nodules.Approach:Firstly, 1862 cases of thyroid nodules with the results of USE and FNA were collected as the dataset. Then, an improved U-Net++ model was utilized for segmenting thyroid nodules in ultrasound images automatically. Finally, the segmentation result was imported into a multi-task convolutional neural network (MT-CNN), the design of which was based on the clinical guideline called KWAK TI-RADS. Apart from the category of benign and malignant, the MT-CNN also exported the classification result of four malignant features, Solid Component (SC), Hypoechogenicity or Marked Hypoechogenicity (HMH), Microlobulated or Irregular Margin (MIM), Microcalcification (MC), which were used for counting the risk level in KWAK TI-RADS.Main results:The performance of the improved U-Net++ was evaluated on our test set, including 302 cases. The Dice coefficient and Intersection over Union (IoU) reached 0.899, 0.816, respectively. The classification accuracy rates of SC, HMH, MIM, MC, were 94.5%, 92.8%, 86.1%, 88.9%, while the false positive (FP) rate was 6.0%, 5.6%, 10.6%, 12.9% separately. As for the category of benign and malignant, the precision and recall rates were 93.7% and 94.4%.Significance:The proposed CAD method showed favourable performance in the diagnosis of thyroid nodules. Compared with other methods, it could provide reports closer to clinical practice for doctors.

show abstract

Multi‐objectives optimisation of features selection for the classification of thyroid nodules in ultrasound images

Cited by 26 publications

References 35 publications

Improving Ultrasound Image Classification with Local Texture Quantisation

Improving Ultrasound Image Classification with Local Texture Quantisation

A novel ultrasound image diagnostic method for thyroid nodules

Automatic ultrasound diagnosis of thyroid nodules: a combination of deep learning and KWAK TI-RADS

Contact Info

Product

Resources

About