The Ki67 index (KI) is a standard clinical marker for tumor proliferation;however, its application is hindered by intratumoral heterogeneity. In this study, we used digital image analysis to comprehensively analyze Ki...The Ki67 index (KI) is a standard clinical marker for tumor proliferation;however, its application is hindered by intratumoral heterogeneity. In this study, we used digital image analysis to comprehensively analyze Ki67 heterogeneity and distribution patterns in breast carcinoma. Using Smart Pathology software, we digitized and analyzed 42 excised breast carcinoma Ki67 slides. Boxplots, histograms, and heat maps were generated to illustrate the KI distribution. We found that 30% of cases (13/42) exhibited discrepancies between global and hotspot KI when using a 14% KI threshold for classification. Patients with higher global or hotspot KI values displayed greater heterogenicity. Ki67 distribution patterns were categorized as randomly distributed (52%, 22/42), peripheral (43%, 18/42), and centered (5%, 2/42). Our sampling simulator indicated analyzing more than 10 high-power fields was typically required to accurately estimate global KI, with sampling size being correlated with heterogeneity. In conclusion, using digital image analysis in whole-slide images allows for comprehensive Ki67 profile assessment, shedding light on heterogeneity and distribution patterns. This spatial information can facilitate KI surveys of breast cancer and other malignancies.展开更多
AIM:To automate breast cancer diagnosis and to study the inter-observer and intra-observer variations in the manual evaluations.METHODS:Breast tissue specimens from sixty cases were stained separately for estrogen rec...AIM:To automate breast cancer diagnosis and to study the inter-observer and intra-observer variations in the manual evaluations.METHODS:Breast tissue specimens from sixty cases were stained separately for estrogen receptor(ER),progesterone receptor(PR)and human epidermal growth factor receptor-2(HER-2/neu).All cases were assessed by manual grading as well as image analysis.The manual grading was performed by an experienced expert pathologist.To study inter-observer and intra-observer variations,we obtained readings from another pathologist as the second observer from a different laboratory who has a little less experience than the first observer.We also took a second reading from the second observer to study intra-observer variations.Image analysis was carried out using in-house developed software(TissueQuant).A comparison of the results from image analysis and manual scoring of ER,PR and HER-2/neu was also carried out.RESULTS:The performance of the automated analysis in the case of ER,PR and HER-2/neu expressions was compared with the manual evaluations.The performance of the automated system was found to correlate well with the manual evaluations.The inter-observer variations were measured using Spearman correlation coefficient r and 95%confidence interval.In the case of ER expression,Spearman correlation r=0.53,in the case of PR expression,r=0.63,and in the case of HER-2/neu expression,r=0.68.Similarly,intra-observer variations were also measured.In the case of ER,PR and HER-2/neu expressions,r=0.46,0.66 and 0.70,respectively.CONCLUSION:The automation of breast cancer diagnosis from immunohistochemically stained specimens is very useful for providing objective and repeatable evaluations.展开更多
Objective To investigate the difference in texture features on diffusion weighted imaging(DWI) images between breast benign and malignant tumors.Methods Patients including 56 with mass-like breast cancer, 16 with brea...Objective To investigate the difference in texture features on diffusion weighted imaging(DWI) images between breast benign and malignant tumors.Methods Patients including 56 with mass-like breast cancer, 16 with breast fibroadenoma, and 4 with intraductal papilloma of breast treated in the Hainan Hospital of Chinese PLA General Hospital were retrospectively enrolled in this study, and allocated to the benign group(20 patients) and the malignant group(56 patients) according to the post-surgically pathological results. Texture analysis was performed on axial DWI images, and five characteristic parameters including Angular Second Moment(ASM), Contrast, Correlation, Inverse Difference Moment(IDM), and Entropy were calculated. Independent sample t-test and Mann-Whitney U test were performed for intergroup comparison. Regression model was established by using Binary Logistic regression analysis, and receiver operating characteristic curve(ROC) analysis was carried out to evaluate the diagnostic efficiency. Results The texture features ASM, Contrast, Correlation and Entropy showed significant differences between the benign and malignant breast tumor groups(PASM= 0.014, Pcontrast= 0.019, Pcorrelation= 0.010, Pentropy= 0.007). The area under the ROC curve was 0.685, 0.681, 0.754, and 0.683 respectively for the positive texture variables mentioned above, and that for the combined variables(ASM, Contrast, and Entropy) was 0.802 in the model of Logistic regression. Binary Logistic regression analysis demonstrated that ASM, Contrast and Entropy were considered as thespecific imaging variables for the differential diagnosis of breast benign and malignant tumors.Conclusion The texture analysis of DWI may be a simple and effective tool in the differential diagnosis between breast benign and malignant tumors.展开更多
BACKGROUND A positive resection margin is a major risk factor for local breast cancer recurrence after breast-conserving surgery(BCS).Preoperative imaging examinations are frequently employed to assess the surgical ma...BACKGROUND A positive resection margin is a major risk factor for local breast cancer recurrence after breast-conserving surgery(BCS).Preoperative imaging examinations are frequently employed to assess the surgical margin.AIM To investigate the role and value of preoperative imaging examinations[magnetic resonance imaging(MRI),molybdenum target,and ultrasound]in evaluating margins for BCS.METHODS A retrospective study was conducted on 323 breast cancer patients who met the criteria for BCS and consented to the procedure from January 2014 to July 2021.The study gathered preoperative imaging data(MRI,ultrasound,and molybdenum target examination)and intraoperative and postoperative pathological information.Based on their BCS outcomes,patients were categorized into positive and negative margin groups.Subsequently,the patients were randomly split into a training set(226 patients,approximately 70%)and a validation set(97 patients,approximately 30%).The imaging and pathological information was analyzed and summarized using R software.Non-conditional logistic regression and LASSO regression were conducted in the validation set to identify factors that might influence the failure of BCS.A column chart was generated and applied to the validation set to examine the relationship between pathological margin range and prognosis.This study aims to identify the risk factors associated with failure in BCS.RESULTS The multivariate non-conditional logistic regression analysis demonstrated that various factors raise the risk of positive margins following BCS.These factors comprise non-mass enhancement(NME)on dynamic contrastenhanced MRI,multiple focal vascular signs around the lesion on MRI,tumor size exceeding 2 cm,type III timesignal intensity curve,indistinct margins on molybdenum target examination,unclear margins on ultrasound examination,and estrogen receptor(ER)positivity in immunohistochemistry.LASSO regression was additionally employed in this study to identify four predictive factors for the model:ER,molybdenum target tumor type(MT Xmd Shape),maximum intensity projection imaging feature,and lesion type on MRI.The model constructed with these predictive factors exhibited strong consistency with the real-world scenario in both the training set and validation set.Particularly,the outcomes of the column chart model accurately predicted the likelihood of positive margins in BCS.CONCLUSION The proposed column chart model effectively predicts the success of BCS for breast cancer.The model utilizes preoperative ultrasound,molybdenum target,MRI,and core needle biopsy pathology evaluation results,all of which align with the real-world scenario.Hence,our model can offer dependable guidance for clinical decisionmaking concerning BCS.展开更多
AIM: To build and evaluate predictive models for contrast-enhanced ultrasound(CEUS) of the breast to distinguish between benign and malignant lesions. METHODS: A total of 235 breast imaging reporting and data system(B...AIM: To build and evaluate predictive models for contrast-enhanced ultrasound(CEUS) of the breast to distinguish between benign and malignant lesions. METHODS: A total of 235 breast imaging reporting and data system(BI-RADS) 4 solid breast lesions were imaged via CEUS before core needle biopsy or surgical resection. CEUS results were analyzed on 10 enhancing patterns to evaluate diagnostic performance of three benign and three malignant CEUS models, with pathological results used as the gold standard. A logistic regression model was developed basing on the CEUS results, and then evaluated with receiver operating curve(ROC). RESULTS: Except in cases of enhanced homogeneity, the rest of the 9 enhancement appearances were statistically significant(P < 0.05). These 9 enhancement patterns were selected in the final step of the logistic regression analysis, with diagnostic sensitivity and specificity of 84.4% and 82.7%, respectively, and the area under the ROC curve of 0.911. Diagnostic sensitivity, specificity, and accuracy of the malignant vs benign CEUS models were 84.38%, 87.77%, 86.38% and 86.46%, 81.29% and 83.40%, respectively. CONCLUSION: The breast CEUS models can predict risk of malignant breast lesions more accurately, decrease false-positive biopsy, and provide accurate BIRADS classification.展开更多
A lump growing in the breast may be referred to as a breast mass related to the tumor.However,not all tumors are cancerous or malignant.Breast masses can cause discomfort and pain,depending on the size and texture of ...A lump growing in the breast may be referred to as a breast mass related to the tumor.However,not all tumors are cancerous or malignant.Breast masses can cause discomfort and pain,depending on the size and texture of the breast.With an appropriate diagnosis,non-cancerous breast masses can be diagnosed earlier to prevent their cultivation from being malignant.With the development of the artificial neural network,the deep discriminative model,such as a convolutional neural network,may evaluate the breast lesion to distinguish benign and malignant cancers frommammogram breast masses images.This work accomplished breastmasses classification relative to benign and malignant cancers using a digital database for screening mammography image datasets.A residual neural network 50(ResNet50)model along with an adaptive gradient algorithm,adaptive moment estimation,and stochastic gradient descent optimizers,as well as data augmentations and fine-tuning methods,were implemented.In addition,a learning rate scheduler and 5-fold cross-validation were applied with 60 training procedures to determine the best models.The results of training accuracy,p-value,test accuracy,area under the curve,sensitivity,precision,F1-score,specificity,and kappa for adaptive gradient algorithm 25%,75%,100%,and stochastic gradient descent 100%fine-tunings indicate that the classifier is feasible for categorizing breast cancer into benign and malignant from the mammographic breast masses images.展开更多
In this paper,a novel hybrid texture feature set and fractional derivative filter-based breast cancer detection model is introduced.This paper also introduces the application of a histogram of linear bipolar pattern f...In this paper,a novel hybrid texture feature set and fractional derivative filter-based breast cancer detection model is introduced.This paper also introduces the application of a histogram of linear bipolar pattern features(HLBP)for breast thermogram classification.Initially,breast tissues are separated by masking operation and filtered by Gr¨umwald–Letnikov fractional derivative-based Sobel mask to enhance the texture and rectify the noise.A novel hybrid feature set usingHLBP and other statistical feature sets is derived and reduced by principal component analysis.Radial basis function kernel-based support vector machine is employed for detecting the abnormality in the thermogram.The performance parameters are calculated using five-fold cross-validation scheme using MATLAB 2015a simulation software.The proposedmodel achieves the classification accuracy,sensitivity,specificity,and area under the curve of 94.44%,95.55%,92.22%,96.11%,respectively.A comparative investigation of different texture features with respect to fractional orderαto classify the breast malignancy is also presented.The proposed model is also compared with a few existing state-of-art schemes which verifies the efficacy of the model.Fractional orderαoffers extra adaptability in overcoming the limitations of thermal imaging techniques and assists radiologists in prior breast cancer detection.The proposed model is more generalized which can be used with different thermal image acquisition protocols and IoT based applications.展开更多
Earlier recognition of breast cancer is crucial to decrease the severity and optimize the survival rate.One of the commonly utilized imaging modalities for breast cancer is histopathological images.Since manual inspec...Earlier recognition of breast cancer is crucial to decrease the severity and optimize the survival rate.One of the commonly utilized imaging modalities for breast cancer is histopathological images.Since manual inspection of histopathological images is a challenging task,automated tools using deep learning(DL)and artificial intelligence(AI)approaches need to be designed.The latest advances of DL models help in accomplishing maximum image classification performance in several application areas.In this view,this study develops a Deep Transfer Learning with Rider Optimization Algorithm for Histopathological Classification of Breast Cancer(DTLRO-HCBC)technique.The proposed DTLRO-HCBC technique aims to categorize the existence of breast cancer using histopathological images.To accomplish this,the DTLRO-HCBC technique undergoes pre-processing and data augmentation to increase quantitative analysis.Then,optimal SqueezeNet model is employed for feature extractor and the hyperparameter tuning process is carried out using the Adadelta optimizer.Finally,rider optimization with deep feed forward neural network(RO-DFFNN)technique was utilized employed for breast cancer classification.The RO algorithm is applied for optimally adjusting the weight and bias values of the DFFNN technique.For demonstrating the greater performance of the DTLRO-HCBC approach,a sequence of simulations were carried out and the outcomes reported its promising performance over the current state of art approaches.展开更多
Breast cancer recognition is an important issue in elastography diagnostic imaging. Breast tumor biopsy has been for many years the reference procedure to assess histological definition for breast diseases. But biopsy...Breast cancer recognition is an important issue in elastography diagnostic imaging. Breast tumor biopsy has been for many years the reference procedure to assess histological definition for breast diseases. But biopsy measurement is an invasive method besides it takes larger time. So, fast and improved methods are needed. Using elastography technology, a digital image correlation technique can be used to calculate the displacement of breast tissue after it has suffered a compression force. This displacement is related to tissue stiffness, and breast cancer can be classified into benign or malignant according to that displacement. The value of compression force affects the displacement of tissue, and then affects the results of the breast cancer recognition. Finite element method was being used to simulate a model for the breast cancer as a phantom to be used in measurements and study of breast cancer diagnosis. The breast cancer using this phantom can be recognized within a short time. The proposed work succeeded in recognizing breast tumor phantom by an average correct recognition ratio CRR of about 94.25% on a simulation environment. The strain ratio SR for benign and malignant models is also computed. The result of the simulated breast tumor model is compared with real data of 10 lesion cases (6 benign and 4 malignant). The coefficient of variation CV between the simulated SR and the SR using real data reaches to about 5% for benign lesions and 4.78% for malignant lesions. The results of CRR and CV in this proposed work assure that the proposed breast cancer model using finite element modeling is a robust technique for breast tumor simulation where the behavior of real data of breast cancer can be predicted.展开更多
Breast cancer represents the most common malignancy in women,being one of the most frequent cause of cancer-related mortality.Ultrasound,mammography,and magnetic resonance imaging(MRI)play a pivotal role in the diagno...Breast cancer represents the most common malignancy in women,being one of the most frequent cause of cancer-related mortality.Ultrasound,mammography,and magnetic resonance imaging(MRI)play a pivotal role in the diagnosis of breast lesions,with different levels of accuracy.Particularly,dynamic contrastenhanced MRI has shown high diagnostic value in detecting multifocal,multicentric,or contralateral breast cancers.Radiomics is emerging as a promising tool for quantitative tumor evaluation,allowing the extraction of additional quantitative data from radiological imaging acquired with different modalities.Radiomics analysis may provide novel information through the quantification of lesions heterogeneity,that may be relevant in clinical practice for the characterization of breast lesions,prediction of tumor response to systemic therapies and evaluation of prognosis in patients with breast cancers.Several published studies have explored the value of radiomics with good-to-excellent diagnostic and prognostic performances for the evaluation of breast lesions.Particularly,the integrations of radiomics data with other clinical and histopathological parameters have demonstrated to improve the prediction of tumor aggressiveness with high accuracy and provided precise models that will help to guide clinical decisions and patients management.The purpose of this article in to describe the current application of radiomics in breast dynamic contrast-enhanced MRI.展开更多
To study quantitative index of bci-2, P53, Nroliferating cell nuclear antigen (PCNA),ER and PR in breast carcinoma and their correiation and their relatiousbip with prognosis, the ex expression of bcl-2, P53 and PCNA ...To study quantitative index of bci-2, P53, Nroliferating cell nuclear antigen (PCNA),ER and PR in breast carcinoma and their correiation and their relatiousbip with prognosis, the ex expression of bcl-2, P53 and PCNA were studied by immunohistochemical technique. The measurementof ER and PR used enzyme linked affinuity histochemical methods. The quantitative index was analyzed by image technique. All analyses were hased on 60 breast carcinomas. The results were as follows:the more bcl-2 protein, the lower histological graded the longer survival term and the highersurvival rate (P< 0. 05). The quautitative measurement of bcl-2, P53 and PCNA expression were ofvalue in evaluating the degree of differentiation and prognosis in breast carcinoma. The quantitativeand qualitative measurement or p53 protein expression showed a Ⅰwerful evidence in evaluatingprognosis of bcl-2 were more significant in evaluating poor prognosis of breast carcinoma. A relationship between bcl-2 and ER, PR showed a better value for response to endocrine therapy in breastcarcinoma patients.展开更多
文摘The Ki67 index (KI) is a standard clinical marker for tumor proliferation;however, its application is hindered by intratumoral heterogeneity. In this study, we used digital image analysis to comprehensively analyze Ki67 heterogeneity and distribution patterns in breast carcinoma. Using Smart Pathology software, we digitized and analyzed 42 excised breast carcinoma Ki67 slides. Boxplots, histograms, and heat maps were generated to illustrate the KI distribution. We found that 30% of cases (13/42) exhibited discrepancies between global and hotspot KI when using a 14% KI threshold for classification. Patients with higher global or hotspot KI values displayed greater heterogenicity. Ki67 distribution patterns were categorized as randomly distributed (52%, 22/42), peripheral (43%, 18/42), and centered (5%, 2/42). Our sampling simulator indicated analyzing more than 10 high-power fields was typically required to accurately estimate global KI, with sampling size being correlated with heterogeneity. In conclusion, using digital image analysis in whole-slide images allows for comprehensive Ki67 profile assessment, shedding light on heterogeneity and distribution patterns. This spatial information can facilitate KI surveys of breast cancer and other malignancies.
文摘AIM:To automate breast cancer diagnosis and to study the inter-observer and intra-observer variations in the manual evaluations.METHODS:Breast tissue specimens from sixty cases were stained separately for estrogen receptor(ER),progesterone receptor(PR)and human epidermal growth factor receptor-2(HER-2/neu).All cases were assessed by manual grading as well as image analysis.The manual grading was performed by an experienced expert pathologist.To study inter-observer and intra-observer variations,we obtained readings from another pathologist as the second observer from a different laboratory who has a little less experience than the first observer.We also took a second reading from the second observer to study intra-observer variations.Image analysis was carried out using in-house developed software(TissueQuant).A comparison of the results from image analysis and manual scoring of ER,PR and HER-2/neu was also carried out.RESULTS:The performance of the automated analysis in the case of ER,PR and HER-2/neu expressions was compared with the manual evaluations.The performance of the automated system was found to correlate well with the manual evaluations.The inter-observer variations were measured using Spearman correlation coefficient r and 95%confidence interval.In the case of ER expression,Spearman correlation r=0.53,in the case of PR expression,r=0.63,and in the case of HER-2/neu expression,r=0.68.Similarly,intra-observer variations were also measured.In the case of ER,PR and HER-2/neu expressions,r=0.46,0.66 and 0.70,respectively.CONCLUSION:The automation of breast cancer diagnosis from immunohistochemically stained specimens is very useful for providing objective and repeatable evaluations.
文摘Objective To investigate the difference in texture features on diffusion weighted imaging(DWI) images between breast benign and malignant tumors.Methods Patients including 56 with mass-like breast cancer, 16 with breast fibroadenoma, and 4 with intraductal papilloma of breast treated in the Hainan Hospital of Chinese PLA General Hospital were retrospectively enrolled in this study, and allocated to the benign group(20 patients) and the malignant group(56 patients) according to the post-surgically pathological results. Texture analysis was performed on axial DWI images, and five characteristic parameters including Angular Second Moment(ASM), Contrast, Correlation, Inverse Difference Moment(IDM), and Entropy were calculated. Independent sample t-test and Mann-Whitney U test were performed for intergroup comparison. Regression model was established by using Binary Logistic regression analysis, and receiver operating characteristic curve(ROC) analysis was carried out to evaluate the diagnostic efficiency. Results The texture features ASM, Contrast, Correlation and Entropy showed significant differences between the benign and malignant breast tumor groups(PASM= 0.014, Pcontrast= 0.019, Pcorrelation= 0.010, Pentropy= 0.007). The area under the ROC curve was 0.685, 0.681, 0.754, and 0.683 respectively for the positive texture variables mentioned above, and that for the combined variables(ASM, Contrast, and Entropy) was 0.802 in the model of Logistic regression. Binary Logistic regression analysis demonstrated that ASM, Contrast and Entropy were considered as thespecific imaging variables for the differential diagnosis of breast benign and malignant tumors.Conclusion The texture analysis of DWI may be a simple and effective tool in the differential diagnosis between breast benign and malignant tumors.
文摘BACKGROUND A positive resection margin is a major risk factor for local breast cancer recurrence after breast-conserving surgery(BCS).Preoperative imaging examinations are frequently employed to assess the surgical margin.AIM To investigate the role and value of preoperative imaging examinations[magnetic resonance imaging(MRI),molybdenum target,and ultrasound]in evaluating margins for BCS.METHODS A retrospective study was conducted on 323 breast cancer patients who met the criteria for BCS and consented to the procedure from January 2014 to July 2021.The study gathered preoperative imaging data(MRI,ultrasound,and molybdenum target examination)and intraoperative and postoperative pathological information.Based on their BCS outcomes,patients were categorized into positive and negative margin groups.Subsequently,the patients were randomly split into a training set(226 patients,approximately 70%)and a validation set(97 patients,approximately 30%).The imaging and pathological information was analyzed and summarized using R software.Non-conditional logistic regression and LASSO regression were conducted in the validation set to identify factors that might influence the failure of BCS.A column chart was generated and applied to the validation set to examine the relationship between pathological margin range and prognosis.This study aims to identify the risk factors associated with failure in BCS.RESULTS The multivariate non-conditional logistic regression analysis demonstrated that various factors raise the risk of positive margins following BCS.These factors comprise non-mass enhancement(NME)on dynamic contrastenhanced MRI,multiple focal vascular signs around the lesion on MRI,tumor size exceeding 2 cm,type III timesignal intensity curve,indistinct margins on molybdenum target examination,unclear margins on ultrasound examination,and estrogen receptor(ER)positivity in immunohistochemistry.LASSO regression was additionally employed in this study to identify four predictive factors for the model:ER,molybdenum target tumor type(MT Xmd Shape),maximum intensity projection imaging feature,and lesion type on MRI.The model constructed with these predictive factors exhibited strong consistency with the real-world scenario in both the training set and validation set.Particularly,the outcomes of the column chart model accurately predicted the likelihood of positive margins in BCS.CONCLUSION The proposed column chart model effectively predicts the success of BCS for breast cancer.The model utilizes preoperative ultrasound,molybdenum target,MRI,and core needle biopsy pathology evaluation results,all of which align with the real-world scenario.Hence,our model can offer dependable guidance for clinical decisionmaking concerning BCS.
文摘AIM: To build and evaluate predictive models for contrast-enhanced ultrasound(CEUS) of the breast to distinguish between benign and malignant lesions. METHODS: A total of 235 breast imaging reporting and data system(BI-RADS) 4 solid breast lesions were imaged via CEUS before core needle biopsy or surgical resection. CEUS results were analyzed on 10 enhancing patterns to evaluate diagnostic performance of three benign and three malignant CEUS models, with pathological results used as the gold standard. A logistic regression model was developed basing on the CEUS results, and then evaluated with receiver operating curve(ROC). RESULTS: Except in cases of enhanced homogeneity, the rest of the 9 enhancement appearances were statistically significant(P < 0.05). These 9 enhancement patterns were selected in the final step of the logistic regression analysis, with diagnostic sensitivity and specificity of 84.4% and 82.7%, respectively, and the area under the ROC curve of 0.911. Diagnostic sensitivity, specificity, and accuracy of the malignant vs benign CEUS models were 84.38%, 87.77%, 86.38% and 86.46%, 81.29% and 83.40%, respectively. CONCLUSION: The breast CEUS models can predict risk of malignant breast lesions more accurately, decrease false-positive biopsy, and provide accurate BIRADS classification.
基金This research was supported by the National Research Foundation of Korea(NRF)grant funded by the Korean government(MSIT)[NRF-2019R1F1A1062397,NRF-2021R1F1A1059665]Brain Korea 21 FOUR Project(Dept.of IT Convergence Engineering,Kumoh National Institute of Technology)This paper was supported by Korea Institute for Advancement of Technology(KIAT)grant funded by the Korea Government(MOTIE)[P0017123,The Competency Development Program for Industry Specialist].
文摘A lump growing in the breast may be referred to as a breast mass related to the tumor.However,not all tumors are cancerous or malignant.Breast masses can cause discomfort and pain,depending on the size and texture of the breast.With an appropriate diagnosis,non-cancerous breast masses can be diagnosed earlier to prevent their cultivation from being malignant.With the development of the artificial neural network,the deep discriminative model,such as a convolutional neural network,may evaluate the breast lesion to distinguish benign and malignant cancers frommammogram breast masses images.This work accomplished breastmasses classification relative to benign and malignant cancers using a digital database for screening mammography image datasets.A residual neural network 50(ResNet50)model along with an adaptive gradient algorithm,adaptive moment estimation,and stochastic gradient descent optimizers,as well as data augmentations and fine-tuning methods,were implemented.In addition,a learning rate scheduler and 5-fold cross-validation were applied with 60 training procedures to determine the best models.The results of training accuracy,p-value,test accuracy,area under the curve,sensitivity,precision,F1-score,specificity,and kappa for adaptive gradient algorithm 25%,75%,100%,and stochastic gradient descent 100%fine-tunings indicate that the classifier is feasible for categorizing breast cancer into benign and malignant from the mammographic breast masses images.
基金Praveen Agarwal,thanks to the SERB(Project TAR/2018/000001)DST(Projects DST/INT/DAAD/P-21/2019 and INT/RUS/RFBR/308)NBHM(DAE)(Project 02011/12/2020 NBHM(R.P)/RD II/7867).
文摘In this paper,a novel hybrid texture feature set and fractional derivative filter-based breast cancer detection model is introduced.This paper also introduces the application of a histogram of linear bipolar pattern features(HLBP)for breast thermogram classification.Initially,breast tissues are separated by masking operation and filtered by Gr¨umwald–Letnikov fractional derivative-based Sobel mask to enhance the texture and rectify the noise.A novel hybrid feature set usingHLBP and other statistical feature sets is derived and reduced by principal component analysis.Radial basis function kernel-based support vector machine is employed for detecting the abnormality in the thermogram.The performance parameters are calculated using five-fold cross-validation scheme using MATLAB 2015a simulation software.The proposedmodel achieves the classification accuracy,sensitivity,specificity,and area under the curve of 94.44%,95.55%,92.22%,96.11%,respectively.A comparative investigation of different texture features with respect to fractional orderαto classify the breast malignancy is also presented.The proposed model is also compared with a few existing state-of-art schemes which verifies the efficacy of the model.Fractional orderαoffers extra adaptability in overcoming the limitations of thermal imaging techniques and assists radiologists in prior breast cancer detection.The proposed model is more generalized which can be used with different thermal image acquisition protocols and IoT based applications.
基金This project was funded by the Deanship of Scientific Research(DSR),King Abdulaziz University,Jeddah,under grant no.(D-773-130-1443).
文摘Earlier recognition of breast cancer is crucial to decrease the severity and optimize the survival rate.One of the commonly utilized imaging modalities for breast cancer is histopathological images.Since manual inspection of histopathological images is a challenging task,automated tools using deep learning(DL)and artificial intelligence(AI)approaches need to be designed.The latest advances of DL models help in accomplishing maximum image classification performance in several application areas.In this view,this study develops a Deep Transfer Learning with Rider Optimization Algorithm for Histopathological Classification of Breast Cancer(DTLRO-HCBC)technique.The proposed DTLRO-HCBC technique aims to categorize the existence of breast cancer using histopathological images.To accomplish this,the DTLRO-HCBC technique undergoes pre-processing and data augmentation to increase quantitative analysis.Then,optimal SqueezeNet model is employed for feature extractor and the hyperparameter tuning process is carried out using the Adadelta optimizer.Finally,rider optimization with deep feed forward neural network(RO-DFFNN)technique was utilized employed for breast cancer classification.The RO algorithm is applied for optimally adjusting the weight and bias values of the DFFNN technique.For demonstrating the greater performance of the DTLRO-HCBC approach,a sequence of simulations were carried out and the outcomes reported its promising performance over the current state of art approaches.
文摘Breast cancer recognition is an important issue in elastography diagnostic imaging. Breast tumor biopsy has been for many years the reference procedure to assess histological definition for breast diseases. But biopsy measurement is an invasive method besides it takes larger time. So, fast and improved methods are needed. Using elastography technology, a digital image correlation technique can be used to calculate the displacement of breast tissue after it has suffered a compression force. This displacement is related to tissue stiffness, and breast cancer can be classified into benign or malignant according to that displacement. The value of compression force affects the displacement of tissue, and then affects the results of the breast cancer recognition. Finite element method was being used to simulate a model for the breast cancer as a phantom to be used in measurements and study of breast cancer diagnosis. The breast cancer using this phantom can be recognized within a short time. The proposed work succeeded in recognizing breast tumor phantom by an average correct recognition ratio CRR of about 94.25% on a simulation environment. The strain ratio SR for benign and malignant models is also computed. The result of the simulated breast tumor model is compared with real data of 10 lesion cases (6 benign and 4 malignant). The coefficient of variation CV between the simulated SR and the SR using real data reaches to about 5% for benign lesions and 4.78% for malignant lesions. The results of CRR and CV in this proposed work assure that the proposed breast cancer model using finite element modeling is a robust technique for breast tumor simulation where the behavior of real data of breast cancer can be predicted.
文摘Breast cancer represents the most common malignancy in women,being one of the most frequent cause of cancer-related mortality.Ultrasound,mammography,and magnetic resonance imaging(MRI)play a pivotal role in the diagnosis of breast lesions,with different levels of accuracy.Particularly,dynamic contrastenhanced MRI has shown high diagnostic value in detecting multifocal,multicentric,or contralateral breast cancers.Radiomics is emerging as a promising tool for quantitative tumor evaluation,allowing the extraction of additional quantitative data from radiological imaging acquired with different modalities.Radiomics analysis may provide novel information through the quantification of lesions heterogeneity,that may be relevant in clinical practice for the characterization of breast lesions,prediction of tumor response to systemic therapies and evaluation of prognosis in patients with breast cancers.Several published studies have explored the value of radiomics with good-to-excellent diagnostic and prognostic performances for the evaluation of breast lesions.Particularly,the integrations of radiomics data with other clinical and histopathological parameters have demonstrated to improve the prediction of tumor aggressiveness with high accuracy and provided precise models that will help to guide clinical decisions and patients management.The purpose of this article in to describe the current application of radiomics in breast dynamic contrast-enhanced MRI.
文摘To study quantitative index of bci-2, P53, Nroliferating cell nuclear antigen (PCNA),ER and PR in breast carcinoma and their correiation and their relatiousbip with prognosis, the ex expression of bcl-2, P53 and PCNA were studied by immunohistochemical technique. The measurementof ER and PR used enzyme linked affinuity histochemical methods. The quantitative index was analyzed by image technique. All analyses were hased on 60 breast carcinomas. The results were as follows:the more bcl-2 protein, the lower histological graded the longer survival term and the highersurvival rate (P< 0. 05). The quautitative measurement of bcl-2, P53 and PCNA expression were ofvalue in evaluating the degree of differentiation and prognosis in breast carcinoma. The quantitativeand qualitative measurement or p53 protein expression showed a Ⅰwerful evidence in evaluatingprognosis of bcl-2 were more significant in evaluating poor prognosis of breast carcinoma. A relationship between bcl-2 and ER, PR showed a better value for response to endocrine therapy in breastcarcinoma patients.
