Predicting the trabecular bone apparent stiffness tensor with spherical convolutional neural networks

Bone protective effects of the polysaccharides from Grifola frondosa on ovariectomy-induced osteoporosis in mice via inhibiting PINK1/Parkin signaling, oxidative stress and inflammation

BACKGROUND

Polysaccharides from Grifola frondosa(GFP) have gained worldwide attention owing to their promising biological activities and potential health benefits.

PURPOSE

This study aimed to investigate the effects of GFP on alleviation of osteoporosis in ovariectomized (OVX) mice and examine the underlying mechanism.

METHOD

A mouse model of postmenopausal osteoporosis was established by OVX method, Forty eight C57BL/6 female mice were randomly divided into Normal group, OVX alone (Model group, n = 8), OVX + 10 mg/kg GFP (GFP-L group, n = 8), OVX + 20 mg/kg GFP (GFP-M group, n = 8), OVX + 40 mg/kg GFP (GFP-H group, n = 8), OVX + 10 mg/kg Estradiol valerate (Positive group, n = 8).

RESULTS

The results showed that compared with Model group, the concentrations of interleukin (IL)-1β, interleukin (IL)-6 and Tumor necrosis factor-α (TNF-α) were significantly reduced, the activity of superoxide dismutase (SOD) and glutathione (GSH) were significantly increased, the content of myeloperoxidase (MPO) and malondialdehyde (MDA) were significantly reduced, and the proteins levels of PINK1, Parkin, Beclin-1 and LC3-II were significantly decreased in the GFP groups.

CONCLUSION

This study demonstrates that GFP alleviates ovariectomy-induced osteoporosis via reduced secretion of inflammatory cytokines, improvement in the oxidative stress status in the body, and inhibition of the PINK1/Parkin signaling pathway.

Spherical Convolutional Neural Networks for Survival Rate Prediction in Cancer Patients

Objective

Survival Rate Prediction (SRP) is a valuable tool to assist in the clinical diagnosis and treatment planning of lung cancer patients. In recent years, deep learning (DL) based methods have shown great potential in medical image processing in general and SRP in particular. This study proposes a fully-automated method for SRP from computed tomography (CT) images, which combines an automatic segmentation of the tumor and a DL-based method for extracting rotational-invariant features.

Methods

In the first stage, the tumor is segmented from the CT image of the lungs. Here, we use a deep-learning-based method that entails a variational autoencoder to provide more information to a U-Net segmentation model. Next, the 3D volumetric image of the tumor is projected onto 2D spherical maps. These spherical maps serve as inputs for a spherical convolutional neural network that approximates the log risk for a generalized Cox proportional hazard model.

Results

The proposed method is compared with 17 baseline methods that combine different feature sets and prediction models using three publicly-available datasets: Lung1 (n=422), Lung3 (n=89), and H&N1 (n=136). We observed comparable C-index scores compared to the best-performing baseline methods in a 5-fold cross-validation on Lung1 (0.59 ± 0.03 vs. 0.62 ± 0.04). In comparison, it slightly outperforms all methods in inter-data set evaluation (0.64 vs. 0.63). The best-performing method from the first experiment reduced its performance to 0.61 and 0.62 for Lung3 and H&N1, respectively.

Discussion

The experiments suggest that the performance of spherical features is comparable with previous approaches, but they generalize better when applied to unseen datasets. That might imply that orientation-independent shape features are relevant for SRP. The performance of the proposed method was very similar, using manual and automatic segmentation methods. This makes the proposed model useful in cases where expert annotations are not available or difficult to obtain.