Breast cancer detection heavily relies on medical imaging, particularly ultrasound, for early diagnosis and effectivetreatment. This research addresses the challenges associated with computer-aided diagnosis (CAD) of ...Breast cancer detection heavily relies on medical imaging, particularly ultrasound, for early diagnosis and effectivetreatment. This research addresses the challenges associated with computer-aided diagnosis (CAD) of breastcancer fromultrasound images. The primary challenge is accurately distinguishing between malignant and benigntumors, complicated by factors such as speckle noise, variable image quality, and the need for precise segmentationand classification. The main objective of the research paper is to develop an advanced methodology for breastultrasound image classification, focusing on speckle noise reduction, precise segmentation, feature extraction, andmachine learning-based classification. A unique approach is introduced that combines Enhanced Speckle ReducedAnisotropic Diffusion (SRAD) filters for speckle noise reduction, U-NET-based segmentation, Genetic Algorithm(GA)-based feature selection, and Random Forest and Bagging Tree classifiers, resulting in a novel and efficientmodel. To test and validate the hybrid model, rigorous experimentations were performed and results state thatthe proposed hybrid model achieved accuracy rate of 99.9%, outperforming other existing techniques, and alsosignificantly reducing computational time. This enhanced accuracy, along with improved sensitivity and specificity,makes the proposed hybrid model a valuable addition to CAD systems in breast cancer diagnosis, ultimatelyenhancing diagnostic accuracy in clinical applications.展开更多
Deep neural network(DNN)based computer-aided breast tumor diagnosis(CABTD)method plays a vital role in the early detection and diagnosis of breast tumors.However,a Brightness mode(B-mode)ultrasound image derives train...Deep neural network(DNN)based computer-aided breast tumor diagnosis(CABTD)method plays a vital role in the early detection and diagnosis of breast tumors.However,a Brightness mode(B-mode)ultrasound image derives training feature samples that make closer isolation toward the infection part.Hence,it is expensive due to a metaheuristic search of features occupying the global region of interest(ROI)structures of input images.Thus,it may lead to the high computational complexity of the pre-trained DNN-based CABTD method.This paper proposes a novel ensemble pretrained DNN-based CABTD method using global-and local-ROI-structures of B-mode ultrasound images.It conveys the additional consideration of a local-ROI-structures for further enhan-cing the pretrained DNN-based CABTD method’s breast tumor diagnostic performance without degrading its visual quality.The features are extracted at various depths(18,50,and 101)from the global and local ROI structures and feed to support vector machine for better classification.From the experimental results,it has been observed that the combined local and global ROI structure of small depth residual network ResNet18(0.8 in%)has produced significant improve-ment in pixel ratio as compared to ResNet50(0.5 in%)and ResNet101(0.3 in%),respectively.Subsequently,the pretrained DNN-based CABTD methods have been tested by influencing local and global ROI structures to diagnose two specific breast tumors(Benign and Malignant)and improve the diagnostic accuracy(86%)compared to Dense Net,Alex Net,VGG Net,and Google Net.Moreover,it reduces the computational complexity due to the small depth residual network ResNet18,respectively.展开更多
BACKGROUND Primary breast diffuse large B-cell lymphoma(PB-DLBCL)is a rare subtype of non-Hodgkin lymphoma that accounts for<3%of extranodal lymphomas and 1%of breast tumors.Its diagnosis and management are challen...BACKGROUND Primary breast diffuse large B-cell lymphoma(PB-DLBCL)is a rare subtype of non-Hodgkin lymphoma that accounts for<3%of extranodal lymphomas and 1%of breast tumors.Its diagnosis and management are challenging because of its rarity,heterogeneity,and aggressive behavior.Conventional ultrasound(US)is the first-line imaging modality for breast lesions;however,it has limited specificity and accuracy for PB-DLBCL.Shear wave elastography(SWE)is a novel US technique that measures tissue stiffness and may reflect the histological characteristics and biological behavior of breast lesions.AIM To compare the conventional US and SWE features of PB-DLBCL and evaluate their diagnostic performance and prognostic value.METHODS We retrospectively reviewed the clinical data and US images of 32 patients with pathologically confirmed PB-DLBCL who underwent conventional US and SWE before treatment.We analyzed conventional US features(shape,margin,orientation,echo,posterior acoustic features,calcification,and vascularity)and SWE features(mean elasticity value,standard deviation,minimum elasticity value,maximum elasticity value,and lesion-to-fat ratio)of the PB-DLBCL lesions.Using receiver operating characteristic curve analysis,we determined the optimal cutoff values and diagnostic performance of conventional US and SWE features.We also performed a survival analysis to assess the prognostic value of conventional US and SWE features.RESULTS The results showed that the PB-DLBCL lesions were mostly irregular in shape(84.4%),microlobulated or spiculated in margins(75%),parallel in orientation(65.6%),hypoechoic in echo(87.5%),and had posterior acoustic enhancement(65.6%).Calcification was rare(6.3%)and vascularity was variable(31.3%avascular,37.5%hypovascular,and 31.3%hypervascular).The mean elasticity value of PB-DLBCL lesions was significantly higher than that of benign breast lesions(113.4±46.9 kPa vs 27.8±16.4 kPa,P<0.001).The optimal cutoff value of the mean elasticity for distinguishing PB-DLBCL from benign breast lesions was 54.5 kPa,with a sensitivity of 93.8%,specificity of 92.9%,positive predictive value of 93.8%,negative predictive value of 92.9%,and accuracy of 93.3%.The mean elasticity value was also significantly correlated with Ki-67 expression level(r=0.612,P<0.001),which is a marker of tumor proliferation and aggressiveness.Survival analysis showed that patients with higher mean elasticity values(>54.5 kPa)had worse overall survival(OS)and progression-free survival(PFS)than those with lower mean elasticity values(<54.5 kPa)(P=0.038 for OS and P=0.027 for PFS).CONCLUSION Conventional US and SWE provide useful information for diagnosing and forecasting PB-DLBCL.SWE excels in distinguishing PB-DLBCL from benign breast lesions,reflects tumor proliferation and aggressiveness,and improves disease management.展开更多
Objective: To discuss the value of contrast-enhanced ultrasound parameters for assessing the nature of breast tumor and their relationship with the malignant biological behavior of tumor. Methods: A total of 176 patie...Objective: To discuss the value of contrast-enhanced ultrasound parameters for assessing the nature of breast tumor and their relationship with the malignant biological behavior of tumor. Methods: A total of 176 patients with breast tumor who received surgical treatment in the hospital between January 2015 and January 2017 were divided into breast adenoma group (n=110) and breast cancer group (n=66) according to pathological results. The differences in breast contrast-enhanced ultrasound parameter levels as well as the expression of proliferation, invasion and autophagy genes in tumor tissues were compared between the two groups of patients. Pearson test was used to evaluate the correlation between contrast-enhanced ultrasound parameters and malignant molecule expression in lesion tissue of patients with breast cancer. Results: PI and AUC levels in breast cancer group were greatly higher than those in breast adenoma group;proliferation genes CXCL1 and Notch1 mRNA expression in tumor tissue were higher than those of breast adenoma group whereas HPK1 mRNA expression was lower than that of breast adenoma group;invasion genes Gab2, NUAK1 and FOXF1 mRNA expression in tumor tissue were higher than those of breast adenoma group whereas NDRG1 mRNA expression was lower than that of breast adenoma group;autophagy genes ATG2B, ATG4D and ATG9B mRNA expression in tumor tissue were lower than those of breast adenoma group. Pearson test showed that the contrast-enhanced ultrasound parameters PI and AUC levels in patients with breast cancer were directly correlated with the expression of proliferation, invasion and autophagy molecules in the lesion tissue. Conclusion: The contrast-enhanced ultrasound parameter levels in patients with breast cancer are different from those in patients with benign tumors, and the specific parameter levels are directly correlated with the tumor malignancy and can be used as the reliable means for early disease screening and malignancy assessment.展开更多
Objective: To study the correlation of contrast-enhanced ultrasound parameters with oncogene expression and cell proliferation activity in breast cancer. Methods: Breast cancer lesions and benign breast lesions surgic...Objective: To study the correlation of contrast-enhanced ultrasound parameters with oncogene expression and cell proliferation activity in breast cancer. Methods: Breast cancer lesions and benign breast lesions surgically removed in Zigong Third People's Hospital between May 2014 and February 2017 were selected, contrast-enhanced ultrasound was done before operation to draw the time-intensity curve and calculate the area under the curve (AUC), and the expression of proliferation molecules and tumor suppressor genes were detected after operation. Results:The contrast-enhanced ultrasound parameter AUC of the breast cancer lesion was greatly higher than that of the benign breast lesion;ECT2, ZKSCAN3, USP39 and EphA2 mRNA expression in breast cancer lesions were obviously higher than those in benign breast lesions whereas HPK1, TCEAL17, CCN5, ATG2B and ATG4D mRNA expression were greatly lower than those in benign breast lesions;ECT2, ZKSCAN3, USP39 and EphA2 mRNA expression in breast cancer lesions with high AUC were greatly higher than those in breast cancer lesions with low AUC whereas HPK1, TCEAL17, CCN5, ATG2B and ATG4D mRNA expression were greatly lower than those in breast cancer lesions with low AUC. Conclusion: The contrast-enhanced ultrasound parameter AUC of breast cancer lesion significantly increases and is closely related to the higher expression of pro-proliferation molecules and the lower expression of tumor suppressor genes.展开更多
目的:探究超声造影与核磁共振成像(contrast-enhanced ultrasound and magnetic resonance imaging, MRI)在诊断乳腺叶状肿瘤(phyllodes tumor of the breast, PTB)良恶性方面的诊断效能。方法:回顾性分析39例PTB(病理诊断)的超声造影和...目的:探究超声造影与核磁共振成像(contrast-enhanced ultrasound and magnetic resonance imaging, MRI)在诊断乳腺叶状肿瘤(phyllodes tumor of the breast, PTB)良恶性方面的诊断效能。方法:回顾性分析39例PTB(病理诊断)的超声造影和MRI特征,比较两种检查方法的诊断效能。结果:良性叶状肿瘤27例,交界性5例,恶性7例。PTB超声造影多呈向心性不均匀高增强伴灌注缺损、增强后边界清晰,其中增强强度、灌注缺损在三组间差异明显(P=0.015、0.009)。PTB的MRI特征以T2WI低信号裂隙、强化不均匀及Ⅱ型TIC曲线类型为主;且T2WI低信号分隔、强化特征及ADC值差异具有统计学意义(P=0.027、0.001、0.042)。MRI对PTB的诊断效能高于CEUS,但差异无统计学意义。结论:超声造影在PTB的表现上具有一定特征性,在诊断PTB良恶性方面与MRI具有较好的一致性。展开更多
基金funded through Researchers Supporting Project Number(RSPD2024R996)King Saud University,Riyadh,Saudi Arabia。
文摘Breast cancer detection heavily relies on medical imaging, particularly ultrasound, for early diagnosis and effectivetreatment. This research addresses the challenges associated with computer-aided diagnosis (CAD) of breastcancer fromultrasound images. The primary challenge is accurately distinguishing between malignant and benigntumors, complicated by factors such as speckle noise, variable image quality, and the need for precise segmentationand classification. The main objective of the research paper is to develop an advanced methodology for breastultrasound image classification, focusing on speckle noise reduction, precise segmentation, feature extraction, andmachine learning-based classification. A unique approach is introduced that combines Enhanced Speckle ReducedAnisotropic Diffusion (SRAD) filters for speckle noise reduction, U-NET-based segmentation, Genetic Algorithm(GA)-based feature selection, and Random Forest and Bagging Tree classifiers, resulting in a novel and efficientmodel. To test and validate the hybrid model, rigorous experimentations were performed and results state thatthe proposed hybrid model achieved accuracy rate of 99.9%, outperforming other existing techniques, and alsosignificantly reducing computational time. This enhanced accuracy, along with improved sensitivity and specificity,makes the proposed hybrid model a valuable addition to CAD systems in breast cancer diagnosis, ultimatelyenhancing diagnostic accuracy in clinical applications.
文摘Deep neural network(DNN)based computer-aided breast tumor diagnosis(CABTD)method plays a vital role in the early detection and diagnosis of breast tumors.However,a Brightness mode(B-mode)ultrasound image derives training feature samples that make closer isolation toward the infection part.Hence,it is expensive due to a metaheuristic search of features occupying the global region of interest(ROI)structures of input images.Thus,it may lead to the high computational complexity of the pre-trained DNN-based CABTD method.This paper proposes a novel ensemble pretrained DNN-based CABTD method using global-and local-ROI-structures of B-mode ultrasound images.It conveys the additional consideration of a local-ROI-structures for further enhan-cing the pretrained DNN-based CABTD method’s breast tumor diagnostic performance without degrading its visual quality.The features are extracted at various depths(18,50,and 101)from the global and local ROI structures and feed to support vector machine for better classification.From the experimental results,it has been observed that the combined local and global ROI structure of small depth residual network ResNet18(0.8 in%)has produced significant improve-ment in pixel ratio as compared to ResNet50(0.5 in%)and ResNet101(0.3 in%),respectively.Subsequently,the pretrained DNN-based CABTD methods have been tested by influencing local and global ROI structures to diagnose two specific breast tumors(Benign and Malignant)and improve the diagnostic accuracy(86%)compared to Dense Net,Alex Net,VGG Net,and Google Net.Moreover,it reduces the computational complexity due to the small depth residual network ResNet18,respectively.
基金This study was reviewed and approved by the Ethics Committee of the Affiliated Hospital of Guizhou Medical University.
文摘BACKGROUND Primary breast diffuse large B-cell lymphoma(PB-DLBCL)is a rare subtype of non-Hodgkin lymphoma that accounts for<3%of extranodal lymphomas and 1%of breast tumors.Its diagnosis and management are challenging because of its rarity,heterogeneity,and aggressive behavior.Conventional ultrasound(US)is the first-line imaging modality for breast lesions;however,it has limited specificity and accuracy for PB-DLBCL.Shear wave elastography(SWE)is a novel US technique that measures tissue stiffness and may reflect the histological characteristics and biological behavior of breast lesions.AIM To compare the conventional US and SWE features of PB-DLBCL and evaluate their diagnostic performance and prognostic value.METHODS We retrospectively reviewed the clinical data and US images of 32 patients with pathologically confirmed PB-DLBCL who underwent conventional US and SWE before treatment.We analyzed conventional US features(shape,margin,orientation,echo,posterior acoustic features,calcification,and vascularity)and SWE features(mean elasticity value,standard deviation,minimum elasticity value,maximum elasticity value,and lesion-to-fat ratio)of the PB-DLBCL lesions.Using receiver operating characteristic curve analysis,we determined the optimal cutoff values and diagnostic performance of conventional US and SWE features.We also performed a survival analysis to assess the prognostic value of conventional US and SWE features.RESULTS The results showed that the PB-DLBCL lesions were mostly irregular in shape(84.4%),microlobulated or spiculated in margins(75%),parallel in orientation(65.6%),hypoechoic in echo(87.5%),and had posterior acoustic enhancement(65.6%).Calcification was rare(6.3%)and vascularity was variable(31.3%avascular,37.5%hypovascular,and 31.3%hypervascular).The mean elasticity value of PB-DLBCL lesions was significantly higher than that of benign breast lesions(113.4±46.9 kPa vs 27.8±16.4 kPa,P<0.001).The optimal cutoff value of the mean elasticity for distinguishing PB-DLBCL from benign breast lesions was 54.5 kPa,with a sensitivity of 93.8%,specificity of 92.9%,positive predictive value of 93.8%,negative predictive value of 92.9%,and accuracy of 93.3%.The mean elasticity value was also significantly correlated with Ki-67 expression level(r=0.612,P<0.001),which is a marker of tumor proliferation and aggressiveness.Survival analysis showed that patients with higher mean elasticity values(>54.5 kPa)had worse overall survival(OS)and progression-free survival(PFS)than those with lower mean elasticity values(<54.5 kPa)(P=0.038 for OS and P=0.027 for PFS).CONCLUSION Conventional US and SWE provide useful information for diagnosing and forecasting PB-DLBCL.SWE excels in distinguishing PB-DLBCL from benign breast lesions,reflects tumor proliferation and aggressiveness,and improves disease management.
文摘Objective: To discuss the value of contrast-enhanced ultrasound parameters for assessing the nature of breast tumor and their relationship with the malignant biological behavior of tumor. Methods: A total of 176 patients with breast tumor who received surgical treatment in the hospital between January 2015 and January 2017 were divided into breast adenoma group (n=110) and breast cancer group (n=66) according to pathological results. The differences in breast contrast-enhanced ultrasound parameter levels as well as the expression of proliferation, invasion and autophagy genes in tumor tissues were compared between the two groups of patients. Pearson test was used to evaluate the correlation between contrast-enhanced ultrasound parameters and malignant molecule expression in lesion tissue of patients with breast cancer. Results: PI and AUC levels in breast cancer group were greatly higher than those in breast adenoma group;proliferation genes CXCL1 and Notch1 mRNA expression in tumor tissue were higher than those of breast adenoma group whereas HPK1 mRNA expression was lower than that of breast adenoma group;invasion genes Gab2, NUAK1 and FOXF1 mRNA expression in tumor tissue were higher than those of breast adenoma group whereas NDRG1 mRNA expression was lower than that of breast adenoma group;autophagy genes ATG2B, ATG4D and ATG9B mRNA expression in tumor tissue were lower than those of breast adenoma group. Pearson test showed that the contrast-enhanced ultrasound parameters PI and AUC levels in patients with breast cancer were directly correlated with the expression of proliferation, invasion and autophagy molecules in the lesion tissue. Conclusion: The contrast-enhanced ultrasound parameter levels in patients with breast cancer are different from those in patients with benign tumors, and the specific parameter levels are directly correlated with the tumor malignancy and can be used as the reliable means for early disease screening and malignancy assessment.
文摘Objective: To study the correlation of contrast-enhanced ultrasound parameters with oncogene expression and cell proliferation activity in breast cancer. Methods: Breast cancer lesions and benign breast lesions surgically removed in Zigong Third People's Hospital between May 2014 and February 2017 were selected, contrast-enhanced ultrasound was done before operation to draw the time-intensity curve and calculate the area under the curve (AUC), and the expression of proliferation molecules and tumor suppressor genes were detected after operation. Results:The contrast-enhanced ultrasound parameter AUC of the breast cancer lesion was greatly higher than that of the benign breast lesion;ECT2, ZKSCAN3, USP39 and EphA2 mRNA expression in breast cancer lesions were obviously higher than those in benign breast lesions whereas HPK1, TCEAL17, CCN5, ATG2B and ATG4D mRNA expression were greatly lower than those in benign breast lesions;ECT2, ZKSCAN3, USP39 and EphA2 mRNA expression in breast cancer lesions with high AUC were greatly higher than those in breast cancer lesions with low AUC whereas HPK1, TCEAL17, CCN5, ATG2B and ATG4D mRNA expression were greatly lower than those in breast cancer lesions with low AUC. Conclusion: The contrast-enhanced ultrasound parameter AUC of breast cancer lesion significantly increases and is closely related to the higher expression of pro-proliferation molecules and the lower expression of tumor suppressor genes.
文摘目的:探究超声造影与核磁共振成像(contrast-enhanced ultrasound and magnetic resonance imaging, MRI)在诊断乳腺叶状肿瘤(phyllodes tumor of the breast, PTB)良恶性方面的诊断效能。方法:回顾性分析39例PTB(病理诊断)的超声造影和MRI特征,比较两种检查方法的诊断效能。结果:良性叶状肿瘤27例,交界性5例,恶性7例。PTB超声造影多呈向心性不均匀高增强伴灌注缺损、增强后边界清晰,其中增强强度、灌注缺损在三组间差异明显(P=0.015、0.009)。PTB的MRI特征以T2WI低信号裂隙、强化不均匀及Ⅱ型TIC曲线类型为主;且T2WI低信号分隔、强化特征及ADC值差异具有统计学意义(P=0.027、0.001、0.042)。MRI对PTB的诊断效能高于CEUS,但差异无统计学意义。结论:超声造影在PTB的表现上具有一定特征性,在诊断PTB良恶性方面与MRI具有较好的一致性。