Ultrasonic Assessment of Liver Fibrosis Using One-Dimensional Convolutional Neural Networks Based on Frequency Spectra of Radiofrequency Signals with Deep Learning Segmentation of Liver Regions in B-Mode Images: A Feasibility Study.

The early detection of liver fibrosis is of significant importance. Deep learning analysis of ultrasound backscattered radiofrequency (RF) signals is emerging for tissue characterization as the RF signals carry abundant information related to tissue microstructures. However, the existing methods only used the time-domain information of the RF signals for liver fibrosis assessment, and the liver region of interest (ROI) is outlined manually. In this study, we proposed an approach for liver fibrosis assessment using deep learning models on ultrasound RF signals. The proposed method consisted of two-dimensional (2D) convolutional neural networks (CNNs) for automatic liver ROI segmentation from reconstructed B-mode ultrasound images and one-dimensional (1D) CNNs for liver fibrosis stage classification based on the frequency spectra (amplitude, phase, and power) of the segmented ROI signals. The Fourier transform was used to obtain the three kinds of frequency spectra. Two classical 2D CNNs were employed for liver ROI segmentation: U-Net and Attention U-Net. ROI spectrum signals were normalized and augmented using a sliding window technique. Ultrasound RF signals collected (with a 3-MHz transducer) from 613 participants (Group A) were included for liver ROI segmentation and those from 237 participants (Group B) for liver fibrosis stage classification, with a liver biopsy as the reference standard (Fibrosis stage: F0 = 27, F1 = 49, F2 = 51, F3 = 49, F4 = 61). In the test set of Group A, U-Net and Attention U-Net yielded Dice similarity coefficients of 95.05% and 94.68%, respectively. In the test set of Group B, the 1D CNN performed the best when using ROI phase spectrum signals to evaluate liver fibrosis stages ≥F1 (area under the receive operating characteristic curve, AUC: 0.957; accuracy: 89.19%; sensitivity: 85.17%; specificity: 93.75%), ≥F2 (AUC: 0.808; accuracy: 83.34%; sensitivity: 87.50%; specificity: 78.57%), and ≥F4 (AUC: 0.876; accuracy: 85.71%; sensitivity: 77.78%; specificity: 94.12%), and when using the power spectrum signals to evaluate ≥F3 (AUC: 0.729; accuracy: 77.14%; sensitivity: 77.27%; specificity: 76.92%). The experimental results demonstrated the feasibility of both the 2D and 1D CNNs in liver parenchyma detection and liver fibrosis characterization. The proposed methods have provided a new strategy for liver fibrosis assessment based on ultrasound RF signals, especially for early fibrosis detection. The findings of this study shed light on deep learning analysis of ultrasound RF signals in the frequency domain with automatic ROI segmentation.

Abnormal white matter structural network topological property in patients with temporal lobe epilepsy.

BACKGROUND: Diffusion tensor imaging (DTI) studies have demonstrated white matter (WM) abnormalities in patients with temporal lobe epilepsy (TLE). However, alterations in the topological properties of the WM structural network in patients with TLE remain unclear. Graph theoretical analysis provides a new perspective for evaluating the connectivity of WM structural networks. METHODS: DTI was used to map the structural networks of 18 patients with TLE (10 males and 8 females) and 29 (17 males and 12 females) age- and gender-matched normal controls (NC). Graph theory was used to analyze the whole-brain networks and their topological properties between the two groups. Finally, partial correlation analyses were performed on the weighted network properties and clinical characteristics, namely, duration of epilepsy, verbal intelligence quotient (IQ), and performance IQ. RESULTS: Patients with TLE exhibited reduced global efficiency and increased characteristic path length. A total of 31 regions with nodal efficiency alterations were detected in the fractional anisotropy_ weighted network of the patients. Communication hubs, such as the middle temporal gyrus, right inferior temporal gyrus, left calcarine, and right superior parietal gyrus, were also differently distributed in the patients compared with the NC. Several node regions showed close relationships with duration of epilepsy, verbal IQ, and performance IQ. CONCLUSIONS: Our results demonstrate the disruption of the WM structural network in TLE patients. This study may contribute to the further understanding of the pathological mechanism of TLE.

Temperature distribution analysis of tissue water vaporization during microwave ablation: experiments and simulations.

The temperature distribution in the region near a microwave antenna is a critical factor that affects the entire temperature field during microwave ablation of tissue. It is challenging to predict this distribution precisely, because the temperature in the near-antenna region varies greatly. The effects of water vaporisation and subsequent tissue carbonisation in an ex vivo porcine liver were therefore studied experimentally and in simulations. The enthalpy and high-temperature specific absorption rate (SAR) of liver tissues were calculated and incorporated into the simulation process. The accuracy of predictions for near-field temperatures in our simulations has reached the level where the average maximum error is less than 5°C. In addition, a modified thermal model that accounts for water vaporisation and the change in the SAR distribution pattern is proposed and validated with experiment. The results from this study may be useful in the clinical practice of microwave ablation and can be applied to predict the temperature field in surgical planning.