Automatic Respiratory Gating Hepatic DCEUS-based Dual-phase Multi-parametric Functional Perfusion Imaging using a Derivative Principal Component Analysis

Classification of multi-feature fusion ultrasound images of breast tumor within category 4 using convolutional neural networks

BACKGROUND

Breast tumor is a fatal threat to the health of women. Ultrasound (US) is a common and economical method for the diagnosis of breast cancer. Breast imaging reporting and data system (BI-RADS) category 4 has the highest false-positive value of about 30% among five categories. The classification task in BI-RADS category 4 is challenging and has not been fully studied.

PURPOSE

This work aimed to use convolutional neural networks (CNNs) for breast tumor classification using B-mode images in category 4 to overcome the dependence on operator and artifacts. Additionally, this work intends to take full advantage of morphological and textural features in breast tumor US images to improve classification accuracy.

METHODS

First, original US images coming directly from the hospital were cropped and resized. In 1385 B-mode US BI-RADS category 4 images, the biopsy eliminated 503 samples of benign tumor and left 882 of malignant. Then, K-means clustering algorithm and entropy of sliding windows of US images were conducted. Considering the diversity of different characteristic information of malignant and benign represented by original B-mode images, K-means clustering images and entropy images, they are fused in a three-channel form multi-feature fusion images dataset. The training, validation, and test sets are 969, 277, and 139. With transfer learning, 11 CNN models including DenseNet and ResNet were investigated. Finally, by comparing accuracy, precision, recall, F1-score, and area under curve (AUC) of the results, models which had better performance were selected. The normality of data was assessed by Shapiro-Wilk test. DeLong test and independent t-test were used to evaluate the significant difference of AUC and other values. False discovery rate was utilized to ultimately evaluate the advantages of CNN with highest evaluation metrics. In addition, the study of anti-log compression was conducted but no improvement has shown in CNNs classification results.

RESULTS

With multi-feature fusion images, DenseNet121 has highest accuracy of 80.22 ± 1.45% compared to other CNNs, precision of 77.97 ± 2.89% and AUC of 0.82 ± 0.01. Multi-feature fusion improved accuracy of DenseNet121 by 1.87% from classification of original B-mode images (p < 0.05).

CONCLUSION

The CNNs with multi-feature fusion show a good potential of reducing the false-positive rate within category 4. The work illustrated that CNNs and fusion images have the potential to reduce false-positive rate in breast tumor within US BI-RADS category 4, and make the diagnosis of category 4 breast tumors to be more accurate and precise.

Very Low Frequency Radial Modulation for Deep Penetration Contrast-Enhanced Ultrasound Imaging

Contrast-enhanced ultrasound imaging allows vascular imaging in a variety of diseases. Radial modulation imaging is a contrast agent-specific imaging approach for improving microbubble detection at high imaging frequencies (≥7.5 MHz), with imaging depth limited to a few centimeters. To provide high-sensitivity contrast-enhanced ultrasound imaging at high penetration depths, a new radial modulation imaging strategy using a very low frequency (100 kHz) ultrasound modulation wave in combination with imaging pulses ≤5 MHz is proposed. Microbubbles driven at 100 kHz were imaged in 10 successive oscillation states by manipulating the pulse repetition frequency to unlock the frame rate from the number of oscillation states. Tissue background was suppressed using frequency domain radial modulation imaging (F-RMI) and singular value decomposition-based radial modulation imaging (S-RMI). One hundred-kilohertz modulation resulted in significantly higher microbubble signal magnitude (63-88 dB) at the modulation frequency relative to that without 100-kHz modulation (51-59 dB). F-RMI produced images with high contrast-to-tissue ratios (CTRs) of 15 to 22 dB in a stationary tissue phantom, while S-RMI further improved the CTR (19-26 dB). These CTR values were significantly higher than that of amplitude modulation pulse inversion images (11.9 dB). In the presence of tissue motion (1 and 10 mm/s), S-RMI produced high-contrast images with CTR up to 18 dB; however, F-RMI resulted in minimal contrast enhancement in the presence of tissue motion. Finally, in transcranial ultrasound imaging studies through a highly attenuating ex vivo cranial bone, CTR values with S-RMI were as high as 23 dB. The proposed technique demonstrates successful modulation of microbubble response at 100 kHz for the first time. The presented S-RMI low-frequency radial modulation imaging strategy represents the first demonstration of real-time (20 frames/s), high-penetration-depth radial modulation imaging for contrast-enhanced ultrasound imaging.