Semi-supervised deep learning for breast anatomy decomposition in ultrasound images

Li, Yongshuai; Liu, Yuan; Wang, Zhili; Luo, Jianwen

doi:10.1109/ius52206.2021.9593480

Cited by 2 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This model employed an integration strategy to generate relatively stable pseudo-labels and incorporated an uncertainty mapping to further ensure the reliability of these pseudo-labels. Li et al [47] generated pseudolabels for unlabeled data by smoothing network predictions using a simple linear iterative clustering (SLIC) superpixel algorithm.…”

Section: Pseudo-labels-based Suvosmentioning

confidence: 99%

Semi-supervised Learning for Real-time Segmentation of Ultrasound Video Objects: A Review

Jin,

Zhaojun,

Yanping

2023

Advanced Ultrasound in Diagnosis and Therapy

View full text Add to dashboard Cite

Real-time intelligent segmentation of ultrasound video object is a demanding task in the field of medical image processing and serves as an essential and critical step in image-guided clinical procedures. However, obtaining reliable and accurate medical image annotations often necessitates expert guidance, making the acquisition of large-scale annotated datasets challenging and costly. This presents obstacles for traditional supervised learning methods. Consequently, semi-supervised learning (SSL) has emerged as a promising solution, capable of utilizing unlabeled data to enhance model performance and has been widely adopted in medical image segmentation tasks. However, striking a balance between segmentation accuracy and inference speed remains a challenge for real-time segmentation. This paper provides a comprehensive review of research progress in real-time intelligent semi-supervised ultrasound video object segmentation (SUVOS) and offers insights into future developments in this area.

show abstract

Section: Pseudo-labels-based Suvosmentioning

confidence: 99%

Semi-supervised Learning for Real-time Segmentation of Ultrasound Video Objects: A Review

Jin,

Zhaojun,

Yanping

2023

Advanced Ultrasound in Diagnosis and Therapy

View full text Add to dashboard Cite

show abstract

“…In the same paper, the authors also trained and tested their best model on the EchoNet-Dynamic dataset [16], obtaining aaFD of 2.30 for ED and 3.49 for ES frames. Li et al [17] stated only one pair of ED/ES is labeled in each video of the EchoNet-Dynamic dataset, which indicates only a few frames are available for training and makes it challenging for echo phase detection. To address this, they formulated the regression problem as semi-supervised learning, resulting in faster convergence and aaFD of 2.1 frames for ED and 1.7 frames for ES phases.…”

Section: Introductionmentioning

confidence: 99%

Preprocessing and Frame Level Classification Framework for Cardiac Phase Detection in 2D Echocardiography

Singh,

Darji,

Sarvaiya

et al. 2023

Preprint

View full text Add to dashboard Cite

The first and critical step of accurate heart function analysis (i.e., ventricles volume, Ejection Fraction, and stroke volume measurement) in echocardiography is the detection of end-diastole (ED) and end-systole (ES) frames. Detecting these frames is challenging due to variations in cardiac structure, heart rate associated with clinical conditions, and the low-resolution nature of echo sequences. Several deep learning techniques have recently emerged, primarily combining convolutional neural networks (CNN) and recurrent neural networks (RNN). These models were trained on a range of datasets, including open-access sources like CAMUS, PACS, and EchoNet-Dynamic, as well as private datasets. The largest open-access dataset, EchoNet Dynamic, has poor echo phase detection compared to other datasets. This suggests that the dataset has noise, which is removed in this study using three preprocessing steps: Noise reduction using heart rate formulation, Video frame synchronization, and Non-oscillating Mean absolute frame difference (MAFD). Additionally, this study formulates the echo phase detection problem as frame-level binary classification, distinguishing between diastole and systole phases. The proposed architecture takes the echo sequence as an input and a customized time-distributed CNN extracts spatial features from each frame. These frame features are fed into a bidirectional long short-term memory (BLSTM) network, capturing temporal information, followed by a classification layer. The model is trained on the pre-processed EchoNet-Dynamic dataset. Compared to true labels, an average absolute frame distance (aaFD) of 1.02 and 1.04 frames is achieved for ED and ES frames, respectively. Moreover, it operates with an inference time of less than 65 ms for an input sequence consisting of 32 frames on a graphics processing unit (GPU).

show abstract

Semi-supervised deep learning for breast anatomy decomposition in ultrasound images

Cited by 2 publications

References 14 publications

Semi-supervised Learning for Real-time Segmentation of Ultrasound Video Objects: A Review

Semi-supervised Learning for Real-time Segmentation of Ultrasound Video Objects: A Review

Preprocessing and Frame Level Classification Framework for Cardiac Phase Detection in 2D Echocardiography

Contact Info

Product

Resources

About