Ultrasonic Diagnosis of Breast Nodules Using Modified Faster R-CNN

Zhang, Zihao; Zhang, Xuesheng; Lin, Xiaona; Dong, Licong; Zhang, Sure; Zhang, Xueling; Sun, Desheng; Yuan, Kehong

doi:10.1177/0161734619882683

Cited by 25 publications

(24 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further, Yolo V3 employs the sigmoid function as a tool for predicting the activation function. The sigmoid function solves the problem efficiently, while the equal target has two labels [39,41,42].…”

Section: Traffic Sign Recognition With You Only Look Once (Yolo) V3mentioning

confidence: 99%

Deep Learning for Traffic Sign Recognition Based on Spatial Pyramid Pooling with Scale Analysis

et al. 2020

View full text Add to dashboard Cite

In the area of traffic sign detection (TSD) methods, deep learning has been implemented and achieves outstanding performance. The detection of a traffic sign, as it has a dual function in monitoring and directing the driver, is a big concern for driver support systems. A core feature of autonomous vehicle systems is the identification of the traffic sign. This article focuses on the prohibitive sign. The objective is to detect in real-time and reduce processing time considerably. In this study, we implement the spatial pyramid pooling (SPP) principle to boost Yolo V3’s backbone network for the extraction of functionality. Our work uses SPP for more comprehensive learning of multiscale object features. Then, perform a comparative investigation of Yolo V3 and Yolo V3 SPP across various scales to recognize the prohibitory sign. Comparisons with Yolo V3 SPP models reveal that their mean average precision (mAP) is higher than Yolo V3. Furthermore, the test accuracy findings indicate that the Yolo V3 SPP model performs better than Yolo V3 for different sizes.

show abstract

Section: Traffic Sign Recognition With You Only Look Once (Yolo) V3mentioning

confidence: 99%

Deep Learning for Traffic Sign Recognition Based on Spatial Pyramid Pooling with Scale Analysis

et al. 2020

View full text Add to dashboard Cite

show abstract

“…An AI-based colonoscopy system has been shown to accurately differentiate neoplastic lesions from nonneoplastic lesions on stained endocytoscopic images and endocytoscopic narrow-band images in endoscopic evaluation of small colon polyps (7). The application of AI for US-based diagnosis has been mainly reported for the diagnosis of malignant tumors, such as mammary and thyroid cancers (8)(9)(10)(11). Le et al reported an AI model for the diagnosis of thyroid cancer pre-trained with 312,399 B-mode US images of cancer and healthy controls (12).…”

Section: History and Recent Progress Of Ai In Medical Imagingmentioning

confidence: 99%

Artificial Intelligence in Medical Imaging and Its Application in Sonography for the Management of Liver Tumor

Nishida

Kudo

2020

Front. Oncol.

View full text Add to dashboard Cite

Recent advancement in artificial intelligence (AI) facilitate the development of AI-powered medical imaging including ultrasonography (US). However, overlooking or misdiagnosis of malignant lesions may result in serious consequences; the introduction of AI to the imaging modalities may be an ideal solution to prevent human error. For the development of AI for medical imaging, it is necessary to understand the characteristics of modalities on the context of task setting, required data sets, suitable AI algorism, and expected performance with clinical impact. Regarding the AI-aided US diagnosis, several attempts have been made to construct an image database and develop an AI-aided diagnosis system in the field of oncology. Regarding the diagnosis of liver tumors using US images, 4- or 5-class classifications, including the discrimination of hepatocellular carcinoma (HCC), metastatic tumors, hemangiomas, liver cysts, and focal nodular hyperplasia, have been reported using AI. Combination of radiomic approach with AI is also becoming a powerful tool for predicting the outcome in patients with HCC after treatment, indicating the potential of AI for applying personalized medical care. However, US images show high heterogeneity because of differences in conditions during the examination, and a variety of imaging parameters may affect the quality of images; such conditions may hamper the development of US-based AI. In this review, we summarized the development of AI in medical images with challenges to task setting, data curation, and focus on the application of AI for the managements of liver tumor, especially for US diagnosis.

show abstract

“…Since the initial release of the BI-RADS lexicon for ultrasound (BI-RADS-US) in 2003 [ 4 ], together with the rapid development of technique, ultrasound is getting more used for breast screening, especially in Asian women with dense breasts [ 5 ]. Compare to mammography, ultrasound examination is a portable, real-time, non-invasive, non-radiative, inexpensive, and highly reproducible method, and well accepted by patients [ 5 – 7 ]. BI-RADS-US is confirmed to be feasible to offer typical features for benign or malignant breast lesions [ 8 , 9 ].…”

Section: Introductionmentioning

confidence: 99%

Breast lesions excised via vacuum-assisted system: could we get any clues for B3 lesions before excision biopsy?

Zheng

Xing

et al. 2021

BMC Cancer

View full text Add to dashboard Cite

Background The purpose of this study was to determine the validity of the ultrasound features as well as patient characteristics assigned to B3 (uncertain malignant potential) breast lesions before vacuum-assisted excision biopsy (VAEB). Methods This study population consisted of 2245 women with breast-nodular abnormalities, which were conducted ultrasound-guided VAEB (US-VAEB). Patient’s clinical and anamnestic data and lesion-related ultrasonic feature variables of B3 captured before US-VAEB were compared with those of benign or malignant cases, using histopathological results as a benchmark. Results The proportions of benign, B3 and malignant breast lesions diagnosed post-US-VAEB were 88.5, 8.2 and 3.4% respectively. B3 high frequent occurred in BI-RADS-US grade 3 (7.7%), grade 4a (11.0%) and grade 4b (9.1%). The overall malignancy underestimation rate of B3 was 4.4% (8/183). Malignant lesions were found mostly in the range of BI-RADS grade 4b (27.3%), grade 4c (33.3%) and grade 5 (100%). Multivariate binary logistic regression analyses (B3 vs benign) showed that non-menopausal patients (95% CI 1.628–8.616, P = 0.002), single (95% CI 1.370–2.650, P = 0.000) or vascularity (95% CI 1.745–4.150, P = 0.000) nodules in ultrasonic features were significant risk factors for B3 occurrences. In addition, patients elder than 50 years (95% CI 3.178–19.816, P = 0.000), unclear margin (95% CI 3.571–14.119, P = 0.000) or suspicious calcification (95% CI 4.010–30.733, P = 0.000) lesions were significantly associated with higher risks of malignant potentials for B3 cases (malignant vs B3). Conclusion The results of this study indicate that ultrasound findings and patients’ characteristics might provide valuable information for distinguishing B3 lesions from benign breast abnormalities before VAEB, and help to reduce malignancy underestimation of B3.

show abstract

Ultrasonic Diagnosis of Breast Nodules Using Modified Faster R-CNN

Cited by 25 publications

References 21 publications

Deep Learning for Traffic Sign Recognition Based on Spatial Pyramid Pooling with Scale Analysis

Deep Learning for Traffic Sign Recognition Based on Spatial Pyramid Pooling with Scale Analysis

Artificial Intelligence in Medical Imaging and Its Application in Sonography for the Management of Liver Tumor

Breast lesions excised via vacuum-assisted system: could we get any clues for B3 lesions before excision biopsy?

Contact Info

Product

Resources

About