Differentiation of speech in Parkinson's disease and spinocerebellar degeneration using deep neural networks.

INTRODUCTION: Assessing dysarthria features in patients with neurodegenerative diseases helps diagnose underlying pathologies. Although deep neural network (DNN) techniques have been widely adopted in various audio processing tasks, few studies have tested whether DNNs can help differentiate neurodegenerative diseases using patients' speech data. This study evaluated whether a DNN model using a transformer architecture could differentiate patients with Parkinson's disease (PD) from patients with spinocerebellar degeneration (SCD) using speech data. METHODS: Speech data were obtained from 251 and 101 patients with PD and SCD, respectively, while they read a passage. We fine-tuned a pre-trained DNN model using log-mel spectrograms generated from speech data. The DNN model was trained to predict whether the input spectrogram was generated from patients with PD or SCD. We used fivefold cross-validation to evaluate the predictive performance using the area under the receiver operating characteristic curve (AUC) and accuracy, sensitivity, and specificity. RESULTS: Average ± standard deviation of the AUC, accuracy, sensitivity, and specificity of the trained model for the fivefold cross-validation were 0.93 ± 0.04, 0.87 ± 0.03, 0.83 ± 0.05, and 0.89 ± 0.05, respectively. CONCLUSION: The DNN model can differentiate speech data of patients with PD from that of patients with SCD with relatively high accuracy and AUC. The proposed method can be used as a non-invasive, easy-to-perform screening method to differentiate PD from SCD using patient speech and is expected to be applied to telemedicine.

Functional outcomes and associated factors of cerebral infarction and intracerebral hemorrhage in an area with aging populations in change over time: Evidence from the Akita Stroke Registry.

AIM: To examine secular change in functional outcomes and associated factors of stroke in a rapidly aging region. METHODS: We retrospectively analyzed cerebral infarction and intracerebral hemorrhage incidence registered cases in the Akita Stroke Registry from 1985 to 2014, divided into three of 10 years each. Functional outcome was defined as good with a modified Rankin scale score of 0-1 and poor with a score of 3-6 at discharge. Mixed effects logistic regression analysis with the location of medical facility as a random effects variable by disease type was used to examine the results. RESULTS: There were 81 254 eligible patients (cerebral infarction: 58 217, intracerebral hemorrhage: 23 037). Age at onset increased over time in both diseases (cerebral infarction: median [interquartile range] age, 70 [63-77] years in 1985-1994 to 77 [69-83] years in 2005-2014; intracerebral hemorrhage: 64 [56-72] years in 1985-1994 to 72 [61-80] years in 2005-2014). Multivariate analysis showed that the odds ratio associated with good outcomes increased over time for cerebral infarction, and cerebral hemorrhage increased in periods 2 and 3 compared with period 1, but decreased from period 2 to period 3. For cerebral infarction, the odds ratios of prior diabetes associated with poor outcomes decreased over time. CONCLUSION: The age at onset increased over time. In cerebral infarction, functional outcomes improved over time, and the association between diabetes and poor outcome declined over time. It was speculated that these results were related to advances in the healthcare system and improved management of vascular risk factors during the study period. Intracerebral hemorrhage improved during the first 20 years, with no apparent improvement thereafter. Geriatr Gerontol Int 2023; 23: 486-492.

Pre- or post-treatment with hepatocyte growth factor prevents glycerol-induced acute renal failure.

Hepatocyte growth factor (HGF) is known to have beneficial effects against damage in various organs, including liver, kidney and lung, in disease models. Previously, we reported that repeated administration of HGF ameliorates renal dysfunction and histological alteration of glycerol-injected rats, an animal model for severe acute renal failure (ARF). In the present study, we investigated in more detail the efficacy of pre- and post-treatment of HGF in this model. ARF was induced by intramuscular injection of glycerol into the hind limbs of male Wistar rats. The efficacy of pre-treatment was studied by intravenous injection of HGF (1 mg/kg) or vehicle 1 and 18 hours prior to glycerol injection. Pre-treatment of HGF dramatically protected glycerol-induced ARF rats against death, and prevented deterioration of biochemical parameters for renal function. We also analyzed expression of heme oxygenase-1 (HO-1), a cytoprotective protein, in kidney of HGF-injected rats. Intravenous administration of HGF enhanced renal expression of HO-1 mRNA from 1 to 3 hours after injection. Next, as a post-treatment study, HGF (1 mg/kg/3 hours) with dopamine was infused into glycerol-induced ARF rats 7 hours after glycerol injection. Intravenous infusion of HGF after ARF onset also ameliorated renal biochemical parameters. These results indicate that pre-treatment of HGF can improve ARF, and induction of HO-1 expression in kidney may be a cause of the protective effect. In addition, post-treatment of HGF with dopamine was also effective against the establishment of ARF.

Ameliorative effect of hepatocyte growth factor on glycerol-induced acute renal failure with acute tubular necrosis.

BACKGROUND/AIMS: Hepatocyte growth factor (HGF), a multi-potent growth factor, is known to promote regeneration of damaged renal epithelial cells. Glycerol injection into rats induces severe acute renal failure (ARF) with ischemia and tubular necrosis, a model which shares many features with human ARF or rhabdomyolysis. We investigated the efficacy of HGF in this glycerol-induced ARF rat model. METHODS: ARF was induced by intramuscular injection of glycerol into the hind limbs of male Wistar rats. HGF (0.25 mg/kg/shot) or vehicle was administered intravenously 1 h before and 1, 3, 5, 8, 24 and 36 h after glycerol injection. Biochemical parameters for serum and urine were measured and histological analyses of the kidneys were performed. We also analyzed endogenous HGF expression and phosphorylation of c-Met/HGF receptor in the kidneys of glycerol-induced ARF rats. RESULTS: Glycerol treatment caused severe ARF which invariably led to death of the rats. Repeated administration of HGF protected rats from death caused by severe ARF. Histological analyses revealed that HGF treatment reduced necrosis of tubular cells in the renal cortex. Serum/urine biochemical parameters also showed that renal dysfunction was improved by HGF administration. Intravenous administration of HGF enhanced phosphorylation of the c-Met/HGF receptor and mitogen-activated protein kinase in the kidney. In the vehicle-treated group the renal endogenous HGF concentration decreased and there was no change in c-Met/HGF receptor phosphorylation. CONCLUSION: These results indicate that HGF effectively accelerated the recovery of renal function and improved survival in glycerol-induced ARF rats.

Hepatocyte growth factor protects functional and histological disorders of HgCl(2)-induced acute renal failure mice.

Hepatocyte growth factor (HGF) enhances proliferation of renal epithelial cells as well as hepatocytes. HGF accelerates recovery from acute renal failure (ARF) in animal models. However, pharmacological profiles of HGF including its action mechanism has not been studied in detail. An HgCl(2)-induced ARF mouse was used in this study to evaluate the efficacy of HGF. Single administrations of recombinant human HGF or vehicle were given to ARF mice 30 min after HgCl(2) injection. Renal function was monitored by measuring serum creatinine, blood urea nitrogen and creatinine clearance. In the ARF mice, there was a deterioration of renal function biochemical parameters and histological evidence of renal damage including acute tubular necrosis of proximal tubules. These were both significantly ameliorated by a single HGF administration. The effect of HGF was noticeable in the early phase of ARF (1 day after onset) when there was no histological evidence of increased labeling indexes in renal tubular epithelial cells. Western blot analysis of the c-Met/HGF receptor showed that tyrosine phosphorylation was enhanced immediately after HGF administration indicating direct activation of renal epithelial cells. HGF prevented increase of apoptotic nuclei with DNA fragmentation in renal epithelial cells which suggests cytoprotective activity of HGF on renal epithelial cells in the ARF mice.