A Deep Learning Approach for Biped Robot Locomotion Interface Using a Single Inertial Sensor.

In this study, we introduce a novel framework that combines human motion parameterization from a single inertial sensor, motion synthesis from these parameters, and biped robot motion control using the synthesized motion. This framework applies advanced deep learning methods to data obtained from an IMU attached to a human subject's pelvis. This minimalistic sensor setup simplifies the data collection process, overcoming price and complexity challenges related to multi-sensor systems. We employed a Bi-LSTM encoder to estimate key human motion parameters: walking velocity and gait phase from the IMU sensor. This step is followed by a feedforward motion generator-decoder network that accurately produces lower limb joint angles and displacement corresponding to these parameters. Additionally, our method also introduces a Fourier series-based approach to generate these key motion parameters solely from user commands, specifically walking speed and gait period. Hence, the decoder can receive inputs either from the encoder or directly from the Fourier series parameter generator. The output of the decoder network is then utilized as a reference motion for the walking control of a biped robot, employing a constraint-consistent inverse dynamics control algorithm. This framework facilitates biped robot motion planning based on data from either a single inertial sensor or two user commands. The proposed method was validated through robot simulations in the MuJoco physics engine environment. The motion controller achieved an error of ≤5° in tracking the joint angles demonstrating the effectiveness of the proposed framework. This was accomplished using minimal sensor data or few user commands, marking a promising foundation for robotic control and human-robot interaction.

Comparison of physician- and self-assessed pubertal onset in Japanese children.

Introduction: Physical examinations to assess pubertal development are challenging in large epidemiological surveys. This study aimed to assess the reliability of judgment of pubertal onset in Japanese children by the original pubertal self-assessment sheet. Methods: A total of 144 children aged 10 or 12 years were recruited between March 2019 and September 2020 from the pediatric endocrine outpatient clinics of participating institutions. Agreement between the physician- and participantassessed pubertal onsets was determined using unweighted kappa (UK) and Gwet's agreement coefficient (AC1). Results: The physician's assessment of pubertal onset was in slight agreement with that of the self-assessment sheet in 10-year-old boys (UK: 0.23 and AC1: 0.14), whereas the agreement between the physician's assessment and self-assessment sheet results was good and the physician's assessment was fair (UK: 0.64 and AC1: 0.94) in 12-year-old boys. The physician's assessment of pubertal onset were in good and moderate agreement with the self-assessment sheet in 10-year-old girls (UK/AC1: 0.74/0.78, respectively). In 12-year-old girls, although it showed poor agreement with UK (0.46), there was a very good agreement with AC1 (0.88). Conclusions: Although self-assessment of breast development was in good agreement with that of the physician's assessment for determining pubertal onset in girls, large-scale epidemiological studies are difficult to conduct for adolescent boys, especially for those in the early pubertal stage.

Deep-Learning-Based Character Recognition from Handwriting Motion Data Captured Using IMU and Force Sensors.

Digitizing handwriting is mostly performed using either image-based methods, such as optical character recognition, or utilizing two or more devices, such as a special stylus and a smart pad. The high-cost nature of this approach necessitates a cheaper and standalone smart pen. Therefore, in this paper, a deep-learning-based compact smart digital pen that recognizes 36 alphanumeric characters was developed. Unlike common methods, which employ only inertial data, handwriting recognition is achieved from hand motion data captured using an inertial force sensor. The developed prototype smart pen comprises an ordinary ballpoint ink chamber, three force sensors, a six-channel inertial sensor, a microcomputer, and a plastic barrel structure. Handwritten data of the characters were recorded from six volunteers. After the data was properly trimmed and restructured, it was used to train four neural networks using deep-learning methods. These included Vision transformer (ViT), DNN (deep neural network), CNN (convolutional neural network), and LSTM (long short-term memory). The ViT network outperformed the others to achieve a validation accuracy of 99.05%. The trained model was further validated in real-time where it showed promising performance. These results will be used as a foundation to extend this investigation to include more characters and subjects.

SOX10 Mutation Screening for 117 Patients with Kallmann Syndrome.

INTRODUCTION: Kallmann syndrome (KS) is a genetically heterogeneous condition characterized by hypogonadotropic hypogonadism (HH) and olfactory dysfunction. Although SOX10, a causative gene for Waardenburg syndrome (WS) and peripheral demyelinating neuropathy, central demyelination, WS, and Hirschsprung disease (PCWH) has previously been implicated in KS, the clinical significance of SOX10 variants as the cause of KS remains uncertain. PATIENTS AND METHODS: A total of 117 patients with KS underwent mutation screening of SOX10 and 14 other causative genes for KS/HH. Rare SOX10 variants were subjected to in silico and in vitro analyses. We also examined clinical data of the patients and their parents with SOX10 variants. RESULTS: Sequence analysis identified 2 heterozygous variants of SOX10 (c.1225G > T, p.Gly409* and c.475C > T, p.Arg159Trp) in patients 1-3, as well as in the parents of patients 1 and 3. The variants were assessed as pathogenic/likely pathogenic, according to the American College of Medical Genomics guidelines. Both variants lacked in vitro transactivating activity for the MITF promoter and exerted no dominant-negative effects. Patients 1-3 carried no pathogenic variants in other genes examined. The patients presented with typical KS, while such features were absent in the parents of patients 1 and 3. None of the 5 variant-positive individuals exhibited hypopigmentation, while 1 and 2 individuals exhibited complete and partial hearing loss, respectively. CONCLUSION: These results provide evidence that SOX10 haploinsufficiency accounts for a small percentage of KS cases. SOX10 haploinsufficiency is likely to be associated with a broad phenotypic spectrum, which includes KS without other clinical features of WS/PCWH.

New Sensor Data Structuring for Deeper Feature Extraction in Human Activity Recognition.

For the effective application of thriving human-assistive technologies in healthcare services and human-robot collaborative tasks, computing devices must be aware of human movements. Developing a reliable real-time activity recognition method for the continuous and smooth operation of such smart devices is imperative. To achieve this, light and intelligent methods that use ubiquitous sensors are pivotal. In this study, with the correlation of time series data in mind, a new method of data structuring for deeper feature extraction is introduced herein. The activity data were collected using a smartphone with the help of an exclusively developed iOS application. Data from eight activities were shaped into single and double-channels to extract deep temporal and spatial features of the signals. In addition to the time domain, raw data were represented via the Fourier and wavelet domains. Among the several neural network models used to fit the deep-learning classification of the activities, a convolutional neural network with a double-channeled time-domain input performed well. This method was further evaluated using other public datasets, and better performance was obtained. The practicability of the trained model was finally tested on a computer and a smartphone in real-time, where it demonstrated promising results.

Pituitary stalk compression by the dorsum sellae: possible cause for late childhood onset growth disorders.

PURPOSE: The purpose of this study was to evaluate the relationship between pituitary stalk compression by the dorsum sellae and clinical or laboratory findings in short stature children. MATERIALS AND METHODS: We retrospectively reviewed magnetic resonance images of the pituitary gland and pituitary stalk for 34 short stature children with growth hormone (GH) deficiency and 24 age-matched control cases. We evaluated the degree of pituitary stalk compression caused by the dorsum sellae. Body height, GH level, pituitary height and onset age of the short stature were statistically compared between cases of pituitary stalk compression with associated stalk deformity and cases without compression. RESULTS: Compression of the pituitary stalk with associated stalk deformity was seen in nine cases within the short stature group. There were no cases observed in the control group. There were no significant differences found for body height, GH level and pituitary height between the cases of pituitary stalk compression with associated stalk deformity and cases without compression. However, a significant difference was seen in the onset age between cases with and without stalk compression. CONCLUSION: Pituitary stalk compression with stalk deformity caused by the dorsum sellae was significantly correlated with late childhood onset of short stature.

Rapid onset of glycogen storage hepatomegaly in a type-2 diabetic patient after a massive dose of long-acting insulin and large doses of glucose.

We present a case of rapid onset of glycogen storage hepatomegaly, caused by a massive dose of long-acting insulin and large doses of glucose, in a type-2 diabetic patient. A 41-year-old man was admitted to our hospital because of hypoglycemia and unconsciousness following subcutaneous administration of 180 units of insulin glargine in a suicide attempt. Despite continuous hypercaloric infusion with additional intravenous glucose injections, hypoglycemia persisted for 36 hours. Although the hepatic function was normal and no hepatomegaly was detected on admission, the liver function tests became abnormal and hepatomegaly was detected on hospitalization day 3. Plain abdominal computed tomography (CT) scanning confirmed liver enlargement, with hepatic CT attenuation markedly elevated at 83.7 HU. Liver biopsy revealed hepatocytic glycogen deposition with edematous degeneration. Based on these findings, the diagnosis was made as rapid onset glycogen storage hepatomegaly caused by administration of a massive dose of long-acting insulin and supplementation with large doses of glucose. With improved glycemic control, the liver function improved, the CT findings of hepatomegaly improved, and the hepatic CT attenuation decreased. Repeat liver biopsy also confirmed almost complete disappearance of glycogen deposits. When hepatic dysfunction or hepatomegaly is detected during treatment with insulin, the possibility of hepatic glycogen deposition should be considered. CT scanning and liver biopsy were useful in diagnosing this case.

Glucocorticoid therapy ameliorated hypoglycemia in insulin-like growth factor-II-producing solitary fibrous tumor.

A 54-year-old man with a past history of multi-operation for solitary fibrous tumor was hospitalized for hypoglycemia. Computed tomography demonstrated multiple tumors in the liver but not in the pancreas. His serum insulin-like growth factor (IGF)-II level was normal, but immunoblot analysis and immunohistochemistry revealed a high molecular weight form of IGF-II in the serum and tumor. Surgical resection was impossible, because of tumor dissemination. Hypoglycemia was repeated despite infusion of glucose and glucagon. Glucocorticoid dramatically and continuously abolished hypoglycemia. In cases of inoperable IGF-II-producing tumor, glucocorticoid therapy may be promising for hypoglycemia.

Cerebral hemorrhage in turner syndrome: a case report.

We report the case of a 21-yr-old female with Turner syndrome associated with cerebral hemorrhage (CH). She was transferred to our hospital for loss of consciousness and was diagnosed with right putaminal hemorrhage. Following surgical removal of the hematoma, she regained consciousness, and her left hemiplegia gradually improved after surgery. Angiography revealed absence of vascular abnormality of the cerebral artery, aorta, and renal arteries. Hypertension was noted on arrival at the hospital and persisted after surgery. A slight hypertensive change was observed in her retinas. Plasma renin activity was elevated (20 ng/ml/h) and renovascular hypertension was suspected. In this patient, CH was suspected to have occurred due to hypertension. This case emphasizes the necessity to carefully monitor the blood pressure in Turner syndrome cases, even during childhood.