2019
Papadopoulos, Alexandros; Kyritsis, Konstantinos; Klingelhoefer, Lisa; Bostanjopoulou, Sevasti; Chaudhuri, Ray K; Delopoulos, Anastasios
Detecting Parkinsonian Tremor from IMU Data Collected In-The-Wild using Deep Multiple-Instance Learning Journal Article
In: IEEE Journal of Biomedical and Health Informatics, 2019.
Abstract | Links | BibTeX | Tags: Accelerometers, Annotations, Diseases, Feature extraction, Rigidity, Sensors, Standards
@article{papadopoulos2019detecting,
title = {Detecting Parkinsonian Tremor from IMU Data Collected In-The-Wild using Deep Multiple-Instance Learning},
author = {Alexandros Papadopoulos and Konstantinos Kyritsis and Lisa Klingelhoefer and Sevasti Bostanjopoulou and Ray K Chaudhuri and Anastasios Delopoulos},
doi = {10.1109/JBHI.2019.2961748},
year = {2019},
date = {2019-01-01},
journal = {IEEE Journal of Biomedical and Health Informatics},
publisher = {IEEE},
abstract = {Parkinson's Disease (PD) is a slowly evolving neurological disease that affects about 1% of the population above 60 years old, causing symptoms that are subtle at first, but whose intensity increases as the disease progresses. Automated detection of these symptoms could offer clues as to the early onset of the disease, thus improving the expected clinical outcomes of the patients via appropriately targeted interventions. This potential has led many researchers to develop methods that use widely available sensors to measure and quantify the presence of PD symptoms such as tremor, rigidity and braykinesia. However, most of these approaches operate under controlled settings, such as in lab or at home, thus limiting their applicability under free-living conditions. In this work, we present a method for automatically identifying tremorous episodes related to PD, based on IMU signals captured via a smartphone device. We propose a Multiple-Instance Learning approach, wherein a subject is represented as an unordered bag of accelerometer signal segments and a single, expert-provided, tremor annotation. Our method combines deep feature learning with a learnable pooling stage that is able to identify key instances within the subject bag, while still being trainable end-to-end. We validate our algorithm on a newly introduced dataset of 45 subjects, containing accelerometer signals collected entirely in-the-wild. The good classification performance obtained in the conducted experiments suggests that the proposed method can efficiently navigate the noisy environment of in-the-wild recordings.},
keywords = {Accelerometers, Annotations, Diseases, Feature extraction, Rigidity, Sensors, Standards},
pubstate = {published},
tppubtype = {article}
}
Parkinson's Disease (PD) is a slowly evolving neurological disease that affects about 1% of the population above 60 years old, causing symptoms that are subtle at first, but whose intensity increases as the disease progresses. Automated detection of these symptoms could offer clues as to the early onset of the disease, thus improving the expected clinical outcomes of the patients via appropriately targeted interventions. This potential has led many researchers to develop methods that use widely available sensors to measure and quantify the presence of PD symptoms such as tremor, rigidity and braykinesia. However, most of these approaches operate under controlled settings, such as in lab or at home, thus limiting their applicability under free-living conditions. In this work, we present a method for automatically identifying tremorous episodes related to PD, based on IMU signals captured via a smartphone device. We propose a Multiple-Instance Learning approach, wherein a subject is represented as an unordered bag of accelerometer signal segments and a single, expert-provided, tremor annotation. Our method combines deep feature learning with a learnable pooling stage that is able to identify key instances within the subject bag, while still being trainable end-to-end. We validate our algorithm on a newly introduced dataset of 45 subjects, containing accelerometer signals collected entirely in-the-wild. The good classification performance obtained in the conducted experiments suggests that the proposed method can efficiently navigate the noisy environment of in-the-wild recordings.
Papadopoulos, Alexandros; Kyritsis, Konstantinos; Bostanjopoulou, Sevasti; Klingelhoefer, Lisa; Chaudhuri, Ray K; Delopoulos, Anastasios
Multiple-instance learning for in-the-wild parkinsonian tremor detection Inproceedings
In: 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 6188–6191, IEEE 2019.
Abstract | Links | BibTeX | Tags: Accelerometers, Diseases, Mathematical model, Sensors, Smart phones, Standards, Training
@inproceedings{papadopoulos2019multiple,
title = {Multiple-instance learning for in-the-wild parkinsonian tremor detection},
author = {Alexandros Papadopoulos and Konstantinos Kyritsis and Sevasti Bostanjopoulou and Lisa Klingelhoefer and Ray K Chaudhuri and Anastasios Delopoulos},
doi = {10.1109/EMBC.2019.8856314},
year = {2019},
date = {2019-01-01},
booktitle = {2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)},
pages = {6188--6191},
organization = {IEEE},
abstract = {Parkinson's Disease (PD) is a neurodegenerative disorder that manifests through slowly progressing symptoms, such as tremor, voice degradation and bradykinesia. Automated detection of such symptoms has recently received much attention by the research community, owing to the clinical benefits associated with the early diagnosis of the disease. Unfortunately, most of the approaches proposed so far, operate under a strictly laboratory setting, thus limiting their potential applicability in real world conditions. In this work, we present a method for automatically detecting tremorous episodes related to PD, based on acceleration signals. We propose to address the problem at hand, as a case of Multiple-Instance Learning, wherein a subject is represented as an unordered bag of signal segments and a single, expert-provided, ground-truth. We employ a deep learning approach that combines feature learning and a learnable pooling stage and is trainable end-to-end. Results on a newly introduced dataset of accelerometer signals collected in-the-wild confirm the validity of the proposed approach.},
keywords = {Accelerometers, Diseases, Mathematical model, Sensors, Smart phones, Standards, Training},
pubstate = {published},
tppubtype = {inproceedings}
}
Parkinson's Disease (PD) is a neurodegenerative disorder that manifests through slowly progressing symptoms, such as tremor, voice degradation and bradykinesia. Automated detection of such symptoms has recently received much attention by the research community, owing to the clinical benefits associated with the early diagnosis of the disease. Unfortunately, most of the approaches proposed so far, operate under a strictly laboratory setting, thus limiting their potential applicability in real world conditions. In this work, we present a method for automatically detecting tremorous episodes related to PD, based on acceleration signals. We propose to address the problem at hand, as a case of Multiple-Instance Learning, wherein a subject is represented as an unordered bag of signal segments and a single, expert-provided, ground-truth. We employ a deep learning approach that combines feature learning and a learnable pooling stage and is trainable end-to-end. Results on a newly introduced dataset of accelerometer signals collected in-the-wild confirm the validity of the proposed approach.