Dual-Intended Deep Learning Model for Breast Cancer Diagnosis in Ultrasound Imaging.
Ontology highlight
ABSTRACT:
SUBMITTER: Vigil N
PROVIDER: S-EPMC9179519 | biostudies-literature | 2022 May
REPOSITORIES: biostudies-literature
Publications
Dual-Intended Deep Learning Model for Breast Cancer Diagnosis in Ultrasound Imaging.
Vigil Nicolle N Barry Madeline M Amini Arya A Akhloufi Moulay M Maldague Xavier P V XPV Ma Lan L Ren Lei L Yousefi Bardia B
Cancers 20220527 11
Automated medical data analysis demonstrated a significant role in modern medicine, and cancer diagnosis/prognosis to achieve highly reliable and generalizable systems. In this study, an automated breast cancer screening method in ultrasound imaging is proposed. A convolutional deep autoencoder model is presented for simultaneous segmentation and radiomic extraction. The model segments the breast lesions while concurrently extracting radiomic features. With our deep model, we perform breast lesi ...[more]
Similar Datasets
Project description:Conventional deep learning (DL) algorithm requires full supervision of annotating the region of interest (ROI) that is laborious and often biased. We aimed to develop a weakly-supervised DL algorithm that diagnosis breast cancer at ultrasound without image annotation. Weakly-supervised DL algorithms were implemented with three networks (VGG16, ResNet34, and GoogLeNet) and trained using 1000 unannotated US images (500 benign and 500 malignant masses). Two sets of 200 images (100 benign and 100 malignant masses) were used for internal and external validation sets. For comparison with fully-supervised algorithms, ROI annotation was performed manually and automatically. Diagnostic performances were calculated as the area under the receiver operating characteristic curve (AUC). Using the class activation map, we determined how accurately the weakly-supervised DL algorithms localized the breast masses. For internal validation sets, the weakly-supervised DL algorithms achieved excellent diagnostic performances, with AUC values of 0.92-0.96, which were not statistically different (all Ps > 0.05) from those of fully-supervised DL algorithms with either manual or automated ROI annotation (AUC, 0.92-0.96). For external validation sets, the weakly-supervised DL algorithms achieved AUC values of 0.86-0.90, which were not statistically different (Ps > 0.05) or higher (P = 0.04, VGG16 with automated ROI annotation) from those of fully-supervised DL algorithms (AUC, 0.84-0.92). In internal and external validation sets, weakly-supervised algorithms could localize 100% of malignant masses, except for ResNet34 (98%). The weakly-supervised DL algorithms developed in the present study were feasible for US diagnosis of breast cancer with well-performing localization and differential diagnosis.
| S-EPMC8692405 | biostudies-literature
Project description:Background: Breast ultrasound is the first choice for breast tumor diagnosis in China, but the Breast Imaging Reporting and Data System (BI-RADS) categorization routinely used in the clinic often leads to unnecessary biopsy. Radiologists have no ability to predict molecular subtypes with important pathological information that can guide clinical treatment. Materials and Methods: This retrospective study collected breast ultrasound images from two hospitals and formed training, test and external test sets after strict selection, which included 2,822, 707, and 210 ultrasound images, respectively. An optimized deep learning model (DLM) was constructed with the training set, and the performance was verified in both the test set and the external test set. Diagnostic results were compared with the BI-RADS categorization determined by radiologists. We divided breast cancer into different molecular subtypes according to hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) expression. The ability to predict molecular subtypes using the DLM was confirmed in the test set. Results: In the test set, with pathological results as the gold standard, the accuracy, sensitivity and specificity were 85.6, 98.7, and 63.1%, respectively, according to the BI-RADS categorization. The same set achieved an accuracy, sensitivity, and specificity of 89.7, 91.3, and 86.9%, respectively, when using the DLM. For the test set, the area under the curve (AUC) was 0.96. For the external test set, the AUC was 0.90. The diagnostic accuracy was 92.86% with the DLM in BI-RADS 4a patients. Approximately 70.76% of the cases were judged as benign tumors. Unnecessary biopsy was theoretically reduced by 67.86%. However, the false negative rate was 10.4%. A good prediction effect was shown for the molecular subtypes of breast cancer with the DLM. The AUC were 0.864, 0.811, and 0.837 for the triple-negative subtype, HER2 (+) subtype and HR (+) subtype predictions, respectively. Conclusion: This study showed that the DLM was highly accurate in recognizing breast tumors from ultrasound images. Thus, the DLM can greatly reduce the incidence of unnecessary biopsy, especially for patients with BI-RADS 4a. In addition, the predictive ability of this model for molecular subtypes was satisfactory,which has specific clinical application value.
| S-EPMC7973262 | biostudies-literature
Project description:Deep vein thrombosis (DVT) causes significant healthcare burdens worldwide. This study aims to establish a deep learning model for the diagnosis of DVT from the assessment of vein compressibility. Considering the complexity of ultrasound images, convolutional neural networks with UNet and residual neural network (ResNet) are established for image segmentation, from venous duplex ultrasonographic video images, obtained through standard and portable handheld ultrasound methods. To further evaluate the similarity between the predicted and ground truth images, the structural similarity index (SSIM) is employed. Our deep learning model achieves over 90% accuracy, providing an innovative tool for both images and videos. This study harnesses the power of machine learning to develop an automatic labeling tool that can diagnose DVT by analyzing ultrasonography images. To make the tool more accessible to front-line clinicians, a user-friendly application is created to quickly assess possible clinical severity and enable prompt medical intervention, reducing disease progression.
| S-EPMC11647138 | biostudies-literature
Project description:Deep learning (DL) has been widely investigated in breast ultrasound (US) for distinguishing between benign and malignant breast masses. This systematic review of test diagnosis aims to examine the accuracy of DL, compared to human readers, for the diagnosis of breast cancer in the US under clinical settings. Our literature search included records from databases including PubMed, Embase, Scopus, and Cochrane Library. Test accuracy outcomes were synthesized to compare the diagnostic performance of DL and human readers as well as to evaluate the assistive role of DL to human readers. A total of 16 studies involving 9238 female participants were included. There were no prospective studies comparing the test accuracy of DL versus human readers in clinical workflows. Diagnostic test results varied across the included studies. In 14 studies employing standalone DL systems, DL showed significantly lower sensitivities in 5 studies with comparable specificities and outperformed human readers at higher specificities in another 4 studies; in the remaining studies, DL models and human readers showed equivalent test outcomes. In 12 studies that assessed assistive DL systems, no studies proved the assistive role of DL in the overall diagnostic performance of human readers. Current evidence is insufficient to conclude that DL outperforms human readers or enhances the accuracy of diagnostic breast US in a clinical setting. Standardization of study methodologies is required to improve the reproducibility and generalizability of DL research, which will aid in clinical translation and application.
| S-EPMC10821881 | biostudies-literature
Project description:SignificanceUltrasound (US)-guided diffuse optical tomography (DOT) has demonstrated great potential for breast cancer diagnosis in which real-time or near real-time diagnosis with high accuracy is desired.AimWe aim to use US-guided DOT to achieve an automated, fast, and accurate classification of breast lesions.ApproachWe propose a two-stage classification strategy with deep learning. In the first stage, US images and histograms created from DOT perturbation measurements are combined to predict benign lesions. Then the non-benign suspicious lesions are passed through to the second stage, which combine US image features, DOT histogram features, and 3D DOT reconstructed images for final diagnosis.ResultsThe first stage alone identified 73.0% of benign cases without image reconstruction. In distinguishing between benign and malignant breast lesions in patient data, the two-stage classification approach achieved an area under the receiver operating characteristic curve of 0.946, outperforming the diagnoses of all single-modality models and of a single-stage classification model that combines all US images, DOT histogram, and imaging features.ConclusionsThe proposed two-stage classification strategy achieves better classification accuracy than single-modality-only models and a single-stage classification model that combines all features. It can potentially distinguish breast cancers from benign lesions in near real-time.
| S-EPMC10457211 | biostudies-literature
Project description:BackgroundAccurate diagnosis of unexplained cervical lymphadenopathy (CLA) using medical images heavily relies on the experience of radiologists, which is even worse for CLA patients in underdeveloped countries and regions, because of lack of expertise and reliable medical history. This study aimed to develop a deep learning (DL) radiomics model based on B-mode and color Doppler ultrasound images for assisting radiologists to improve their diagnoses of the etiology of unexplained CLA.MethodsPatients with unexplained CLA who received ultrasound examinations from three hospitals located in underdeveloped areas of China were retrospectively enrolled. They were all pathologically confirmed with reactive hyperplasia, tuberculous lymphadenitis, lymphoma, or metastatic carcinoma. By mimicking the diagnosis logic of radiologists, three DL sub-models were developed to achieve the primary diagnosis of benign and malignant, the secondary diagnosis of reactive hyperplasia and tuberculous lymphadenitis in benign candidates, and of lymphoma and metastatic carcinoma in malignant candidates, respectively. Then, a CLA hierarchical diagnostic model (CLA-HDM) integrating all sub-models was proposed to classify the specific etiology of each unexplained CLA. The assistant effectiveness of CLA-HDM was assessed by comparing six radiologists between without and with using the DL-based classification and heatmap guidance.ResultsA total of 763 patients with unexplained CLA were enrolled and were split into the training cohort (n=395), internal testing cohort (n=171), and external testing cohorts 1 (n=105) and 2 (n=92). The CLA-HDM for diagnosing four common etiologies of unexplained CLA achieved AUCs of 0.873 (95% CI: 0.838-0.908), 0.837 (95% CI: 0.789-0.889), and 0.840 (95% CI: 0.789-0.898) in the three testing cohorts, respectively, which was systematically more accurate than all the participating radiologists. With its assistance, the accuracy, sensitivity, and specificity of six radiologists with different levels of experience were generally improved, reducing the false-negative rate of 2.2-10% and the false-positive rate of 0.7-3.1%.ConclusionsMulti-cohort testing demonstrated our DL model integrating dual-modality ultrasound images achieved accurate diagnosis of unexplained CLA. With its assistance, the gap between radiologists with different levels of experience was narrowed, which is potentially of great significance for benefiting CLA patients in underdeveloped countries and regions worldwide.
| S-EPMC9410737 | biostudies-literature
Project description:Color Doppler is used in the clinic for visually assessing the vascularity of breast masses on ultrasound, to aid in determining the likelihood of malignancy. In this study, quantitative color Doppler radiomics features were algorithmically extracted from breast sonograms for machine learning, producing a diagnostic model for breast cancer with higher performance than models based on grayscale and clinical category from the Breast Imaging Reporting and Data System for ultrasound (BI-RADSUS). Ultrasound images of 159 solid masses were analyzed. Algorithms extracted nine grayscale features and two color Doppler features. These features, along with patient age and BI-RADSUS category, were used to train an AdaBoost ensemble classifier. Though training on computer-extracted grayscale features and color Doppler features each significantly increased performance over that of models trained on clinical features, as measured by the area under the receiver operating characteristic (ROC) curve, training on both color Doppler and grayscale further increased the ROC area, from 0.925 ± 0.022 to 0.958 ± 0.013. Pruning low-confidence cases at 20% improved this to 0.986 ± 0.007 with 100% sensitivity, whereas 64% of the cases had to be pruned to reach this performance without color Doppler. Fewer borderline diagnoses and higher ROC performance were both achieved for diagnostic models of breast cancer on ultrasound by machine learning on color Doppler features.
| S-EPMC7555557 | biostudies-literature
Project description:This study aims to restore grating lobe artifacts and improve the image resolution of sparse array ultrasonography via a deep learning predictive model. A deep learning assisted sparse array was developed using only 64 or 16 channels out of the 128 channels in which the pitch is two or eight times the original array. The deep learning assisted sparse array imaging system was demonstrated on ex vivo porcine teeth. 64- and 16-channel sparse array images were used as the input and corresponding 128-channel dense array images were used as the ground truth. The structural similarity index measure, mean squared error, and peak signal-to-noise ratio of predicted images improved significantly (p < 0.0001). The resolution of predicted images presented close values to ground truth images (0.18 mm and 0.15 mm versus 0.15 mm). The gingival thickness measurement showed a high level of agreement between the predicted sparse array images and the ground truth images, as indicated with a bias of -0.01 mm and 0.02 mm for the 64- and 16-channel predicted images, respectively, and a Pearson's r = 0.99 (p < 0.0001) for both. The gingival thickness bias measured by deep learning assisted sparse array imaging and clinical probing needle was found to be <0.05 mm. Additionally, the deep learning model showed capability of generalization. To conclude, the deep learning assisted sparse array can reconstruct high-resolution ultrasound image using only 16 channels of 128 channels. The deep learning model performed generalization capability for the 64-channel array, while the 16-channel array generalization would require further optimization.
| S-EPMC10615290 | biostudies-literature
Project description:This study aimed to assess the diagnostic performance of deep convolutional neural networks (DCNNs) in classifying breast microcalcification in screening mammograms. To this end, 1579 mammographic images were collected retrospectively from patients exhibiting suspicious microcalcification in screening mammograms between July 2007 and December 2019. Five pre-trained DCNN models and an ensemble model were used to classify the microcalcifications as either malignant or benign. Approximately one million images from the ImageNet database had been used to train the five DCNN models. Herein, 1121 mammographic images were used for individual model fine-tuning, 198 for validation, and 260 for testing. Gradient-weighted class activation mapping (Grad-CAM) was used to confirm the validity of the DCNN models in highlighting the microcalcification regions most critical for determining the final class. The ensemble model yielded the best AUC (0.856). The DenseNet-201 model achieved the best sensitivity (82.47%) and negative predictive value (NPV; 86.92%). The ResNet-101 model yielded the best accuracy (81.54%), specificity (91.41%), and positive predictive value (PPV; 81.82%). The high PPV and specificity achieved by the ResNet-101 model, in particular, demonstrated the model effectiveness in microcalcification diagnosis, which, in turn, may considerably help reduce unnecessary biopsies.
| S-EPMC8671560 | biostudies-literature
Project description:ObjectiveEchocardiography (ECG) is the most common method used to diagnose heart failure (HF). However, its accuracy relies on the experience of the operator. Additionally, the video format of the data makes it challenging for patients to bring them to referrals and reexaminations. Therefore, this study used a deep learning approach to assist physicians in assessing cardiac function to promote the standardization of echocardiographic findings and compatibility of dynamic and static ultrasound data.MethodsA deep spatio-temporal convolutional model r2plus1d-Pan (trained on dynamic data and applied to static data) was improved and trained using the idea of "regression training combined with classification application," which can be generalized to dynamic ECG and static cardiac ultrasound views to identify HF with a reduced ejection fraction (EF < 40%). Additionally, three independent datasets containing 8976 cardiac ultrasound views and 10085 cardiac ultrasound videos were established. Subsequently, a multinational, multi-center dataset of EF was labeled. Furthermore, model training and independent validation were performed. Finally, 15 registered ultrasonographers and cardiologists with different working years in three regional hospitals specialized in cardiovascular disease were recruited to compare the results.ResultsThe proposed deep spatio-temporal convolutional model achieved an area under the receiveroperating characteristic curve (AUC) value of 0.95 (95% confidence interval [CI]: 0.947 to 0.953) on the training set of dynamic ultrasound data and an AUC of 1 (95% CI, 1 to 1) on the independent validation set. Subsequently, the model was applied to the static cardiac ultrasound view (validation set) with simultaneous input of 1, 2, 4, and 8 images of the same heart, with classification accuracies of 85%, 81%, 93%, and 92%, respectively. On the static data, the classification accuracy of the artificial intelligence (AI) model was comparable with the best performance of ultrasonographers and cardiologists with more than 3 working years (P = 0.344), but significantly better than the median level (P = 0.0000008).ConclusionA new deep spatio-temporal convolution model was constructed to identify patients with HF with reduced EF accurately (< 40%) using dynamic and static cardiac ultrasound images. The model outperformed the diagnostic performance of most senior specialists. This may be the first HF-related AI diagnostic model compatible with multi-dimensional cardiac ultrasound data, and may thereby contribute to the improvement of HF diagnosis. Additionally, the model enables patients to carry "on-the-go" static ultrasound reports for referral and reexamination, thus saving healthcare resources.
| S-EPMC10680380 | biostudies-literature