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.

Prevalence of repeat expansions causing autosomal dominant spinocerebellar ataxias in Hokkaido, the northernmost island of Japan.

In Japan, approximately 30% of spinocerebellar degeneration (SCD) is hereditary, and more than 90% of hereditary SCD is autosomal dominant SCD (AD-SCD). We have previously reported the types of AD-SCD in Hokkaido, twice. In this study, we investigated the status of AD-SCD mainly due to repeat expansions, covering the period since the last report. We performed genetic analysis for 312 patients with a clinical diagnosis of SCD, except for multiple system atrophy at medical institutions in Hokkaido between January 2007 and December 2020. The median age at the time of analysis was 58 (1-86) years. Pathogenic variants causing AD-SCD due to repeat expansion were found in 61.5% (192 cases). Spinocerebellar ataxia (SCA) 6 was the most common type in 25.3% (79 cases), followed by Machado-Joseph disease (MJD)/SCA3 in 13.8% (43), SCA1 in 6.4% (20), SCA2 in 5.1% (16), SCA31 in 4.8% (15), dentatorubral-pallidoluysian atrophy in 4.8% (15), SCA7 in 0.6% (2), and SCA8 in 0.6% (2). SCA17, 27B, 36, and 37 were not found. Compared to previous reports, this study found a higher prevalence of SCA6 and a lower prevalence of MJD/SCA3. An increasing number of cases identified by genetic testing, including cases with no apparent family history, accurately revealed the distribution of disease types in Hokkaido.

A retrospective study of autoimmune cerebellar ataxia over a 20-year period in a single institution.

BACKGROUND: It is important to differentiate autoimmune cerebellar ataxia (ACA) from neurodegenerative CA, but this is sometimes difficult. We performed a retrospective study in a single institution in Japan over a 20-year period to reveal the clinical features of ACA. METHODS: Patients with CA as the primary neurological symptom were enrolled from those admitted to the Department of Neurology, Hokkaido University Hospital between April 2002 and March 2022. ACA was diagnosed retrospectively according to the following criteria: (1) CA being the predominant symptom; (2) identification of cancer within 2 years of onset; (3) improvement in cerebellar symptoms following immunotherapy; and (4) ruling out alternative causes of CA. Patients fulfilling criteria (1), (2), and (4) were classified as paraneoplastic cerebellar degeneration (PCD), while those fulfilling (1), (3), and (4) were classified as non-PCD and enrolled as patients with ACA. Neurodegenerative diseases, e.g., multiple system atrophy (MSA), were confirmed retrospectively based on generally used diagnostic criteria and enrolled. Furthermore, the ACA diagnostic criteria proposed by Dalmau and Graus were applied retrospectively to the ACA patients to examine the validity of the diagnoses. RESULTS: Among the 243 patients with CA, 13 were enrolled as ACA; five were PCD and eight were non-PCD. Eight of these cases met the proposed diagnostic criteria by Dalmau and Graus. MSA was the most prevalent disease among CA patients, with 93 cases. The incidence of cerebellar atrophy was significantly lower in ACA (3/13) than in MSA (92/92). Cerebrospinal fluid (CSF) pleocytosis was significantly more frequent in ACA than in MSA (4/13 vs. 2/55, respectively). However, there was no significant difference in the presence of oligoclonal bands, increased protein in CSF, and laterality differences in ataxia. CONCLUSION: ACA was present in ~ 5% of Japanese CA patients. The absence of cerebellar atrophy, despite the presence of CA, strongly supports ACA over MSA. While CSF pleocytosis was observed more often in ACA, the positivity rate was only ~ 30%. Since ACA is treatable, further studies are needed to identify additional clinical features and accurate diagnostic biomarkers.

Case series: Downbeat nystagmus in SCA27B.

BACKGROUND: Spinocerebellar ataxia (SCA) 27B, first reported in late 2022, is caused by the abnormal expansion of GAA repeats in the first intron of the FGF14 gene, which encodes the fibroblast growth factor 14. CASE PRESENTATION: We present two late-onset cases, each manifesting mild cerebellar ataxia accompanied by omnidirectional downbeat nystagmus, which was enhanced in a suspended head position. None of the patients exhibited impaired head impulse or caloric tests. Repeat-primed PCR and targeted long-read nanopore sequence analysis of the FGF14 GAA repeat site identified more than 250 repeats, leading to the diagnosis of SCA27B. DISCUSSION: Downbeat nystagmus is reported to be associated with disturbances in the suppression of the vestibulo-ocular reflex (VOR). Our patients with SCA27B demonstrated downbeat nystagmus, likely due to a disruption of the VOR at the level of the cerebellar cortex, a potentially characteristic clinical feature of SCA27B. We have included video footages of eye movements recorded using Frenzel goggles for these cases. CONCLUSIONS: Omnidirectional downbeat nystagmus may be a distinctive clinical feature of SCA27B.

[A case of Seizure-related 6 homolog like 2 (Sez6l2) antibody-associated autoimmune cerebellar ataxia].

A 35-year-old male patient had been repeatedly involved in car accidents since the age of 34 years, had difficulty in tidying his room, and developed speech and gait disorders. At the first examination, he had a hypophonia and poor gait, but he could talk and walk by himself. His Mini-Mental State Examination (MMSE) score was 23 and mild cognitive impairment was observed. Three months later, neurological findings showed subacute progression to loss of speech, and poor right dominant diadochokinesia and finger tapping, poor finger-nose test and heel-knee test, and loss of ability to stand and walk. Cerebellar atrophy was observed on head MRI, and the patient was positive for Seizure-related 6 homolog like 2 (Sez6l2) antibody, leading to a diagnosis of Sez6l2 antibody-associated autoimmune cerebellar ataxia. Immunotherapy halted the progression of symptoms, and the patient showed slight improvement. Sez6l2 antibody measurement and immunotherapy were considered necessary for subacute progressive cerebellar ataxia of unknown cause.

Gait video-based prediction of unified Parkinson's disease rating scale score: a retrospective study.

BACKGROUND: The diagnosis of Parkinson's disease (PD) and evaluation of its symptoms require in-person clinical examination. Remote evaluation of PD symptoms is desirable, especially during a pandemic such as the coronavirus disease 2019 pandemic. One potential method to remotely evaluate PD motor impairments is video-based analysis. In this study, we aimed to assess the feasibility of predicting the Unified Parkinson's Disease Rating Scale (UPDRS) score from gait videos using a convolutional neural network (CNN) model. METHODS: We retrospectively obtained 737 consecutive gait videos of 74 patients with PD and their corresponding neurologist-rated UPDRS scores. We utilized a CNN model for predicting the total UPDRS part III score and four subscores of axial symptoms (items 27, 28, 29, and 30), bradykinesia (items 23, 24, 25, 26, and 31), rigidity (item 22) and tremor (items 20 and 21). We trained the model on 80% of the gait videos and used 10% of the videos as a validation dataset. We evaluated the predictive performance of the trained model by comparing the model-predicted score with the neurologist-rated score for the remaining 10% of videos (test dataset). We calculated the coefficient of determination (R2) between those scores to evaluate the model's goodness of fit. RESULTS: In the test dataset, the R2 values between the model-predicted and neurologist-rated values for the total UPDRS part III score and subscores of axial symptoms, bradykinesia, rigidity, and tremor were 0.59, 0.77, 0.56, 0.46, and 0.0, respectively. The performance was relatively low for videos from patients with severe symptoms. CONCLUSIONS: Despite the low predictive performance of the model for the total UPDRS part III score, it demonstrated relatively high performance in predicting subscores of axial symptoms. The model approximately predicted the total UPDRS part III scores of patients with moderate symptoms, but the performance was low for patients with severe symptoms owing to limited data. A larger dataset is needed to improve the model's performance in clinical settings.

Lambert-Eaton Myasthenic Syndrome Complicated by anti-GABAB Receptor Encephalitis.

A 74-year-old man experienced diplopia, generalized muscle weakness, and acute respiratory failure. He was diagnosed with Lambert-Eaton myasthenic syndrome (LEMS) and treated with immunotherapy, but no improvement was observed, and additional symptoms, including central apnea and hallucinations, appeared. Subsequent serum and cerebrospinal fluid (CSF) analyses confirmed the presence of GABAB receptor antibodies, indicating the coexistence of autoimmune encephalitis. Although there were no findings of malignancy, it is highly likely that occult small-cell lung carcinoma was present. When atypical symptoms occur in patients with LEMS, it is important to consider the possibility of concomitant autoimmune encephalitis.

Practice of Hereditary ATTR Amyloidosis in Non-endemic Areas of Japan.

Objective Hereditary ATTR (ATTRv) amyloidosis was once an incurable disease; however, in recent years, disease-modifying therapies, such as tafamidis and patisiran, have become available. We herein report the medical care situation in an ATTRv amyloidosis non-endemic area of Japan. Methods We confirmed the information in the medical records of our department and analyzed the data retrospectively. Patients Patients with ATTRv amyloidosis who were treated in our department between 2010 and 2021 were included. Results A total of 15 ATTRv amyloidosis cases (8 men and 7 women) were treated in our department during the study period; 9 patients had a family history, and the transthyretin V30M (p.V50M) gene mutation was present in 66% of cases. The average age of the onset was 57 years old, with 73% of the initial symptoms being dysesthesia and 13% being autonomic dysfunction. Ten patients were treated with tafamidis and nine with patisiran. Although it took a long time to start treatment among our experienced cases, there were some cases in which treatment could be introduced relatively early. Conclusion ATTRv amyloidosis is treatable and should be included in the differential diagnosis of neuropathy so that it can be diagnosed early and introduced into treatment. In the near future, the presymptomatic diagnosis of ATTRv amyloidosis and genetic counseling will become more important.

Clinical features of anti-mitochondrial M2 antibody-positive myositis: case series of 17 patients.

OBJECTIVE: In 2012, a large number of myositis cases with anti-mitochondrial M2 (AMA-M2) antibody, which had well been known as the serological hallmark for primary biliary cholangitis (PBC), were reported in Japan. Recently, some case series from Japan, France, America, China and India have shown that approximately 2.5% to 19.5% of patients with myositis have AMA-M2 antibody. The objective of this study was to clarify the prevalence, clinical features, treatment outcome, and severity determinants of AMA-M2 positive myositis. METHODS: This study was a multicenter observational study. We enrolled patients who were diagnosed with myositis during a ten-year period between 2012 and 2021. RESULTS: Of the total of 185 patients with inflammatory myopathy, 17 patients were positive for AMA-M2 antibody. The typical symptoms were weakness mainly involving paravertebral muscles, weight loss, respiratory failure, and cardiac complications. Thirteen of the 17 patients had cardiac complications. A strong correlation was found between respiratory failure and modified Rankin Scale (mRS) score. A strong correlation was also found between respiratory failure and body weight, indicating that weight loss can be an indicator of potential progression of respiratory failure. Six of the 17 patients were complicated by malignancy. CONCLUSIONS: This study showed significant correlations between % vital capacity (VC), body mass index (BMI), and mRS score in patients with AMA-M2-positive myositis. Immunotherapy often improved CK level and respiratory dysfunction. We therefore propose that %VC and BMI should be monitored as disease indicators in treatment of AMA-M2-positive myositis.

Very-Late-Onset Neuromyelitis Optica Spectrum Disorder in a Patient with Breast Cancer and Parkinson Disease.

Anti-aquaporin-4 (anti-AQP-4) antibody-positive neuromyelitis optica spectrum disorder (NMOSD) is a rare autoimmune disorder resulting in severe, recurrent optic neuritis, transverse myelitis, brain stem syndrome, and other types of neurological involvement. Its median age of onset has been reported to be around 40 years. We report herein a case of very-late-onset NMOSD (76 years of age) and try to promote its awareness as a type of neurological deterioration in elder patients. A 76-year-old woman suffering from Parkinson disease was admitted to our hospital because of consciousness disturbance. Cranial magnetic resonance imaging revealed the presence of fluid-attenuated inversion recovery high-signal-intensity lesions in the right peri- and intralateral ventricle. Part of this lesion and the meninges showed gadolinium enhancement. Physical examination revealed the presence of a tumor in the right breast, which was later diagnosed as invasive ductal carcinoma. In addition, laboratory examinations led to the detection of anti-AQP-4 antibodies in her serum; consequently, the patient was diagnosed as having NMOSD. She received initial pulsed steroid therapy, followed by right mastectomy. Although the patient's consciousness improved significantly, she developed abrupt-onset bilateral leg weakness and multiple longitudinal spinal cord lesions. Additional steroid therapy ameliorated the patient's leg weakness and reduced the swelling of the spinal cord.