Abstract The vertebral compression is a significant factor for determining the prognosis of osteoporotic vertebral compression fractures and is generally measured manually by specialists. The consequent misdiagnosis or delayed diagnosis can be fatal for patients. In this study, we trained and evaluated the performance of a vertebral body segmentation model and a vertebral compression measurement model based on convolutional neural networks. For vertebral body segmentation, we used a recurrent residual U-Net model, with an average sensitivity of 0.934 (± 0.086), an average specificity of 0.997 (± 0.002), an average accuracy of 0.987 (± 0.005), and an average dice similarity coefficient of 0.923 (± 0.073). We then generated 1134 data points on the images of three vertebral bodies by labeling each segment of the segmented vertebral body. These were used in the vertebral compression measurement model based on linear regression and multi-scale residual dilated blocks. The model yielded an average mean absolute error of 2.637 (± 1.872) (%), an average mean square error of 13.985 (± 24.107) (%), and an average root mean square error of 3.739 (± 2.187) (%) in fractured vertebral body data. The proposed algorithm has significant potential for aiding the diagnosis of vertebral compression fractures.
This study introduces the thickness-tapered channel design for flow field-flow fractionation (FlFFF) for the first time. In this design, the channel thickness linearly decreases along the channel axis such that the flow velocity increases down the channel. Channel thickness is an important variable for controlling retention time and resolution in field-flow fractionation. Especially, in the steric/hyperlayer mode of FlFFF, in which particles (>1 μm) migrate at elevated heights above the channel wall owing to hydrodynamic lift forces, the migration of long-retaining smaller-sized particles can be enhanced in a relatively thin channel or by increasing the migration flow rate; however, an upper size limit that can be resolved is simultaneously sacrificed. A thickness-tapered channel was constructed without a channel spacer by carving the surface of a channel block such that the channel inlet was deeper than the outlet (w = 400 → 200 μm). The performance of a thickness-tapered channel was evaluated using polystyrene standards and compared to that of a channel of uniform thickness (w = 300 μm) with a similar effective channel volume in terms of sample recovery, dynamic size range of separation, and steric transition under different flow rate conditions. The thickness-tapered channel can be an alternative to maintain the resolving power for particles with an upper large-diameter limit, faster separation of particles with a lower limit, and higher elution recovery without implementing the additional field-programming option.
In this study, we aimed to investigate quantitative differences in performance in terms of comparing the automated classification of deep vein thrombosis (DVT) using two categories of artificial intelligence algorithms: deep learning based on convolutional neural networks (CNNs) and conventional machine learning. We retrospectively enrolled 659 participants (DVT patients, 282; normal controls, 377) who were evaluated using contrast-enhanced lower extremity computed tomography (CT) venography. Conventional machine learning consists of logistic regression (LR), support vector machines (SVM), random forests (RF), and extreme gradient boosts (XGB). Deep learning based on CNN included the VGG16, VGG19, Resnet50, and Resnet152 models. According to the mean generated AUC values, we found that the CNN-based VGG16 model showed a 0.007 higher performance (0.982 ± 0.014) as compared with the XGB model (0.975 ± 0.010), which showed the highest performance among the conventional machine learning models. In the conventional machine learning-based classifications, we found that the radiomic features presenting a statistically significant effect were median values and skewness. We found that the VGG16 model within the deep learning algorithm distinguished deep vein thrombosis on CT images most accurately, with slightly higher AUC values as compared with the other AI algorithms used in this study. Our results guide research directions and medical practice.
Seafoods and seaweeds represent some of the most important reservoirs of new therapeutic compounds for humans. Seaweed has been shown to have several biological activities, including anticancer activity. This review focuses on colorectal and breast cancers, which are major causes of cancer-related mortality in men and women. It also describes various compounds extracted from a range of seaweeds that have been shown to eradicate or slow the progression of cancer. Fucoidan extracted from the brown algae Fucus spp. has shown activity against both colorectal and breast cancers. Furthermore, we review the mechanisms through which these compounds can induce apoptosis in vitro and in vivo. By considering the ability of compounds present in seaweeds to act against colorectal and breast cancers, this review highlights the potential use of seaweeds as anticancer agents.
A comprehensive lipid profile was analyzed in patients with non-small cell lung cancer (NSCLC) using nanoflow ultrahigh-performance liquid chromatography-electrospray ionization-tandem mass spectrometry. The study investigated 297, 202, and 166 lipids in saliva, plasma, and fecal samples, respectively, comparing NSCLC patients to healthy controls. Lipids with significant changes (>2-fold, p<0.05) were further analyzed in each sample type. Both saliva and plasma exhibited similar lipid alteration patterns in NSCLC, but saliva showed more pronounced changes and fecal lipids had weak correlation with those of saliva and plasma. Total triglycerides (TGs) increased (>2–3 folds) in plasma and saliva but decreased in fecal samples. Three specific TGs (50:2, 52:5, and 54:6) were significantly increased in NSCLC across all sample types. A common ceramide species (d18:1/24:0) decreased in both plasma and saliva but increased in fecal samples. Additionally, phosphatidylinositol 38:4 decreased by approximately 2-fold in plasma and saliva. Phosphatidylserine 36:1 was selectively detected in saliva and showed a subsequent decrease, making it a potential biomarker for predicting lung cancer. The study identifies 27 salivary, 10 plasma, and 16 fecal lipids as candidate markers for NSCLC by statistical evaluations. Moreover, it highlights the potential of saliva in understanding changes in lipid metabolism associated with NSCLC.
'%3e%3cg id='b'%3e%3cpath id='a' stroke='%23000' stroke-width='2' d='M-6-25H6m-12 3H6m-12 3H6'/%3e%3cuse xlink:href='%23a' y='44'/%3e%3c/g%3e%3cpath stroke='%23fff' d='M0 17v10'/%3e%3ccircle r='12' fill='%23cd2e3a'/%3e%3cpath fill='%230047a0' d='M0-12A6 6 0 0 0 0 0a6 6 0 0 1 0 12 12 12 0 0 1 0-24z'/%3e%3c/g%3e%3cg transform='rotate(-123.69)'%3e%3cuse xlink:href='%23b'/%3e%3cpath stroke='%23fff' d='M0-23.5v3M0 17v3.5m0 3v3'/%3e%3c/g%3e%3c/svg%3e)