Dual-Intended Deep Learning Model for Breast Cancer Diagnosis in Ultrasound Imaging

Vigil, Nicolle; Barry, Madeline; Amini, Arya; Akhloufi, Moulay A.; Maldague, Xavier; Ma, Li; Ren, Lei; Yousefi, Bardia

doi:10.3390/cancers14112663

Cited by 24 publications

(3 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, a constraint of this study has been the limited availability of images in the publicly accessible dataset for breast cancer ultrasound imaging, which impacts the performance of the model. Vigil et al [21] have conducted a study that has explored the results of predictive and diagnostic systems in the context of breast cancer through ultrasound images. They have emphasized the importance of achieving higher accuracy and enabling early diagnosis facilitated by diverse imaging modalities and artificial intelligence models.…”

Section: Related Workmentioning

confidence: 99%

Integrating Deep and Handcrafted Features forEnhanced Decision-Making Assistance in BreastCancer Diagnosis on Ultrasound Images

Abhisheka,

Biswas,

Purkayastha

et al. 2023

Preprint

View full text Add to dashboard Cite

The classification of breast cancer has emerged as a significant concern in thehealthcare sector in recent times. This is primarily due to its status as the sec-ond leading cause of cancer-related fatalities among women. Prior research inthe field of breast cancer classification has mainly focused on either traditionalMachine Learning (ML) based models or employing Deep Learning (DL) mod-els. However, both approaches solely fail to capture the complete spectrum offeatures necessary for accurate classification. Therefore, this paper introduces aclassification model named Hybrid Breast Cancer Prediction System (HBCPS),which utilizes a combination of features derived from deep Convolutional Neu-ral Network (CNN) model and handcrafted features, to improve classificationperformance. The proposed HBCPS uses the pre-trained ResNet50 network toextract the deep features, while the handcrafted features are obtained using His-togram Orientation Gradient (HOG). For classification the proposed approach uses Support Vector Machine (SVM). In addition, the proposed method inte-grates the Block Matching and 3D (BM3D) denoising filter. This filter efficientlyreduces multiplicative noise such as speckle noise from Breast Ultrasound (BUS)images, resulting in improved image quality. This enhancement technique hassignificantly contributed to improve the overall performance of the system. Theproposed framework is evaluated using a widely used Breast Ultrasound Image(BUSI) dataset. The proposed HBCPS exhibited satisfactory performance withan accuracy, precision, recall, F1-score, specificity, and AUC values of 89.02%,87.67%, 87.17%, 87.36%, 83.87%, and 0.8717, respectively. These metrics indicatethe robustness and reliability of the proposed method and offer a comprehensiveframework for breast cancer classification using BUS images.

show abstract

Section: Related Workmentioning

confidence: 99%

Integrating Deep and Handcrafted Features forEnhanced Decision-Making Assistance in BreastCancer Diagnosis on Ultrasound Images

Abhisheka,

Biswas,

Purkayastha

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…91 The current review extracted and organized the data in tabular form and summarized the application of US images in breast cancer diagnosis (Table 2). 6,8,80,83,90,92–118…”

Section: Introductionmentioning

confidence: 99%

Machine learning and new insights for breast cancer diagnosis

Guo,

Zhang,

Yuan

et al. 2024

J Int Med Res

View full text Add to dashboard Cite

Breast cancer (BC) is the most prominent form of cancer among females all over the world. The current methods of BC detection include X-ray mammography, ultrasound, computed tomography, magnetic resonance imaging, positron emission tomography and breast thermographic techniques. More recently, machine learning (ML) tools have been increasingly employed in diagnostic medicine for its high efficiency in detection and intervention. The subsequent imaging features and mathematical analyses can then be used to generate ML models, which stratify, differentiate and detect benign and malignant breast lesions. Given its marked advantages, radiomics is a frequently used tool in recent research and clinics. Artificial neural networks and deep learning (DL) are novel forms of ML that evaluate data using computer simulation of the human brain. DL directly processes unstructured information, such as images, sounds and language, and performs precise clinical image stratification, medical record analyses and tumour diagnosis. Herein, this review thoroughly summarizes prior investigations on the application of medical images for the detection and intervention of BC using radiomics, namely DL and ML. The aim was to provide guidance to scientists regarding the use of artificial intelligence and ML in research and the clinic.

show abstract

“…We introduce a supplementary study of VWT in Appendix A . Yousefi et al [ 17 , 18 , 19 ] employed PCA in the dynamic data to preprocess vein data for further artificial intelligence (A.I.) interpretation.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Vascular Imaging Using Active Breast Thermography

Gershenson

2023

Sensors

View full text Add to dashboard Cite

Mammography is considered the gold standard for breast cancer screening and diagnostic imaging; however, there is an unmet clinical need for complementary methods to detect lesions not characterized by mammography. Far-infrared ‘thermogram’ breast imaging can map the skin temperature, and signal inversion with components analysis can be used to identify the mechanisms of thermal image generation of the vasculature using dynamic thermal data. This work focuses on using dynamic infrared breast imaging to identify the thermal response of the stationary vascular system and the physiologic vascular response to a temperature stimulus affected by vasomodulation. The recorded data are analyzed by converting the diffusive heat propagation into a virtual wave and identifying the reflection using component analysis. Clear images of passive thermal reflection and thermal response to vasomodulation were obtained. In our limited data, the magnitude of vasoconstriction appears to depend on the presence of cancer. The authors propose future studies with supporting diagnostic and clinical data that may provide validation of the proposed paradigm.

show abstract

Dual-Intended Deep Learning Model for Breast Cancer Diagnosis in Ultrasound Imaging

Cited by 24 publications

References 55 publications

Integrating Deep and Handcrafted Features forEnhanced Decision-Making Assistance in BreastCancer Diagnosis on Ultrasound Images

Integrating Deep and Handcrafted Features forEnhanced Decision-Making Assistance in BreastCancer Diagnosis on Ultrasound Images

Machine learning and new insights for breast cancer diagnosis

Dynamic Vascular Imaging Using Active Breast Thermography

Contact Info

Product

Resources

About