A Fully Automated Method for 3D Individual Tooth Identification and Segmentation in Dental CBCT.

Accurate and automatic segmentation of three-dimensional (3D) individual teeth from cone-beam computerized tomography (CBCT) images is a challenging problem because of the difficulty in separating an individual tooth from adjacent teeth and its surrounding alveolar bone. Thus, this paper proposes a fully automated method of identifying and segmenting 3D individual teeth from dental CBCT images. The proposed method addresses the aforementioned difficulty by developing a deep learning-based hierarchical multi-step model. First, it automatically generates upper and lower jaws panoramic images to overcome the computational complexity caused by high-dimensional data and the curse of dimensionality associated with limited training dataset. The obtained 2D panoramic images are then used to identify 2D individual teeth and capture loose- and tight- regions of interest (ROIs) of 3D individual teeth. Finally, accurate 3D individual tooth segmentation is achieved using both loose and tight ROIs. Experimental results showed that the proposed method achieved an F1-score of 93.35 percent for tooth identification and a Dice similarity coefficient of 94.79 percent for individual 3D tooth segmentation. The results demonstrate that the proposed method provides an effective clinical and practical framework for digital dentistry.

Achieving a Low-Voltage, High-Mobility IGZO Transistor through an ALD-Derived Bilayer Channel and a Hafnia-Based Gate Dielectric Stack.

Ultrahigh-resolution displays for augmented reality (AR) and virtual reality (VR) applications require a novel architecture and process. Atomic-layer deposition (ALD) enables the facile fabrication of indium-gallium zinc oxide (IGZO) thin-film transistors (TFTs) on a substrate with a nonplanar surface due to its excellent step coverage and accurate thickness control. Here, we report all-ALD-derived TFTs using IGZO and HfO2 as the channel layer and gate insulator, respectively. A bilayer IGZO channel structure consisting of a 10 nm base layer (In0.52Ga0.29Zn0.19O) with good stability and a 3 nm boost layer (In0.82Ga0.08Zn0.10O) with extremely high mobility was designed based on a cation combinatorial study of the ALD-derived IGZO system. Reducing the thickness of the HfO2 dielectric film by the ALD process offers high areal capacitance in field-effect transistors, which allows low-voltage drivability and enhanced carrier transport. The intrinsic inferior stability of the HfO2 gate insulator was effectively mitigated by the insertion of an ALD-derived 4 nm Al2O3 interfacial layer between HfO2 and the IGZO film. The optimized bilayer IGZO TFTs with HfO2-based gate insulators exhibited excellent performances with a high field-effect mobility of 74.0 ± 0.91 cm2/(V s), a low subthreshold swing of 0.13 ± 0.01 V/dec, a threshold voltage of 0.20 ± 0.24 V, and an ION/OFF of ∼3.2 × 108 in a low-operation-voltage (≤2 V) range. This promising result was due to the synergic effects of a bilayer IGZO channel and HfO2-based gate insulator with a high permittivity, which were mainly attributed to the effective carrier confinement in the boost layer with high mobility, low free carrier density of the base layer with a low VO concentration, and HfO2-induced high effective capacitance.

Automated ultrasound assessment of amniotic fluid index using deep learning.

The estimation of antenatal amniotic fluid (AF) volume (AFV) is important as it offers crucial information about fetal development, fetal well-being, and perinatal prognosis. However, AFV measurement is cumbersome and patient specific. Moreover, it is heavily sonographer-dependent, with measurement accuracy varying greatly depending on the sonographer's experience. Therefore, the development of accurate, robust, and adoptable methods to evaluate AFV is highly desirable. In this regard, automation is expected to reduce user-based variability and workload of sonographers. However, automating AFV measurement is very challenging, because accurate detection of AF pockets is difficult owing to various confusing factors, such as reverberation artifact, AF mimicking region and floating matter. Furthermore, AF pocket exhibits an unspecified variety of shapes and sizes, and ultrasound images often show missing or incomplete structural boundaries. To overcome the abovementioned difficulties, we develop a hierarchical deep-learning-based method, which consider clinicians' anatomical-knowledge-based approaches. The key step is the segmentation of the AF pocket using our proposed deep learning network, AF-net. AF-net is a variation of U-net combined with three complementary concepts - atrous convolution, multi-scale side-input layer, and side-output layer. The experimental results demonstrate that the proposed method provides a measurement of the amniotic fluid index (AFI) that is as robust and precise as the results from clinicians. The proposed method achieved a Dice similarity of 0.877±0.086 for AF segmentation and achieved a mean absolute error of 2.666±2.986 and mean relative error of 0.018±0.023 for AFI value. To the best of our knowledge, our method, for the first time, provides an automated measurement of AFI.

Automatic detection and segmentation of lumbar vertebrae from X-ray images for compression fracture evaluation.

For compression fracture detection and evaluation, an automatic X-ray image segmentation technique that combines deep-learning and level-set methods is proposed. Automatic segmentation is much more difficult for X-ray images than for CT or MRI images because they contain overlapping shadows of thoracoabdominal structures including lungs, bowel gases, and other bony structures such as ribs. Additional difficulties include unclear object boundaries, the complex shape of the vertebra, inter-patient variability, and variations in image contrast. Accordingly, a structured hierarchical segmentation method is presented that combines the advantages of two deep-learning methods. Pose-driven learning is used to selectively identify the five lumbar vertebrae in an accurate and robust manner. With knowledge of the vertebral positions, M-net is employed to segment the individual vertebra. Finally, fine-tuning segmentation is applied by combining the level-set method with the previously obtained segmentation results. The performance of the proposed method was validated by 160 lumbar X-ray images, resulting in a mean Dice similarity metric of 91.60±2.22%. The results show that the proposed method achieves accurate and robust identification of each lumbar vertebra and fine segmentation of individual vertebra.

High-Performance Thin-Film Transistors with an Atomic-Layer-Deposited Indium Gallium Oxide Channel: A Cation Combinatorial Approach.

The effect of gallium (Ga) concentration on the structural evolution of atomic-layer-deposited indium gallium oxide (IGO) (In1-xGaxO) films as high-mobility n-channel semiconducting layers was investigated. Different Ga concentrations in 10-13 nm thick In1-xGaxO films allowed versatile phase structures to be amorphous, highly ordered, and randomly oriented crystalline by thermal annealing at either 400 or 700 °C for 1 h. Heavy Ga concentrations above 34 atom % caused a phase transformation from a polycrystalline bixbyite to an amorphous IGO film at 400 °C, while proper Ga concentration produced a highly ordered bixbyite crystal structure at 700 °C. The resulting highly ordered In0.66Ga0.34O film show unexpectedly high carrier mobility (µFE) values of 60.7 ± 1.0 cm2 V-1 s-1, a threshold voltage (VTH) of -0.80 ± 0.05 V, and an ION/OFF ratio of 5.1 × 109 in field-effect transistors (FETs). In contrast, the FETs having polycrystalline In1-xGaxO films with higher In fractions (x = 0.18 and 0.25) showed reasonable µFE values of 40.3 ± 1.6 and 31.5 ± 2.4 cm2 V-1 s-1, VTH of -0.64 ± 0.40 and -0.43 ± 0.06 V, and ION/OFF ratios of 2.5 × 109 and 1.4 × 109, respectively. The resulting superior performance of the In0.66Ga0.34O-film-based FET was attributed to a morphology having fewer grain boundaries, with higher mass densification and lower oxygen vacancy defect density of the bixbyite crystallites. Also, the In0.66Ga0.34O transistor was found to show the most stable behavior against an external gate bias stress.

Indeterminate observations (liver imaging reporting and data system category 3) on MRI in the cirrhotic liver: fate and clinical implications.

OBJECTIVE: The purpose of this article is to retrospectively evaluate the imaging characteristics and outcomes of indeterminate observations (Liver Imaging Reporting and Data System category 3) detected on MRI in patients with cirrhosis. MATERIALS AND METHODS: Sixty-nine indeterminate observations in 52 patients with cirrhosis were imaged with hepatobiliary contrast agent-enhanced MRI. Observations were evaluated retrospectively in terms of the location, size, enhancement pattern, and follow-up results. Each observation was categorized as stable or progressed observations according to serial follow-up MRI. RESULTS: The mean follow-up interval was 11.2 months. Forty-six (67%) of the total observations showed arterial enhancement, and 23 (33%) observations showed isointense signal or low signal intensity on arterial phase. The indeterminate observations were classified as arterial enhancement with fade-out appearance (41 observations [59%]), arterial enhancement with washout appearance (five observations [7%]), and nonhyperenhancing observations (23 observations [33%]). Two of 69 observations (3%) were hyperintense on T2-weighted images, and four of 55 observations (7%) were hyperintense on hepatocellular phase. On the final follow-up MRI examinations, four (6%) observations proved to be probable or definite hepatocellular carcinoma, 55 (80%) remained stable, and 10 (14%) decreased in size or were no longer visible. CONCLUSION: The most common cause of indeterminate observations on MRI is hypervascular pseudolesions that were clinically stable on follow-up imaging.