RESUMEN
Screening for osteoporosis is crucial for early detection and prevention, yet it faces challenges due to the low accuracy of calcaneal quantitative ultrasound (QUS) and limited access to dual-energy X-ray absorptiometry (DXA) scans. Recent advances in AI offer a promising solution through opportunistic screening using existing medical images. This study aims to utilize deep learning techniques to develop a model that analyzes chest X-ray (CXR) images for osteoporosis screening. This study included the AI model development stage and the clinical validation stage. In the AI model development stage, the combined dataset of 5122 paired CXR images and DXA reports from the patients aged 20 to 98 years at a medical center was collected. The images were enhanced and filtered for hardware retention such as pedicle screws, bone cement, artificial intervertebral discs or severe deformity in target level of T12 and L1. The dataset was then separated into training, validating, and testing datasets for model training and performance validation. In the clinical validation stage, we collected 440 paired CXR images and DXA reports from both the TCVGH and Joy Clinic, including 304 pared data from TCVGH and 136 paired data from Joy Clinic. The pre-clinical test yielded an area under the curve (AUC) of 0.940, while the clinical validation showed an AUC of 0.946. Pearson's correlation coefficient was 0.88. The model demonstrated an overall accuracy, sensitivity, and specificity of 89.0%, 88.7%, and 89.4%, respectively. This study proposes an AI model for opportunistic osteoporosis screening through CXR, demonstrating good performance and suggesting its potential for broad adoption in preliminary screening among high-risk populations.
RESUMEN
Alzheimer's disease (AD) is characterized by extracellular deposition of amyloid plaques, which are predominantly composed of amyloid-ß (Aß) peptide derived from amyloid precursor protein (APP) cleavage. APP interacts with tropomyosin receptor kinase A, a neurotrophic receptor associated with gangliosides and mediating neuronal survival and differentiation through the extracellular signal-regulated protein kinase (ERK) pathway. The ganglioside Hp-s1's analogue Hp-s1A exerts neuritogenic activity; however, its effect on AD pathology remains unknown. To test the hypothesis that Hp-s1A is a potential candidate to treat AD, we established the AD-modeled cell line by expressing human Swedish and Indiana APP gene (APP-Swe/Ind) in N2a mouse neuroblastoma cells. The cells were treated with Hp-s1A or monosialoganglioside GM1 for comparison. The AD model cells expressing APP-Swe/Ind exhibited a significant reduction in viability, as well as neurite outgrowth rate, in comparison to the control cells expressing APP-695. APP C-terminal fragment-ß (CTFß) and Aß42 were increased in the AD cell lysates and the culture media, respectively. With the treatment of either Hp-s1A or GM1 at 1 µM, the AD model cells showed a significant increase in viability; however, only Hp-s1A reduced CTFß levels in these cells. Further analysis of the culture media revealed that Hp-s1A also reduced Aß42 production from AD model cells. The phosphorylation of ERK was elevated and the neurite outgrowth rate was restored with Hp-s1A treatment. In conclusion, the ganglioside analogue Hp-s1A inhibited amyloidogenic processing of APP and promoted neurotrophic activity and survival of AD model cells. Hp-s1A has great potential in AD therapeutic development.