Article
Engineering, Biomedical
Min Chen, Zhanfang Sun, Tao Xin, Yan Chen, Fei Su
Summary: This study implemented an accurate and objective walking detection algorithm for patients with Parkinson's disease (PD) using motion data collected from inertial measurement units. The results showed that the waist sensor had the best classification performance, and visual interpretation-based optimization of the intelligent classification model reduced raw data processing costs while ensuring its performance. This research contributes to the intelligent diagnosis of PD.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Atul Chaudhary, Hari Prabhat Gupta, K. K. Shukla
Summary: This paper proposes an online system that recognizes activities of daily living (ADL) in real-time, considering the long-tailed class distribution. The system generates features using conventional and deep learning, and utilizes ensemble technique for feature concatenation. It minimizes a loss function that addresses the class imbalance problem and enhances the discriminative power of deep learning features.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Engineering, Electrical & Electronic
Ekaterina Kovalenko, Aleksei Shcherbak, Andrey Somov, Ekaterina Bril, Olga Zimniakova, Maksim Semenov, Aleksandr Samoylov
Summary: This study investigates the combination of multiple data sources for the diagnosis of Parkinson's Disease using machine learning algorithms. The results show that using both wearable sensors and video data can improve the accuracy of diagnosis and open up new possibilities for patient-driven data acquisition and healthcare.
IEEE SENSORS JOURNAL
(2022)
Article
Chemistry, Analytical
Roozbeh Atri, Kevin Urban, Barbara Marebwa, Tanya Simuni, Caroline Tanner, Andrew Siderowf, Mark Frasier, Magali Haas, Lee Lancashire
Summary: This study demonstrates the feasibility of using wearable sensor data and human activity recognition techniques to accurately detect the presence or absence of Parkinson's disease, achieving high classification accuracies.
Article
Chemistry, Analytical
Samer A. A. Mohamed, Uriel Martinez-Hernandez
Summary: This paper proposes a light-weight architecture for activity recognition using wearable sensors. Time-domain features of the sensors data are extracted systematically, and a small high-speed artificial neural network is used for activity recognition. The experiments demonstrate that the proposed architecture can achieve accurate and fast activity recognition with reduced computational complexity.
Article
Engineering, Electrical & Electronic
Majd Saleh, Manuel Abbas, Regine Le Bouquin Jeannes
Summary: This article discusses the limitations of wearable fall detection systems, focusing on issues related to datasets and sensors, and proposes a comprehensive data acquisition system. The FallAllD dataset collects a large amount of data and can effectively evaluate the performance of deep learning and classical algorithms.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Aiguo Wang, Shenghui Zhao, Chundi Zheng, Jing Yang, Guilin Chen, Chih-Yung Chang
Summary: This study proposes two methods for processing binary sensor time-series firings for home human activity recognition and evaluates their impact on deep learning models. Results show that the performance of both deep learning and shallow models is closely associated with the raw signal encodings, with one-dimensional convolutional neural networks demonstrating superiority in generalization across scenarios.
IEEE SENSORS JOURNAL
(2021)
Article
Chemistry, Analytical
Angela R. Weston, Brian J. Loyd, Carolyn Taylor, Carrie Hoppes, Leland E. Dibble
Summary: This study aimed to determine the ability of wearable sensors and data processing algorithms to discern motion restrictions during daily living activities. The results showed that wearable sensors accurately captured significant differences in head and trunk kinematics, including rotational velocity, amplitude, and head-trunk coupling, between restricted and non-restricted conditions. These findings support the ecological validity of using wearable sensors to quantify movement alterations during real-world scenarios.
Article
Engineering, Biomedical
Nils Roth, Arne Kuederle, Martin Ullrich, Till Gladow, Franz Marxreiter, Jochen Klucken, Bjoern M. Eskofier, Felix Kluge
Summary: In this study, a comprehensive free-living evaluation dataset was used to assess the segmentation performance of a new HMM-based method compared to DTW on stride data from Parkinson's Disease patients. The results showed that the proposed HMM had a mean F1-score of 92.1% and outperformed the DTW approach significantly. Segmentation performance was found to be dependent on the number of strides within walking bouts, with shorter bouts resulting in worse performance. The HMM also demonstrated promising generalizability when trained on at-lab data only.
JOURNAL OF NEUROENGINEERING AND REHABILITATION
(2021)
Article
Engineering, Electrical & Electronic
Ekaterina Kovalenko, Aleksandr Talitckii, Anna Anikina, Aleksei Shcherbak, Olga Zimniakova, Maksim Semenov, Ekaterina Bril, Dmitry V. Dylov, Andrey Somov
Summary: Parkinson's Disease is currently the fastest growing neurodegenerative disease and impacts patients' quality of life. This study presents a second opinion system based on video analysis and machine learning methods to assist in accurate diagnosis and avoid misdiagnosis of PD as essential tremor.
IEEE SENSORS JOURNAL
(2021)
Article
Clinical Neurology
Carlos Perez-Lopez, Jorge Hernandez-Vara, Nuria Caballol, Angels Bayes, Mariateresa Buongiorno, Nuria Lopez-Ariztegui, Alexandre Gironell, Jose Lopez-Sanchez, Juan Carlos Martinez-Castrillo, M. Alvarez Sauco, Lydia Lopez-Manzanares, Sonia Escalante-Arroyo, David A. Perez-Martinez, Alejandro Rodriguez-Molinero
Summary: This study analyzed the agreement between Hauser diaries and the STAT-ON sensor and found a moderate concordance between the two tools. The correlations between different UPDRS indices were better with the STAT-ON than with the Hauser diary.
FRONTIERS IN NEUROLOGY
(2022)
Review
Environmental Sciences
Nicola Camp, Martin Lewis, Kirsty Hunter, Julie Johnston, Massimiliano Zecca, Alessandro Di Nuovo, Daniele Magistro
Summary: The use of technology to monitor ADL of older adults is increasingly common, but there is significant variation in the recognition of ADL, how ADL are defined, and the types of technology utilized in monitoring systems.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2021)
Article
Engineering, Electrical & Electronic
Satya P. Singh, Madan Kumar Sharma, Aime Lay-Ekuakille, Deepak Gangwar, Sukrit Gupta
Summary: This study introduces a deep neural network architecture for accurate decoding of human activity from multiple sensors, which selects and learns important time points using a self-attention mechanism. Results demonstrate significant performance improvement over previous deep networks and state-of-the-art methods.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Information Systems
Yordan P. Raykov, Luc J. W. Evers, Reham Badawy, Bastiaan R. Bloem, Tom M. Heskes, Marjan J. Meinders, Kasper Claes, Max A. Little
Summary: In this study, a principled modeling approach for free-living gait analysis was developed to support health predictions and clinical diagnosis. Using a dataset of PD patients and controls, the framework's effectiveness in detecting gait and predicting medication-induced fluctuations in PD patients was demonstrated. The approach was shown to be robust to varying sensor locations.
IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
(2021)
Article
Computer Science, Information Systems
Batsergelen Myagmar, Jie Li, Shigetomo Kimura
Summary: In this research, a novel heterogeneous transfer learning algorithm called HDLAL is proposed to derive domain invariant feature representation space from cross-domain data distributions and train a multi-label classifier using ensemble classification algorithm in the new feature space to predict target domain labels. Experimental results on real-world smart home datasets demonstrate that the HDLAL algorithm outperforms common direct learning approaches in predicting labels of activities of daily living (ADL).
IEEE TRANSACTIONS ON BIG DATA
(2021)
Article
Clinical Neurology
Benjamin J. Dorfman, Joohi Jimenez-Shahed
Summary: Deutetrabenazine is effective in reducing dyskinesia in TD, but ongoing studies are exploring drug selection and cost-effectiveness among existing VMAT2 inhibitors. Other areas of investigation include novel anti-dyskinetic agents and the use of deep brain stimulation.
EXPERT REVIEW OF NEUROTHERAPEUTICS
(2021)
Article
Engineering, Biomedical
Behnaz Ghoraani, Lillian N. Boettcher, Murtadha D. Hssayeni, Amie Rosenfeld, Magdalena Tolea, James E. Galvin
Summary: This study introduces a diagnostic algorithm based on gait and machine learning to detect MCI and AD. By collecting and analyzing gait data from participants, the algorithm achieved an accuracy of 78% in distinguishing between healthy, MCI, and AD individuals. This novel approach demonstrates the potential of gait-based cognitive screening and machine learning in early detection and intervention for cognitive impairment.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2021)
Article
Engineering, Biomedical
Murtadha D. Hssayeni, Joohi Jimenez-Shahed, Michelle A. Burack, Behnaz Ghoraani
Summary: This study developed a novel algorithm using wearable technology to continuously estimate the severity of Parkinson's disease at home. The algorithm was successful in estimating PD severity scores without requiring patients to perform specific tasks, showing potential for effective disease management.
BIOMEDICAL ENGINEERING ONLINE
(2021)
Article
Multidisciplinary Sciences
Murtadha D. Hssayeni, Joohi Jimenez-Shahed, Michelle A. Burack, Behnaz Ghoraani
Summary: This study developed a sensor-based assessment system utilizing a deep recurrent model to estimate levodopa-induced dyskinesia severity in Parkinson's disease patients. The model showed a high correlation with expert ratings and consistent performance across different activities of daily living.
SCIENTIFIC REPORTS
(2021)
Editorial Material
Mathematical & Computational Biology
Yunfeng Wu, Sridhar Krishnan, Behnaz Ghoraani
Summary: Biomedical signal processing and data analysis are crucial in advanced medical expert systems. Signal processing tools improve signal quality, while data analysis techniques reduce redundancy and extract important features related to pathological conditions. Recent computational methods have greatly enhanced the efficiency and accuracy of diagnosis and decision-making in medical fields.
COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE
(2022)
Editorial Material
Chemistry, Analytical
Yunfeng Wu, Behnaz Ghoraani
Article
Engineering, Electrical & Electronic
Rodrigo Capobianco Guido, Tulay Adali, Emil Bjoernson, Laure Blanc-Feraud, Ulisses Braga-Neto, Behnaz Ghoraani, Christian Jutten, Alle-Jan Van der Veen, Hong Vicky Zhao, Xiaoxing Zhu
Summary: The IEEE Signal Processing Society has provided 75 years of service to the signal processing community, making significant contributions to technological advancement.
IEEE SIGNAL PROCESSING MAGAZINE
(2023)
Article
Clinical Neurology
Hubert H. Fernandez, Per Odin, David G. Standaert, Tove Henriksen, Joohi Jimenez-shahed, Sharon Metz, Ali Alobaidi, Connie H. Yan, Pavnit Kukreja, Juan Carlos Parra, Jorge Zamudio, Koray Onuk, Jack Wright, Angelo Antonini
Summary: This study categorized Parkinson's disease patients using MANAGE-PD and examined their quality of life, healthcare resource utilization, and discussions about device-aided therapies. The findings showed that patients with poorly controlled symptoms and eligible for device-aided therapies had higher burden on healthcare resources but a significant portion did not discuss device-aided therapies with providers.
PARKINSONISM & RELATED DISORDERS
(2023)
Review
Clinical Neurology
Nbaa Masood, Joohi Jimenez-Shahed
Summary: Motor complications and the lack of disease modifying therapy counterbalance the discovery of levodopa as the cornerstone of PD treatment. Pathological changes in the striatum lead to OFF episodes caused by non-physiologic stimulation of dopamine receptors. Current research focuses on optimizing dopaminergic stimulation and implementing rescue therapies.
NEUROPSYCHIATRIC DISEASE AND TREATMENT
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Murtadha D. Hssayeni, Arash Andalib, Rishabh Singh, Diego Pava, Kan Li, Steven Borzak, Robert Chait, Kaustubh Kale
Summary: ECG signals play a crucial role in diagnosing cardiovascular diseases (CVD). This study presents an algorithm that successfully estimates the key fiducial points in ECG signals using both rule-based method and deep convolutional neural network (CNN). The algorithm was evaluated using multiple datasets and achieved a high accuracy rate, demonstrating its effectiveness in automatic localization of ECG fiducial points.
2022 21ST IEEE INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS, ICMLA
(2022)
Article
Engineering, Biomedical
Mahmoud Seifallahi, Afsoon Hasani Mehraban, James E. Galvin, Behnaz Ghoraani
Summary: This study investigated the feasibility of using a balance and walking assessment tool to detect Alzheimer's disease (AD) and healthy control (HC). The results showed that using signal processing and statistical analysis, along with a support vector machine classifier, could accurately distinguish between the two groups, demonstrating the potential of this method as a new quantitative tool for detecting AD.
IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING
(2022)
Article
Computer Science, Theory & Methods
Murtadha D. Hssayeni, Arjuna Chala, Roger Dev, Lili Xu, Jesse Shaw, Borko Furht, Behnaz Ghoraani
Summary: This study successfully conducted short-term forecasting of COVID-19 outbreaks using mobility data and a deep learning model, achieving good predictive accuracy and capturing the effects of government responses and age demographics on the disease spread.
JOURNAL OF BIG DATA
(2021)