tFUSFormer: Physics-Guided Super-Resolution Transformer for Simulation of Transcranial Focused Ultrasound Propagation in Brain Stimulation.

Transcranial focused ultrasound (tFUS) has emerged as a new mode of non-invasive brain stimulation (NIBS), with its exquisite spatial precision and capacity to reach the deep regions of the brain. The placement of the acoustic focus onto the desired part of the brain is critical for successful tFUS procedures; however, acoustic wave propagation is severely affected by the skull, distorting the focal location/shape and the pressure level. High-resolution (HR) numerical simulation allows for monitoring of acoustic pressure within the skull but with a considerable computational burden. To address this challenge, we employed a 4× super-resolution (SR) Swin Transformer method to improve the precision of estimating tFUS acoustic pressure field, targeting operator-defined brain areas. The training datasets were obtained through numerical simulations at both ultra-low (2.0 [Formula: see text]) and high (0.5 [Formula: see text]) resolutions, conducted on in vivo CT images of 12 human skulls. Our multivariable datasets, which incorporate physical properties of the acoustic pressure field, wave velocity, and skull CT images, were utilized to train three-dimensional SR models. We found that our method yielded 87.99 ± 4.28% accuracy in terms of focal volume conformity under foreseen skull data, and accuracy of 82.32 ± 5.83% for unforeseen skulls, respectively. Moreover, a significant improvement of 99.4% in computational efficiency compared to the traditional 0.5 [Formula: see text] HR numerical simulation was shown. The presented technique, when adopted in guiding the placement of the FUS transducer to engage specific brain targets, holds great potential in enhancing the safety and effectiveness of tFUS therapy.

Super-resolution techniques for biomedical applications and challenges.

Super-resolution (SR) techniques have revolutionized the field of biomedical applications by detailing the structures at resolutions beyond the limits of imaging or measuring tools. These techniques have been applied in various biomedical applications, including microscopy, magnetic resonance imaging (MRI), computed tomography (CT), X-ray, electroencephalogram (EEG), ultrasound, etc. SR methods are categorized into two main types: traditional non-learning-based methods and modern learning-based approaches. In both applications, SR methodologies have been effectively utilized on biomedical images, enhancing the visualization of complex biological structures. Additionally, these methods have been employed on biomedical data, leading to improvements in computational precision and efficiency for biomedical simulations. The use of SR techniques has resulted in more detailed and accurate analyses in diagnostics and research, essential for early disease detection and treatment planning. However, challenges such as computational demands, data interpretation complexities, and the lack of unified high-quality data persist. The article emphasizes these issues, underscoring the need for ongoing development in SR technologies to further improve biomedical research and patient care outcomes.

Multivariable-incorporating super-resolution residual network for transcranial focused ultrasound simulation.

BACKGROUND AND OBJECTIVE: Transcranial focused ultrasound (tFUS) has emerged as a new non-invasive brain stimulation (NIBS) modality, with its exquisite ability to reach deep brain areas at a high spatial resolution. Accurate placement of an acoustic focus to a target region of the brain is crucial during tFUS treatment; however, the distortion of acoustic wave propagation through the intact skull casts challenges. High-resolution numerical simulation allows for monitoring of the acoustic pressure field in the cranium but also demands extensive computational loads. In this study, we adopt a super-resolution residual network technique based on a deep convolution to enhance the prediction quality of the FUS acoustic pressure field in the targeted brain regions. METHODS: The training dataset was acquired by numerical simulations performed at low-(1.0 mm) and high-resolutions (0.5mm) on three ex vivo human calvariae. Five different super-resolution (SR) network models were trained by using a multivariable dataset in 3D, which incorporated information on the acoustic pressure field, wave velocity, and localized skull computed tomography (CT) images. RESULTS: The accuracy of 80.87±4.50% in predicting the focal volume with a substantial improvement of 86.91% in computational cost compared to the conventional high-resolution numerical simulation was achieved. The results suggest that the method can greatly reduce the simulation time without sacrificing accuracy and improve the accuracy further with the use of additional inputs. CONCLUSIONS: In this research, we developed multivariable-incorporating SR neural networks for transcranial focused ultrasound simulation. Our super-resolution technique may contribute to promoting the safety and efficacy of tFUS-mediated NIBS by providing on-site feedback information on the intracranial pressure field to the operator.

Camera orientation estimation using voting approach on the Gaussian sphere for in-vehicle camera.

Calibration of a vehicle camera is a key technology for advanced driver assistance systems (ADAS). This paper presents a novel estimation method to measure the orientation of a camera that is mounted on a driving vehicle. By considering the characteristics of vehicle cameras and driving environment, we detect three orthogonal vanishing points as a basis of the imaging geometry. The proposed method consists of three steps: i) detection of lines projected to the Gaussian sphere and extraction of the plane normal, ii) estimation of the vanishing point about the optical axis using linear Hough transform, and iii) voting for the rest two vanishing points using circular histogram. The proposed method increases both accuracy and stability by considering the practical driving situation using sequentially estimated three vanishing points. In addition, we can rapidly estimate the orientation by converting the voting space into a 2D plane at each stage. As a result, the proposed method can quickly and accurately estimate the orientation of the vehicle camera in a normal driving situation.

Intelligent Image-Based Railway Inspection System Using Deep Learning-Based Object Detection and Weber Contrast-Based Image Comparison.

For sustainable operation and maintenance of urban railway infrastructure, intelligent visual inspection of the railway infrastructure attracts increasing attention to avoid unreliable, manual observation by humans at night, while trains do not operate. Although various automatic approaches were proposed using image processing and computer vision techniques, most of them are focused only on railway tracks. In this paper, we present a novel railway inspection system using facility detection based on deep convolutional neural network and computer vision-based image comparison approach. The proposed system aims to automatically detect wears and cracks by comparing a pair of corresponding image sets acquired at different times. We installed line scan camera on the roof of the train. Unlike an area-based camera, the line scan camera quickly acquires images with a wide field of view. The proposed system consists of three main modules: (i) image reconstruction for registration of facility positions, (ii) facility detection using an improved single shot detector, and (iii) deformed region detection using image processing and computer vision techniques. In experiments, we demonstrate that the proposed system accurately finds facilities and detects their potential defects. For that reason, the proposed system can provide various advantages such as cost reduction for maintenance and accident prevention.

Clinical Application of Real-Time Sonoelastography for Evaluation of Medial Epicondylitis: A Pilot Study.

The aim of this study was to evaluate the diagnostic potential of real-time sonoelastography (RSE) in medial epicondylitis by comparing clinically diagnosed patients and patients without medial elbow pain. From July 2016 to December 2017, gray-scale sonographic findings (swelling, cortical irregularity, hypo-echogenicity, calcification and tear), color Doppler findings (hyperemia) and sonoelastographic findings (elastographic grade on a 3-point visual scale and strain ratio from two regions of interest) for 63 elbows of 56 patients were compared. Twenty-four patients with 29 imaged elbows were clinically diagnosed with medial epicondylitis, and 32 patients with 34 imaged elbows had no medial elbow pain. Cortical irregularity, hypo-echogenicity, calcification, elastographic grade and strain ratio revealed significant differences (p < 0.05). Among these, strain ratio had the highest diagnostic performance (area under the curve: 0.985). Real-time sonoelastography, which can obtain both elastographic grade and strain ratio, is valuable as a supplementary tool in the diagnosis of medial epicondylitis.

[Deep Vein Thrombosis Due to Compression of Huge Hepatic Cyst Successfully Treated by Inferior Vena Cava Filter and Cyst Drainage].

An 88-year-old woman complained of right quadrant abdominal pain and severe edema in both legs. She had a history of pulmonary embolism one month ago. Abdomen CT showed a huge hepatic cyst compressing the intrahepatic portion of the inferior vena cava (IVC). The venogram CT showed multifocal thrombosis in the iliocaval and both lower extremity veins. Percutaneous hepatic cyst drainage was carried out. Fluid analysis presented leukocytosis, which suggested an infected hepatic cyst. To prevent secondary pulmonary thromboembolism, an IVC filter was inserted before catheter drainage for the hepatic cyst. One week later, abdominal pain was relieved. Then, sclerotherapy for the remnant hepatic cyst was performed by ethanol. Follow-up CT showed an increased amount of thrombosis in the iliocaval and left calf vein, but the IVC filter prevented another thromboembolic event successfully. The patient started dabigatran, a new oral anticoagulant, and compression stockings were applied to both legs. After one month, no visible thrombosis in the pelvis or either extremity was detected in abdominal CT. This case suggests that a huge hepatic cyst, especially with infection, should be considered as a possible cause of deep vein thrombosis if no other risk factors for thromboembolism exist.

Cryoablation of Hepatocellular Carcinoma with High-Risk for Percutaneous Ablation: Safety and Efficacy.

PURPOSE: The aim of this study was to evaluate the safety and effectiveness of cryoablation in the treatment of subcapsular hepatocellular carcinoma (HCC) adjacent to various organs. MATERIALS AND METHODS: Twenty-eight patients with subcapsular HCC were treated with cryoablation in our institution. The degree of peri-procedural pain was measured using the visual analog scale (VAS). Technical success, local tumor progression, and overall disease progression rates were calculated. Procedure-related complications were identified by reviewing electronic medical records. Biochemical data, including serum aspartate aminotransferase (AST), alanine aminotransferase (ALT), and total bilirubin levels before and after the procedure were collected. RESULTS: Subcapsular HCC tumors were located near the gallbladder, colon, stomach, kidney, diaphragm, or abdominal wall. The technical success rate of cryoablation was 96.4 % (27/28). Local recurrence- and progression-free survival rates were 96 and 84 % at 6 months, and 82 and 43 % at 1 year, respectively. All patients survived during the follow-up period. The VAS pain score ranged from 0 to 3 (mean, 1.57). A major complication occurred in one patient (3.6 %) and minor complications occurred at a rate of 17.9 %. Transient elevations of serum AST, ALT, and bilirubin levels were observed. CONCLUSION: Cryoablation is a safe and an effective procedure for the treatment of subcapsular HCC adjacent to various major organs.

Transarterial Chemoembolization Using Sorafenib in a Rabbit VX2 Liver Tumor Model: Pharmacokinetics and Antitumor Effect.

PURPOSE: To investigate feasibility, safety, and effect of transarterial chemoembolization using sorafenib on degree of tumor necrosis in a rabbit VX2 liver tumor model. MATERIALS AND METHODS: New Zealand White rabbits (n = 20) with a VX2 tumor were divided into two groups; one group was treated with hepatic arterial administration of 0.5 mL ethiodized oil alone (Lipiodol; Guerbet, Aulnay-sous-Bois, France) (transarterial embolization with Lipiodol [TAE-L] group), and one group was treated with 0.5 mL ethiodized oil plus 10 mg sorafenib (transarterial embolization with sorafenib [TAE-S] group). Liquid chromatography tandem mass spectrometry was used to measure sorafenib concentration in peripheral blood and tissue. Hepatic enzymes, vascular endothelial growth factor (VEGF), and hypoxia-inducible factor 1α (HIF-1α) were measured at 0, 24, and 72 hours after treatment. Histopathologic examination was performed to evaluate extent of tumor necrosis and normal parenchymal damage. RESULTS: Serum sorafenib concentration peaked at 2 hours after treatment. The mean tissue concentration was 406.8 times greater than the serum concentration. Aspartate aminotransferase and alanine aminotransferase levels were significantly elevated in the TAE-S group at 24 hours after treatment. Serum VEGF and HIF-1α concentrations were not significantly different between the TAE-L and TAE-S groups. Hepatic parenchymal damage was more severe in the TAE-S group. Mean fraction of tumor necrosis after treatment was significantly greater in the TAE-S group. CONCLUSIONS: Transarterial chemoembolization using sorafenib resulted in a high intrahepatic concentration of sorafenib. The degree of tumor necrosis was significantly greater in the TAE-S group compared with the TAE-L group, but more severe toxicity of normal liver tissue also occurred.

Plug-Assisted Retrograde Transvenous Obliteration for the Treatment of Gastric Variceal Hemorrhage.

OBJECTIVE: To evaluate the feasibility, safety, and clinical outcomes of plug-assisted retrograde transvenous obliteration (PARTO) to treat gastric variceal hemorrhage in patients with portal hypertension. MATERIALS AND METHODS: From May 2012 to June 2014, 19 patients (11 men and 8 women, median age; 61, with history of gastric variceal hemorrhage; 17, active bleeding; 2) who underwent PARTO using a vascular plug and a gelfoam pledget were retrospectively analyzed. Clinical and laboratory data were examined to evaluate primary (technical and clinical success, complications) and secondary (worsening of esophageal varix [EV], change in liver function) end points. Median follow-up duration was 11 months, from 6.5 to 18 months. The Wilcoxon signed-rank test was used to compare laboratory data before and after the procedure. RESULTS: Technical success (complete occlusion of the efferent shunt and complete filling of gastric varix [GV] with a gelfoam slurry) was achieved in 18 of 19 (94.7%) patients. The embolic materials could not reach the GV in 1 patient who had endoscopic glue injection before our procedure. The clinical success rate (no recurrence of gastric variceal bleeding) was the same because the technically failed patient showed recurrent bleeding later. Acute complications included fever (n = 2), fever and hypotension (n = 2; one diagnosed adrenal insufficiency), and transient microscopic hematuria (n = 3). Ten patients underwent follow-up endoscopy; all exhibited GV improvement, except 2 without endoscopic change. Five patients exhibited aggravated EV, and 2 of them had a bleeding event. Laboratory findings were significantly improved after PARTO. CONCLUSION: PARTO is technically feasible, safe, and effective for gastric variceal hemorrhage in patients with portal hypertension.