Automated segmentation of liver and hepatic vessels on portal venous phase computed tomography images using a deep learning algorithm.

BACKGROUND: CT-image segmentation for liver and hepatic vessels can facilitate liver surgical planning. However, time-consuming process and inter-observer variations of manual segmentation have limited wider application in clinical practice. PURPOSE: Our study aimed to propose an automated deep learning (DL) segmentation algorithm for liver and hepatic vessels on portal venous phase CT images. METHODS: This retrospective study was performed to develop a coarse-to-fine DL-based algorithm that was trained, validated, and tested using private 413, 52, and 50 portal venous phase CT images, respectively. Additionally, the performance of the DL algorithm was extensively evaluated and compared with manual segmentation using an independent clinical dataset of preoperative contrast-enhanced CT images from 44 patients with hepatic focal lesions. The accuracy of DL-based segmentation was quantitatively evaluated using the Dice Similarity Coefficient (DSC) and complementary metrics [Normalized Surface Dice (NSD) and Hausdorff distance_95 (HD95) for liver segmentation, Recall and Precision for hepatic vessel segmentation]. The processing time for DL and manual segmentation was also compared. RESULTS: Our DL algorithm achieved accurate liver segmentation with DSC of 0.98, NSD of 0.92, and HD95 of 1.52 mm. DL-segmentation of hepatic veins, portal veins, and inferior vena cava attained DSC of 0.86, 0.89, and 0.94, respectively. Compared with the manual approach, the DL algorithm significantly outperformed with better segmentation results for both liver and hepatic vessels, with higher accuracy of liver and hepatic vessel segmentation (all p < 0.001) in independent 44 clinical data. In addition, the DL method significantly reduced the manual processing time of clinical postprocessing (p < 0.001). CONCLUSIONS: The proposed DL algorithm potentially enabled accurate and rapid segmentation for liver and hepatic vessels using portal venous phase contrast CT images.

Zinc-Guided 3D Graphene for Thermally Chargeable Supercapacitors to Harvest Low-Grade Heat.

Low-grade heat energy recycling is the key technology of waste-heat utilization, which needs to be improved. Here, we use a zinc-assisted solid-state pyrolysis route to prepare zinc-guided 3D graphene (ZnG), a 3D porous graphene with the interconnected structure. The obtained ZnG, with a high specific surface area of 1817 m2·g-1 and abundant micropores and mesopores, gives a specific capacitance of 139 F·g-1 in a neutral electrolyte when used as electrode material for supercapacitors. At a high current density of 8 A·g-1, the capacitance retention is 93% after 10,000 cycles. When ZnG is used for thermally chargeable supercapacitors, the thermoelectric conversion of the low-grade heat energy is successfully realized. This work thus provides a demonstration for low-grade heat energy conversion.

Enhanced lesion-to-bubble ratio on ultrasonic Nakagami imaging for monitoring of high-intensity focused ultrasound.

OBJECTIVES: This work explored the feasibility of using ultrasonic Nakagami imaging to enhance the contrast between thermal lesions and bubbles induced by high-intensity focused ultrasound (US) in a transparent tissue-mimicking phantom at different acoustic power levels. METHODS: The term "lesion-to-bubble ratio" was proposed and defined as the ratio of the scattered power from the thermal lesion to the scattered power from the bubbles calculated in the various monitoring of images for high-intensity focused US. Two-dimensional radiofrequency data backscattered from the exposed region were captured by a modified diagnostic US scanner to estimate the Nakagami statistical parameter, m, and reconstruct the ultrasonic B-mode images and Nakagami parameter images. The dynamic changes in the lesion-to-bubble ratio over the US exposure procedure were calculated simultaneously and compared among video photos, B-mode images, and Nakagami images for monitoring of high-intensity focused US. RESULTS: After a small thermal lesion was induced by high-intensity focused US in the phantom, the lesion-to-bubble ratio values corresponding to the video photo, B-mode image, and Nakagami image were 5.3, 1, and 9.8 dB, respectively. When a large thermal lesion appeared in the phantom, the ratio values increased to 7.2, 3, and 14 dB. During US exposure, the ratio values calculated for the video photo, B-mode image, and Nakagami image began to increase gradually and rose to peak values of 8.3, 2.9, and 14.8 dB at the end of the US exposure. CONCLUSIONS: This preliminary study on a tissue-mimicking phantom suggests that Nakagami imaging may have a potential use in enhancing the lesion-to-bubble ratio for monitoring high-intensity focused US. Further studies in vivo and in vitro will be needed to evaluate the potential applications for high-intensity focused US.

Feasibility of using Nakagami distribution in evaluating the formation of ultrasound-induced thermal lesions.

The acoustic posterior shadowing effects of bubbles influence the accuracy for defining the location and range of ablated thermal lesions during focused ultrasound surgery when using ultrasonic monitoring imaging. This paper explored the feasibility of using Nakagami distribution to evaluate the ablated region induced by focused ultrasound exposures at different acoustic power levels in transparent tissue-mimicking phantoms. The mean value of the Nakagami parameter m was about 0.5 in the cavitation region and increased to around 1 in the ablated region. Nakagami images were not subject to significant shadowing effects of bubbles. Ultrasound-induced thermal lesions observed in the photos and Nakagami images were overshadowed by bubbles in the B-mode images. The lesion size predicted in the Nakagami images was smaller than that predicted in the photos due to the sub resolvable effect of Nakagami imaging at the interface. This preliminary study on tissue-mimicking phantom suggested that the Nakagami parameter m may have the potential use in evaluating the formation of ultrasound-induced thermal lesion when the shadowing effect of bubbles is strong while the thermal lesion was small. Further studies in vivo and in vitro will be needed to evaluate the potential application.

Compare ultrasound-mediated heating and cavitation between flowing polymer- and lipid-shelled microbubbles during focused ultrasound exposures.

This paper compares the efficiency of flowing polymer- and lipid-shelled microbubbles (MBs) in the heating and cavitation during focused ultrasound exposures. Temperature and cavitation activity were simultaneously measured as the two types of shelled MBs and saline flowing through a 3 mm diameter vessel in the phantom with varying flow velocities (0-20 cm/s) at different acoustic power levels (0.6-20 W) with each exposure for 5 s. Temperature and cavitation for the lipid-shelled MBs were higher than those for the polymer-shelled MBs. Temperature rise decreased with increasing flow velocities for the two types of shelled MBs and saline at acoustic power 1.5 W. At acoustic power 11.1 W, temperature rise increased with increasing flow velocities for the lipid-shelled MBs. For the polymer-shelled MBs, the temperature rise increased with increasing flow velocities from 3-15 cm/s and decreased at 20 cm/s. Cavitation increased with increasing flow velocity for the two shelled MBs and there were no significant changes of cavitation with increasing flow velocities for saline. These results suggested that lipid-shelled MBs may have a greater efficiency than polymer-shelled MBs in heating and cavitation during focused ultrasound exposures.

Palladium-catalyzed hydroboration reaction of unactivated alkynes with bis (pinacolato) diboron in water.

A highly efficient and mild palladium-catalyzed hydroboration of unactivated internal alkynes in water is described. Both aryl- and alkyl-substituted alkynes proceeded smoothly within the reaction time to afford the desired vinylboronates in moderate to high yields. Bis (pinacolato) diboron was used to afford α- and ß-hydroborated products in the presence of HOAc. These reactions showed high reactivities and tolerance, thus providing a promising method for the synthesis of alkenyl boron compounds.