Article
Chemistry, Analytical
Jeremy Watts, Anahita Khojandi, Rama Vasudevan, Fatta B. Nahab, Ritesh A. Ramdhani
Summary: The study proposes using sensor data and machine learning to cluster patients based on their medication responses to enhance treatment planning. A random forest classifier trained on demographic information, MDS-UPDRS-III scores, and PKG time-series data successfully classified subjects of the two most similar clusters with high accuracy and performance.
Article
Computer Science, Artificial Intelligence
Luis Munoz-Saavedra, Elena Escobar-Linero, Lourdes Miro-Amarante, M. Rocio Bohorquez, Manuel Dominguez-Morales
Summary: This work presents the development of a wearable device for capturing physiological signals, the collection of a dataset to induce emotional states, and the development of an automatic classifier based on neural networks. The optimization process results in over 92% accuracy in all cases, significantly improving previous classifiers. The key contributions include the design of a wearable device for non-invasive collection of physiological signals, a data-collection protocol to induce emotional states, and the development and optimization of a machine learning-based system for emotional state classification based on a two-dimensional model.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Aleksandr Talitckii, Anna Anikina, Ekaterina Kovalenko, Aleksei Shcherbak, Oscar Mayora, Olga Zimniakova, Ekaterina Bril, Maxim Semenov, Dmitry Dylov, Andrey Somov
Summary: Our society is seeing a rapid increase in patients with neurodegenerative diseases like Parkinson's disease, with predictions of a potential pandemic within the next two decades. Current research is looking into utilizing machine learning and wearable sensors to identify optimal exercises for efficient detection of PD.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Chemistry, Analytical
Marios G. Krokidis, Georgios N. Dimitrakopoulos, Aristidis G. Vrahatis, Christos Tzouvelekis, Dimitrios Drakoulis, Foteini Papavassileiou, Themis P. Exarchos, Panayiotis Vlamos
Summary: Parkinson's disease is a progressive neurodegenerative disorder associated with dysfunction of dopaminergic neurons in the brain. Early diagnosis of PD is challenging, but sensor-based platforms and machine learning techniques have the potential to improve diagnostic and prognostic capabilities. This work examines the current situation of sensor-based approaches in PD diagnosis and explores the use of machine learning models for personalized risk prediction.
Article
Engineering, Biomedical
Dongli Li, Andre Hallack, Sophie Gwilym, Dongcheng Li, Michele T. Hu, James Cantley
Summary: This study aimed to improve gait in Parkinson's disease patients by applying FOG-initiated vibration cues to the lower-leg via wearable devices. The results showed that responsive cueing significantly improved gait and reduced the occurrence of FOG events. The machine learning algorithm also achieved high accuracy in FOG detection.
BIOMEDICAL ENGINEERING ONLINE
(2023)
Article
Health Care Sciences & Services
Emad Arasteh, Maryam S. Mirian, Wyatt D. Verchere, Pratibha Surathi, Devavrat Nene, Sepideh Allahdadian, Michelle Doo, Kye Won Park, Somdattaa Ray, Martin J. McKeown
Summary: The primary treatment for Parkinson's disease (PD) is supplementation of levodopa (L-dopa). With disease progression, people may experience motor and non-motor fluctuations, whereby the PD symptoms return before the next dose of medication. Early detection of wearing-off before people are consciously aware would be ideal.
JOURNAL OF PERSONALIZED MEDICINE
(2023)
Article
Engineering, Electrical & Electronic
Aleksei Shcherbak, Ekaterina Kovalenko, Andrey Somov
Summary: Parkinson's disease is a common and rapidly growing neurodegenerative disorder that significantly affects patients' physical and social activities. Early diagnosis is challenging due to similar symptoms with other neurodegenerative diseases.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Clinical Neurology
Shinuan Lin, Chao Gao, Hongxia Li, Pei Huang, Yun Ling, Zhonglue Chen, Kang Ren, Shengdi Chen
Summary: This study used machine learning based on gait and postural transition parameters to differentiate early-stage Parkinson's disease (PD) from essential tremor (ET). The results showed that by combining wearable sensors and machine learning, it is possible to successfully distinguish between these two disorders.
JOURNAL OF NEUROLOGY
(2023)
Article
Engineering, Electrical & Electronic
Aleksandr Talitckii, Ekaterina Kovalenko, Aleksei Shcherbak, Anna Anikina, Ekaterina Bril, Olga Zimniakova, Maxim Semenov, Dmitry Dylov, Andrey Somov
Summary: Parkinson's disease is a common neurodegenerative disorder, and current healthcare lacks the means to detect early symptoms and monitor disease progression. This study compares three patient-driven monitoring approaches, using wearable sensor data, video, and handwriting data, analyzed with machine and deep learning methods. Sensor data shows the best performance, while video and handwriting data are more convenient and time-saving.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Review
Clinical Neurology
Ngoc Thai Tran, Huu Nam Tran, Anh Tuan Mai
Summary: In the past 3 years, the focus of medical resources has been on COVID-19, making it challenging to diagnose sleep disorders like OSA due to a shortage of medical staff and equipment. As a result, alternative at-home OSA detection solutions have gained more attention. This study reviews the latest assessment techniques for out-of-center detection of OSA characteristics and the progress in implementing data acquisition, processing, and machine learning for early detection of severe OSA levels.
FRONTIERS IN NEUROLOGY
(2023)
Article
Clinical Neurology
Iria Cabo-Lopez, Alfredo Puy-Nunez, Nuria Redondo-Rafales, Sara Teixeira Baltazar, Beatriz Calderon-Cruz
Summary: This study aims to analyze the correlation between STAT-ON® reports and other assessment tools to identify APD and assess the accuracy of screening tools in APD patients. Although no significant association was found, the CDEPA questionnaire demonstrated the highest sensitivity and VPN values to detect APD candidates. According to the correlation analyses, MANAGE-PD showed the highest degree of concordance with STAT-ON® recordings.
FRONTIERS IN NEUROLOGY
(2023)
Article
Chemistry, Analytical
Eko Sakti Pramukantoro, Akio Gofuku
Summary: Heartbeat monitoring is crucial for early detection of cardiovascular disease. This paper proposes a wearable device-based heartbeat classifier trained using machine learning and deep learning methods, achieving high accuracy for real-time monitoring.
Article
Chemistry, Analytical
Francesco Castelli Gattinara Di Zubiena, Greta Menna, Ilaria Mileti, Alessandro Zampogna, Francesco Asci, Marco Paoloni, Antonio Suppa, Zaccaria Del Prete, Eduardo Palermo
Summary: This study used machine learning to distinguish Parkinson's disease patients from healthy controls based on kinematic measures recorded during dynamic posturography through portable sensors. The results showed that machine learning is a suitable solution for the early identification of balance disorders in Parkinson's disease.
Article
Chemistry, Analytical
Asma Channa, Rares-Cristian Ifrim, Decebal Popescu, Nirvana Popescu
Summary: Parkinson's disease patients commonly experience motor symptoms such as tremor and bradykinesia, with the accuracy of traditional assessment methods still being questioned. This research proposed an objective assessment solution using inertial sensors, which showed promising results in distinguishing patients from healthy controls through the extraction of temporal and spectral features.
Article
Clinical Neurology
Parisa Farzanehfar, Holly Woodrow, Malcolm Horne
Summary: The study found that a high proportion of patients with Parkinson's disease experience motor and non-motor function wearing off, with severity correlated to factors such as disease duration, baseline MDS-UPDRS (motor component), Percent Time in Bradykinesia, Levodopa Equivalent Daily Dose, frequency of Levodopa doses, and age of onset. Patients with more severe wearing off experienced worse motor and non-motor symptoms, resulting in lower quality of life. Quality of life significantly improved in patients with Parkinson's disease when wearing off was treated.
JOURNAL OF NEUROLOGY
(2021)
Article
Clinical Neurology
Seongho Park, Jin Kyung Oh, Jae-Kwan Song, Boseong Kwon, Bum Joon Kim, Jong S. Kim, Dong-Wha Kang, Jun Young Chang, Ji Sung Lee, Sun U. Kwon
Summary: The study showed that transcranial Doppler (TCD) is a reliable tool for evaluating high-risk patent foramen ovale (PFO), with Valsalva maneuver (VM) playing an important role in the assessment. Patients with minimal or no shunt on TCD after VM are unlikely to have high-risk PFO.
JOURNAL OF NEUROIMAGING
(2021)
Article
Computer Science, Artificial Intelligence
Sabyasachi Chakraborty, Satyabrata Aich, Hee-Cheol Kim
Summary: Maintaining the right amount of sleep is crucial for proper health, as inconsistency in sleep can lead to various health issues. This study proposes an algorithm using smart wearables or phones to accurately monitor and score sleep patterns for better health balance. The algorithm outperformed previously developed models, showing its potential in improving health outcomes.
INTERNATIONAL JOURNAL OF SYSTEM DYNAMICS APPLICATIONS
(2021)
Article
Clinical Neurology
Seongho Park, Dong Ah Lee, Ho-Joon Lee, Kyong Jin Shin, Kang Min Park
Summary: This exploratory study aimed to investigate the underlying pathomechanisms of migraine with aura (MA) and migraine without aura (MO) during the interictal phase using connectivity analysis. While there were no differences in global structural connectivity, differences in global functional connectivity were found between patients with MA and MO. Further studies are needed to confirm these findings and explore the potential of using functional connectivity as novel biomarkers in migraine.
ACTA NEUROLOGICA SCANDINAVICA
(2022)
Article
Neurosciences
Sam Yeol Ha, Sung Eun Kim, Kyong Jin Shin, JinSe Park, Kang Min Park, Si Eun Kim, Seongho Park, Dong Ah Lee, David S. Liebeskind
Summary: The study identified a distinct type of internal border zone infarct with accessory lesions in the anteromedial temporal lobe, which is associated with different arterial features but has a similar clinical course to IBZ infarcts without accessory lesions in the ATL. Initial severity at admission and progression after admission were independently associated with poor prognosis in patients with IBZ infarcts. There were no differences in progression rate and clinical outcomes regardless of the presence of lesions in the ATL.
JOURNAL OF STROKE & CEREBROVASCULAR DISEASES
(2021)
Article
Medicine, General & Internal
Sam-Yeol Ha, Yeonah Kang, Ho-Joon Lee, Moonjung Hwang, Jiyeon Baik, Seongho Park
Summary: In this study, we compared the quantification of blood flow velocity using non-contrast 4D flow MRI and transcranial Doppler ultrasound (TCD). We found a strong correlation between intracranial velocity measurements obtained by both modalities. Mean velocities acquired with 4D flow MRI were slightly lower compared to TCD, but 4D flow MRI offers the advantage of three-dimensional blood flow visualization.
Article
Clinical Neurology
Joonwon Lee, Ho-Joon Lee, Sam Yeol Ha, Hyung Chan Kim, Yeonah Kang, Sung-Chul Jin, Seongho Park
Summary: The use of SWI_p imaging technique can accurately measure and locate the thrombus responsible for middle cerebral artery occlusion in patients with acute ischemic stroke, facilitating effective endovascular recanalization therapy (ERT).
JOURNAL OF NEUROIMAGING
(2023)
Article
Engineering, Electrical & Electronic
Mallika Garg, Debashis Ghosh, Pyari Mohan Pradhan
Summary: Dynamic gesture recognition is challenging due to variations in pose, size, and shape of the signer's hand. This letter proposes a Multiscaled Multi-Head Attention Video Transformer Network (MsMHA-VTN) for dynamic hand gesture recognition. The model extracts multiscale features using a pyramid hierarchy and employs different attention dimensions for each head of the transformer, providing attention at the multiscale level. Furthermore, performance using multiple modalities is examined. Extensive experiments demonstrate the superior performance of the proposed MsMHA-VTN with overall accuracy of 88.22% and 99.10% on NVGesture and Briareo datasets, respectively.
IEEE SIGNAL PROCESSING LETTERS
(2023)
Article
Chemistry, Analytical
Min-Ho Park, Sabyasachi Chakraborty, Quang Dao Vuong, Dong-Hyeon Noh, Ji-Woong Lee, Jae-Ung Lee, Jae-Hyuk Choi, Won-Ju Lee
Summary: This study develops anomaly detection algorithms to detect abnormalities of pulsating pressure in hydraulic accumulators. The algorithms include threshold averaging algorithm, support vector machine (SVM), XGBoost, convolutional neural network (CNN), CNN autoencoder, and long short-term memory (LSTM) autoencoder models. The results show that these algorithms and models have high accuracy in detecting anomalies.
Article
Clinical Neurology
Jinse Park, Jin Whan Cho, Jinyoung Youn, Engseok Oh, Wooyoung Jang, Joong-Seok Kim, Yoon-Sang Oh, Hyungyoung Hwang, Chang-Hwan Ryu, Jin-Young Ahn, Jee-Young Lee, Seong-Beom Koh, Jae H. Park, Hee-Tae Kim
Summary: This study investigated the validity and reliability of the Korean-translated version of the ICARS. The results showed that the Korean version of ICARS has good construct validity, internal consistency, and reliability, except for subscale 4 assessing oculomotor disorder which showed moderate internal consistency.
JOURNAL OF MOVEMENT DISORDERS
(2023)
Letter
Clinical Neurology
Hyunyoung Hwang, Jinse Park, Jeong Ik Eun, Kyong Jin Shin, Jongmok Ha, Jinyoung Youn
JOURNAL OF MOVEMENT DISORDERS
(2023)
Article
Engineering, Electrical & Electronic
Parth Sharma, Pyari Mohan Pradhan
Summary: In this paper, a new least mean square (LMS) adaptive filtering algorithm called Amari-Alpha LMS (AALMS) based on Amari-Alpha information theoretic divergence is proposed. The upper bound of step size for tractable analysis is obtained through local convergence and stability analysis. The steady-state performance of the algorithm is analyzed, and the mean-square deviation (MSD) at steady-state is derived. The proposed AALMS algorithm outperforms well-known algorithms in terms of MSD in both stationary and non-stationary scenarios, as shown by simulation results and computational complexity analysis. Comparison of different distance measures derived from Amari-Alpha divergence is also conducted in terms of MSD.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS
(2023)
Article
Neurosciences
Seongho Park, Boseong Kwon, Jin Kyung Oh, Jae-Kwan Song, Ji Sung Lee, Sun U. Kwon
Summary: This study found that CS patients with higher D-dimer levels have a greater risk of recurrent stroke related to PFO, showing a dose-dependent relationship. However, no such effect was observed in patients without PFO.
JOURNAL OF STROKE & CEREBROVASCULAR DISEASES
(2023)
Article
Engineering, Electrical & Electronic
Parth Sharma, Pyari Mohan Pradhan
Summary: Distributed estimation in the wireless sensor network faces challenges with outliers in the desired data. Conventional estimation techniques degrade in performance when outliers are present. This letter proposes a family of incremental distributed estimation techniques based on Bregman divergence, which outperform the incremental BLMS algorithm in terms of mean-square deviation according to simulation results.
IEEE SENSORS LETTERS
(2023)
Article
Clinical Neurology
Jinse Park, Hojin Choi, Jea-Won Jang, Jae-Sung Lim, YoungSoon Yang, Chan-Nyoung Lee, Kee Hyung Park
Summary: The SOC-ADL is a valid and reliable tool for differentiating dementia from mild cognitive impairment based on an assessment of activities of daily living. The scores are strongly correlated with other dementia assessment tools and can be used for screening in elderly individuals.
JOURNAL OF CLINICAL NEUROLOGY
(2021)
Article
Clinical Neurology
Boseong Kwon, Eun-Jae Lee, Seongho Park, Ji Sung Lee, Min Hwan Lee, Daeun Jeong, Dongwhane Lee, Hyuk Sung Kwon, Dae-Il Chang, Jong-Ho Park, Jae-Kwan Cha, Ji Hoe Heo, Sung-Il Sohn, Dong-Eog Kim, Smi Choi-Kwon, Jong S. Kim
Summary: The study found that in the long-term period after stroke, the prevalence and severity of post-stroke depression increased, while those of post-stroke emotional incontinence and post-stroke anger decreased. Social support was found to have an impact on depression and anger, but not on emotional incontinence.