Freezing of gait assessment with inertial measurement units and deep learning: effect of tasks, medication states, and stops.

BACKGROUND: Freezing of gait (FOG) is an episodic and highly disabling symptom of Parkinson's Disease (PD). Traditionally, FOG assessment relies on time-consuming visual inspection of camera footage. Therefore, previous studies have proposed portable and automated solutions to annotate FOG. However, automated FOG assessment is challenging due to gait variability caused by medication effects and varying FOG-provoking tasks. Moreover, whether automated approaches can differentiate FOG from typical everyday movements, such as volitional stops, remains to be determined. To address these questions, we evaluated an automated FOG assessment model with deep learning (DL) based on inertial measurement units (IMUs). We assessed its performance trained on all standardized FOG-provoking tasks and medication states, as well as on specific tasks and medication states. Furthermore, we examined the effect of adding stopping periods on FOG detection performance. METHODS: Twelve PD patients with self-reported FOG (mean age 69.33 ± 6.02 years) completed a FOG-provoking protocol, including timed-up-and-go and 360-degree turning-in-place tasks in On/Off dopaminergic medication states with/without volitional stopping. IMUs were attached to the pelvis and both sides of the tibia and talus. A temporal convolutional network (TCN) was used to detect FOG episodes. FOG severity was quantified by the percentage of time frozen (%TF) and the number of freezing episodes (#FOG). The agreement between the model-generated outcomes and the gold standard experts' video annotation was assessed by the intra-class correlation coefficient (ICC). RESULTS: For FOG assessment in trials without stopping, the agreement of our model was strong (ICC (%TF) = 0.92 [0.68, 0.98]; ICC(#FOG) = 0.95 [0.72, 0.99]). Models trained on a specific FOG-provoking task could not generalize to unseen tasks, while models trained on a specific medication state could generalize to unseen states. For assessment in trials with stopping, the agreement of our model was moderately strong (ICC (%TF) = 0.95 [0.73, 0.99]; ICC (#FOG) = 0.79 [0.46, 0.94]), but only when stopping was included in the training data. CONCLUSION: A TCN trained on IMU signals allows valid FOG assessment in trials with/without stops containing different medication states and FOG-provoking tasks. These results are encouraging and enable future work investigating automated FOG assessment during everyday life.

Wearable-Sensor-Based Weakly Supervised Parkinson's Disease Assessment with Data Augmentation.

Parkinson's disease (PD) is the second most prevalent dementia in the world. Wearable technology has been useful in the computer-aided diagnosis and long-term monitoring of PD in recent years. The fundamental issue remains how to assess the severity of PD using wearable devices in an efficient and accurate manner. However, in the real-world free-living environment, there are two difficult issues, poor annotation and class imbalance, both of which could potentially impede the automatic assessment of PD. To address these challenges, we propose a novel framework for assessing the severity of PD patient's in a free-living environment. Specifically, we use clustering methods to learn latent categories from the same activities, while latent Dirichlet allocation (LDA) topic models are utilized to capture latent features from multiple activities. Then, to mitigate the impact of data imbalance, we augment bag-level data while retaining key instance prototypes. To comprehensively demonstrate the efficacy of our proposed framework, we collected a dataset containing wearable-sensor signals from 83 individuals in real-life free-living conditions. The experimental results show that our framework achieves an astounding 73.48% accuracy in the fine-grained (normal, mild, moderate, severe) classification of PD severity based on hand movements. Overall, this study contributes to more accurate PD self-diagnosis in the wild, allowing doctors to provide remote drug intervention guidance.

Quantifying the potential effects of air pollution reduction on population health and health expenditure in Taiwan.

Air pollution, particularly ambient fine particulate matter (PM2.5) pollution, poses a significant risk to public health, underscoring the importance of comprehending the long-term impact on health burden and expenditure at national and subnational levels. Therefore, this study aims to quantify the disease burden and healthcare expenditure associated with PM2.5 exposure in Taiwan and assess the potential benefits of reducing pollution levels. Using a comparative risk assessment framework that integrates an auto-aggressive integrated moving average model, we evaluated the avoidable burden of cardiopulmonary diseases (including ischemic heart disease, stroke, chronic obstructive pulmonary disease, lung cancer, and diabetes mellitus) and related healthcare expenditure under different air quality target scenarios, including status quo and target scenarios of 15, 10, and 5 µg/m3 reduction in PM2.5 concentration. Our findings indicate that reducing PM2.5 exposure has the potential to significantly alleviate the burden of multiple diseases. Comparing the estimated attributable disease burden and healthcare expenditure between reference and target scenarios from 2022 to 2050, the avoidable disability-adjusted life years were 0.61, 1.83, and 3.19 million for the 15, 10, and 5 µg/m3 target scenarios, respectively. Correspondingly, avoidable healthcare expenditure ranged from US$ 0.63 to 3.67 billion. We also highlighted the unequal allocation of resources and the need for policy interventions to address health disparities due to air pollution. Notably, in the 5 µg/m3 target scenario, Kaohsiung City stands to benefit the most, with 527,368 disability-adjusted life years avoided and US$ 0.53 billion saved from 2022 to 2050. Our findings suggest that adopting stricter emission targets can effectively reduce the health burden and associated healthcare expenditure in Taiwan. Overall, this study provides policymakers in Taiwan with valuable insights for mitigating the negative effects of air pollution by establishing a comprehensive framework for evaluating the co-benefits of air pollution reduction on healthcare expenditure and disease burden.

The effect of multiple interventions to balance healthcare demand for controlling COVID-19 outbreaks: a modelling study.

For controlling recent COVID-19 outbreaks around the world, many countries have implemented suppression and mitigation interventions. This work aims to conduct a feasibility study for accessing the effect of multiple interventions to control the COVID-19 breakouts in the UK and other European countries, accounting for balance of healthcare demand. The model is to infer the impact of mitigation, suppression and multiple rolling interventions for controlling COVID-19 outbreaks in the UK, with two features considered: direct link between exposed and recovered population, and practical healthcare demand by separation of infections. We combined the calibrated model with COVID-19 data in London and non-London regions in the UK during February and April 2020. Our finding suggests that rolling intervention is an optimal strategy to effectively control COVID-19 outbreaks in the UK for balancing healthcare demand and morality ratio. It is better to implement regional based interventions with varied intensities and maintenance periods. We suggest an intervention strategy named as "Besieged and rolling interventions" to the UK that take a consistent suppression in London for 100 days and 3 weeks rolling intervention in other regions. This strategy would reduce the overall infections and deaths of COVID-19 outbreaks, and balance healthcare demand in the UK.

Finding the vulnerable among China's elderly: identifying measures for effective policy targeting.

Facing a rapidly aging population, China has recently started to formulate and implement policies with the aim to provide old-age care. While well-developed old-age care policies commonly include a built-in component that assesses eligibility based on vulnerability, no such process is established in the context of China. Here, based on data from the China Health and Retirement Longitudinal Study collected in both 2011 and 2013, we (a) developed a simple and effective strategy for identifying vulnerable Chinese elderly, which can serve as a basis for policy targeting, and (b) improved the policy relevance and targeting efficiency of this vulnerability measure by including additional health indicators. Our vulnerability measures identify 35% to 46% of Chinese elderly as vulnerable, covering up to 67% of elderly at high risk of death or functional decline. They can serve as an initial screening step for more comprehensive geriatric assessments and enable policy makers to effectively target vulnerable elderly persons in China.

A Hybrid Hierarchical Framework for Free Weight Exercise Recognition and Intensity Measurement with Accelerometer and ECG Data Fusion.

Accurate recognition and effective monitoring of physical activities (PA) in daily life is a goal of many healthcare fields. Existing PA recognition approaches are mostly designed for specific scenarios and often lack extensibility for application in other areas, thereby limiting their usefulness. In this paper, we present a hybrid hierarchical framework that successfully combines two of the main specific-sensor-based PA methods into an effective hybrid solution for general weight exercise applications. The fusion solution separates free weight and non-free weight activities and then further classifies free weight exercises, whilst measuring quantities of repetitions and sets, thus providing a measure of intensity. By fusing accelerometer and electrocardiogram (ECG) data, a One Class Support Vector Machine (OC-SVM) and a Hidden Markov Model (HMM) are applied for exercise recognition and we use semantic inference for determining the intensity of the exercise. The results are based on data from 10 healthy subjects (age: 30 ± 5; BMI: 25 ± 5.5 kg/m^2; body fat: 20.5 ± 5.4), which shows good accuracy in exercise recognition and intensity measurement. This framework can be extended to support additional types of PA recognition in complex applications.

Effects of safety warnings and risk management plan for Thiazolidinediones in Taiwan.

PURPOSE: To evaluate changes in thiazolidinedione use and quality of prescription following safety warnings for thiazolidinediones and cardiac risk in 2007, Risk Management Plan (RMP) policy for rosiglitazone in 2010, and warning for pioglitazone and bladder cancer risk in 2010 in Taiwan. METHODS: We obtained 2003-2011 claims data from Taiwan's National Health Insurance Research Database. Using an interrupted time series design and segmented regression, we estimated changes in monthly prescribing rates for thiazolidinediones among all and prevalent diabetes patients with and without cardiovascular disease history (CV history). We also compared time to prescription of thiazolidinediones among new diabetes patients with CV history before and after each regulatory action using survival analysis. RESULTS: Among prevalent patients with and without CV history, the prescribing rates of rosiglitazone decreased 36.88% and 28.92% after safety warnings in 2007 respectively. Pioglitazone prescriptions increased 13% among patients with CV history, but no changes were detected among patients without CV history. After rosiglitazone's RMP policy in 2010, large reductions in prescriptions were observed in patients with CV history (-101.67%) and those without CV history (-88.04%). Among new diabetes patients with CV history, cardiac safety warnings in 2007 significantly delayed the prescription of rosiglitazone, but no significant change was found for pioglitazone. CONCLUSIONS: The Taiwan FDA regulatory actions for thiazolidinediones communicated possible risks of cardiac events and bladder cancer. Different safety regulatory actions had differential impacts on the use of rosiglitazone and pioglitazone and the quality use of these drugs among the high-risk patients.

Acetazolamide-challenged perfusion magnetic resonance imaging for assessment of cerebrovascular reserve capacity in patients with symptomatic middle cerebral artery stenosis: comparison with technetium-99m-hexamethylpropyleneamine oxime single-photon emission computed tomography.

OBJECTIVES: Acetazolamide-challenged perfusion magnetic resonance imaging (MRI) has been shown as a method for assessment of cerebrovascular reserve (CVR) capacity in patients with atherosclerotic steno-occlusive disease of internal carotid artery. We have assessed the feasibility of the acetazolamide-challenged perfusion MRI for evaluating CVR in symptomatic patients with severe middle cerebral artery (MCA) stenosis (≥70%) by comparison with the acetazolamide-challenged technetium-99m-hexamethylpropyleneamine oxime (HMPAO) single-photon emission computed tomography (SPECT). METHODS: Seventeen prospectively enrolled patients with symptomatic unilateral MCA stenosis underwent technetium-99m-hexamethylpropyleneamine oxime SPECT and perfusion MRI without and with acetazolamide challenge, respectively. Acetazolamide-challenged SPECT and perfusion MRI were compared quantitatively by Region of interest (ROI) analysis. RESULTS: At all ROIs, there were no significant differences in percent change between SPECT and perfusion MRI. Patients with impaired CVR showed significant decreases in the percent changes of respective cerebral blood flow (P=.016) and respective cerebral blood volume (P=.029). CONCLUSION: Acetazolamide-challenged perfusion MRI is feasible for evaluating CVR in symptomatic patients with severe MCA stenosis quantitatively.

Assessment of arterial characteristics in human atherosclerosis by extracting optical properties from polarization-sensitive optical coherence tomography.

In this research, we assessed arterial characteristics in human atherosclerosis by determining both scattering and birefringence properties of vessel tissue from polarization-sensitive optical coherence tomography (PS-OCT) images. We demonstrated that optical characteristics including scattering coefficient (mu(s)), effective anisotropy factor (g(eff)), birefringence coefficient (Delta n), and fast-axis angle (beta) of normal and different kinds of atherosclerotic lesions in human vessels can be quantitatively described. Based on our findings, a quantitative PS-OCT image criterion for plaque characterization was constructed.