RESUMO
The present study aimed to explore the diagnostic values of neutrophil-lymphocyte ratio (NLR) and microRNA (miR)-141 in patients with osteoarthritis and their association with the severity of knee osteoarthritis. In total 142 patients with osteoarthritis (the study group) admitted to Shanghai TCM-Integrated Hospital, Shanghai University of TCM from January 2017 to January 2019 and 150 healthy controls (the control group) were enrolled in the present study. NLR and miR-141 in peripheral blood and their diagnostic values for osteoarthritis were compared between the two groups. The two indicators in the study group were significantly increased (P<0.001), and their combined detection had a better diagnostic value for the disease (P<0.001). Moreover, they were closely associated to the progression of the disease and were independent risk factors (P<0.001). To sum up, NLR and miR-141 were significantly increased in the peripheral blood of patients with osteoarthritis. Their combined detection exhibited a good diagnostic value for the disease and may become a potential therapeutic target osteoarthritis in the future.
RESUMO
The study aimed to explore the application value of lumbar Magnetic Resonance Imaging (MRI) images processed by artificial intelligence algorithms in evaluating the efficacy of chinkuei shin chewan decoction (a traditional Chinese medicine to nourish the kidney) in the treatment of lumbar spinal stenosis (LSS). Specifically, 110 LSS patients admitted to the hospital were selected as the research subjects. They were randomly divided into the control group (n = 55) and experimental group (n = 55) according to different treatment methods. The control group was treated with traditional medicine, and the experimental group additionally took chinkuei shin chewan decoction on its basis. Based on the traditional U-net algorithm, a U-net registration algorithm based on artificial intelligence was designed by introducing the information entropy theory, and the algorithm was applied to the lumbar MRI image evaluation of LSS patients. Compared with the traditional U-net algorithm, the artificial intelligence-based U-net registration algorithm had a decreased noise level (P < 0.05), the Jaccard (J) value (0.84) and the Dice value (0.93) increased significantly versus the traditional algorithm (J = 0.63, Dice = 0.81), and the characteristics of the image were more accurate. Before treatment, the Oswestry Disability Index (ODI) scores of the experimental group and the control group were 44.32 ± 6.45 and 43.32 ± 5.45, respectively. After treatment, the ODI scores of the two groups were 10.21 ± 5.05 and 17.09 ± 5.23, respectively. Both showed significant improvement, while the improvement of the experimental group was more obvious than that of the control group (P < 0.05). The overall effective rates of the two groups of patients were 96.44% and 82.47%, respectively, and the experimental group was significantly higher than the control group (P < 0.05). Under the U-net registration algorithm based on artificial intelligence, the diagnostic accuracy of lumbar MRI in the experimental group was 94.45%, significantly higher than 67.5% before the introduction of the algorithm (P < 0.05). In conclusion, chinkuei shin chewan decoction are effective for the treatment of LSS, and lumbar MRI based on the artificial intelligence U-net registration algorithm can evaluate the efficacy of LSS well and is worthy of promotion.
