Prediction of Parkinson's disease by transcranial sonography-based deep learning.

OBJECTIVES: Transcranial sonography has been used as a valid neuroimaging tool to diagnose Parkinson's disease (PD). This study aimed to develop a modified transcranial sonography (TCS) technique based on a deep convolutional neural network (DCNN) model to predict Parkinson's disease. METHODS: This retrospective diagnostic study was conducted using 1529 transcranial sonography images collected from 854 patients with PD and 775 normal controls admitted to the Second Affiliated Hospital of Soochow University (Suzhou, Jiangsu, China) between September 2019 and May 2022. The data set was divided into training cohorts (570 PD patients and 541 normal controls), and the validation set (184 PD patients and 234 normal controls). Using these datasets, we developed four different DCNN models (ResNet18, ResNet50, ResNet152, and DenseNet121). We then assessed their diagnostic performance, including the area under the receiver operating characteristic (AUROC) curve, specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV), and F1 score and compared with traditional diagnostic criteria. RESULTS: Among the 1529 TCS images, 570 PD patients and 541 normal controls from 4 of 6 sonographers of the TCS team were selected as the training cohort, and 184 PD patients and 234 normal controls from the other 2 sonographers were chosen as the validation cohort. There were no sex and age differences between PD patients and normal control subjects in the training and validation cohorts (P values > 0.05). All DCNN models achieved good performance in distinguishing PD patients from normal control subjects on the validation datasets, with diagnostic AUROCs and accuracy of 0.949 (95% CI 0.925, 0.965) and 86.60 for the RestNet18 model, 0.949 (95% CI 0.929, 0.971) and 87.56 for ResNet50, 0.945 (95% CI 0.931, 0.969) and 88.04 for ResNet152, 0.953 (95% CI 0.935, 0.971) and 87.80 for DenseNet121, respectively. On the other hand, the diagnostic accuracy of the traditional diagnostic method was 82.30. The accuracy of all DCNN models was higher than that of traditional diagnostic method. Moreover, the 5k-fold cross-validation results in train datasets showed that these DCNN models are robust. CONCLUSION: The developed transcranial sonography-based DCNN models performed better than traditional diagnostic criteria, thus improving the sonographer's accuracy in diagnosing PD.

Correlation Between Substantia Nigra Hyperechogenicity and Iron Metabolism in the Postural Instability Gait Difficulty Subtype of Parkinson's Disease.

OBJECTIVE: The correlation between substantia nigra (SN) hyperechogenicity on transcranial sonography (TCS) and serum iron metabolism parameters in patients with the postural instability gait difficulty (PIGD) subtype of Parkinson's disease (PD) was investigated so as to explore the pathological mechanism of SN hyperechogenicity. METHODS: The study enrolled 95 PIGD patients recruited by the Parkinson's Disease Specialty in the Second Affiliated Hospital of Soochow University during June 2019-2021. On the basis of the TCS results, the PIGD patients were assigned to the PD with SN hyperechogenicity (SN+) group (n = 60) and PD without SN hyperechogenicity (SN-) group (n = 35). Meanwhile, 49 sex- and age-matched healthy individuals were included in the control group. All participants underwent blood tests. Differences in the iron metabolism parameters among the three groups and the correlation between SN hyperechogenicity and serum iron metabolism parameters were analyzed. RESULTS: Serum ferritin, ceruloplasmin and transferrin levels were lower in the SN+ and SN- groups than in the control group (all p values <0.001). The serum ceruloplasmin level was lower in the SN+ group (0.23 [0.20, 0.25] g/L) than in the SN- group (0.25 [0.22, 0.29] g/L) (p = 0.001), and the proportion of patients with an abnormal ceruloplasmin level was higher in the SN+ group than in the SN- group (43.3% [26/60] vs. 14.3% [5/35], χ2 = 8.484, p = 0.004). Moreover, the SN hyperechogenicity area was negatively correlated with the serum transferrin level (r = -0.428, p < 0.001). CONCLUSION: Decreased serum ceruloplasmin levels may be associated with SN hyperechogenicity development in PIGD patients. The SN hyperechogenicity area is negatively correlated with the serum transferrin level.

Therapeutic effects of menstrual blood-derived endometrial stem cells on mouse models of streptozotocin-induced type 1 diabetes.

BACKGROUND: Type 1 diabetes (T1D), a chronic metabolic and autoimmune disease, seriously endangers human health. In recent years, mesenchymal stem cell (MSC) transplantation has become an effective treatment for diabetes. Menstrual blood-derived endometrial stem cells (MenSC), a novel MSC type derived from the decidual endometrium during menstruation, are expected to become promising seeding cells for diabetes treatment because of their noninvasive collection procedure, high proliferation rate and high immunomodulation capacity. AIM: To comprehensively compare the effects of MenSC and umbilical cord-derived MSC (UcMSC) transplantation on T1D treatment, to further explore the potential mechanism of MSC-based therapies in T1D, and to provide support for the clinical application of MSC in diabetes treatment. METHODS: A conventional streptozotocin-induced T1D mouse model was established, and the effects of MenSC and UcMSC transplantation on their blood glucose and serum insulin levels were detected. The morphological and functional changes in the pancreas, liver, kidney, and spleen were analyzed by routine histological and immunohistochemical examinations. Changes in the serum cytokine levels in the model mice were assessed by protein arrays. The expression of target proteins related to pancreatic regeneration and apoptosis was examined by western blot. RESULTS: MenSC and UcMSC transplantation significantly improved the blood glucose and serum insulin levels in T1D model mice. Immunofluorescence analysis revealed that the numbers of insulin+ and CD31+ cells in the pancreas were significantly increased in MSC-treated mice compared with control mice. Subsequent western blot analysis also showed that vascular endothelial growth factor (VEGF), Bcl2, Bcl-xL and Proliferating cell nuclear antigen in pancreatic tissue was significantly upregulated in MSC-treated mice compared with control mice. Additionally, protein arrays indicated that MenSC and UcMSC transplantation significantly downregulated the serum levels of interferon γ and tumor necrosis factor α and upregulated the serum levels of interleukin-6 and VEGF in the model mice. Additionally, histological and immunohistochemical analyses revealed that MSC transplantation systematically improved the morphologies and functions of the liver, kidney, and spleen in T1D model mice. CONCLUSION: MenSC transplantation significantly improves the symptoms in T1D model mice and exerts protective effects on their main organs. Moreover, MSC-mediated angiogenesis, antiapoptotic effects and immunomodulation likely contribute to the above improvements. Thus, MenSC are expected to become promising seeding cells for clinical diabetes treatment due to their advantages mentioned above.