More Like this

1. Deep neural networks for wearable sensor-based activity recognition in Parkinson’s disease: investigating generalizability and model complexity

   https://doi.org/10.1186/s12938-024-01214-2  

   Davidashvilly, Shelly ; Cardei, Maria ; Hssayeni, Murtadha ; Chi, Christopher ; Ghoraani, Behnaz ( February 2024 , BioMedical Engineering OnLine)

   The research gap addressed in this study is the applicability of deep neural network (NN) models on wearable sensor data to recognize different activities performed by patients with Parkinson’s Disease (PwPD) and the generalizability of these models to PwPD using labeled healthy data.

   The experiments were carried out utilizing three datasets containing wearable motion sensor readings on common activities of daily living. The collected readings were from two accelerometer sensors. PAMAP2 and MHEALTH are publicly available datasets collected from 10 and 9 healthy, young subjects, respectively. A private dataset of a similar nature collected from 14 PwPD patients was utilized as well. Deep NN models were implemented with varying levels of complexity to investigate the impact of data augmentation, manual axis reorientation, model complexity, and domain adaptation on activity recognition performance.

   A moderately complex model trained on the augmented PAMAP2 dataset and adapted to the Parkinson domain using domain adaptation achieved the best activity recognition performance with an accuracy of 73.02%, which was significantly higher than the accuracy of 63% reported in previous studies. The model’s F1 score of 49.79% significantly improved compared to the best cross-testing of 33.66% F1 score with only data augmentation and 2.88% F1 score without data augmentation or domain adaptation.

   These findings suggest that deep NN models originating on healthy data have the potential to recognize activities performed by PwPD accurately and that data augmentation and domain adaptation can improve the generalizability of models in the healthy-to-PwPD transfer scenario. The simple/moderately complex architectures tested in this study could generalize better to the PwPD domain when trained on a healthy dataset compared to the most complex architectures used. The findings of this study could contribute to the development of accurate wearable-based activity monitoring solutions for PwPD, improving clinical decision-making and patient outcomes based on patient activity levels.

    

   more » « less
2. A phase 1 study of entinostat in children and adolescents with recurrent or refractory solid tumors, including CNS tumors: Trial ADVL1513, Pediatric Early Phase‐Clinical Trial Network (PEP‐CTN)

   https://doi.org/10.1002/pbc.28892  

   Bukowinski, Andrew ; Chang, Bill ; Reid, Joel M. ; Liu, Xiaowei ; Minard, Charles G. ; Trepel, Jane B. ; Lee, Min‐Jung ; Fox, Elizabeth ; Weigel, Brenda J. ( January 2021 , Pediatric Blood & Cancer)

   Entinostat is an oral small molecule inhibitor of class I histone deacetylases (HDAC), which has not previously been evaluated in pediatrics. We conducted a phase I trial to determine the maximum tolerated dose/recommended phase 2 dose (MTD/RP2D), toxicity profile, pharmacokinetics (PK), and pharmacodynamics (PD) of entinostat in children with relapsed or refractory solid tumors including central nervous system (CNS) malignancies.

   A rolling six dose escalation design evaluated two dose levels. Entinostat oral tablet formulation was administered once per week, four doses per 28‐day cycle. PK and PD studies were performed.

   Twenty‐one eligible patients’ median (range) age was 14 years (6‐20). Six subjects were treated at 3 mg/m²dose level and 15 were treated in 4 mg/m²dose level. The study included patients with CNS tumors (n = 12), sarcomas (n = 6), or other solid tumors (n = 3). Eight patients were not fully evaluable for toxicity due to progression of disease prior to receiving the required percentage of protocol therapy. No cycle one dose‐limiting toxicity (DLT) was observed at either dose level. A three‐fold higher area under the curve (AUC) was achieved in our cohort compared to adults using a similar dosing schedule. The PD studies showed increase in acetylated lysine in peripheral blood leukocytes at both doses.

   Entinostat was well tolerated with no DLT observed. All patients experienced progression within the first two cycles, except one patient with ependymoma with stable disease. Based on PK and PD, the R2PD in pediatric patients with solid tumors is 4 mg/m²orally administered once weekly.

    

   more » « less
3. Multimorbidity and Quality of Life: The Mediating Role of ADL, IADL, Loneliness, and Depressive Symptoms

   https://doi.org/10.1093/geroni/igad047  

   Sieber, Stefan ; Roquet, Angélique ; Lampraki, Charikleia ; Jopp, Daniela S. ; Thorpe, Jr, ed., Roland J. ( June 2023 , Innovation in Aging)

   The ubiquity of multimorbidity makes it crucial to examine the intermediary factors linking it with quality of life (QoL). The objective was to examine to what extent the association between multimorbidity and QoL was mediated by functional and emotional/mental health and how these mediation pathways differed by sociodemographic factors (age, gender, education, and financial strain).

   Data from Waves 4 to 8 of 36,908 individuals from the Survey of Health, Aging, and Retirement in Europe (SHARE) were included. Multimorbidity (exposure) was defined as having 2 or more chronic conditions. Mediators included limitations with (instrumental) activities of daily living (ADL and IADL), loneliness, and depressive symptoms. QoL (outcome) was assessed with the CASP-12 scale. Longitudinal model-based causal mediation analyses were performed to decompose the total association between multimorbidity and QoL into direct and indirect effects. Moderated mediation analyses tested for differences in mediation pathways by sociodemographic factors.

   Multimorbidity was significantly associated with lower QoL (direct effect: b = −0.66). This association was mediated by ADL limitations (percentage mediated 0.97%), IADL limitations (3.24%), and depressive symptoms (16.70%), but not by loneliness. The mediation pathways were moderated by age, education, financial strain, and gender.

   ADL, IADL, and depressive symptoms are crucial intermediary factors between multimorbidity and QoL in older European adults, with changing importance according to age, education, financial strain, and gender. The findings may help to increase the QoL of individuals with multimorbidity and redirect care efforts to these factors.

    

   more » « less
4. Effective Brain Connectivity Extraction by Frequency-Domain Convergent Cross-Mapping (FDCCM) and its Application in Parkinson's Disease Classification

   https://doi.org/10.1109/TBME.2023.3250355  

   Avvaru, Sandeep ; Parhi, Keshab K. ( February 2023 , IEEE Transactions on Biomedical Engineering)

   Objective: Inferring causal or effective connectivity between measured timeseries is crucial to understanding directed interactions in complex systems. This task is especially challenging in the brain as the underlying dynamics are not well-understood. This paper aims to introduce a novel causality measure called frequency-domain convergent cross-mapping (FDCCM) that utilizes frequency-domain dynamics through nonlinear state-space reconstruction. Method: Using synthesized chaotic timeseries, we investigate general applicability of FDCCM at different causal strengths and noise levels. We also apply our method on two resting-state Parkinson's datasets with 31 and 54 subjects, respectively. To this end, we construct causal networks, extract network features, and perform machine learning analysis to distinguish Parkinson's disease patients (PD) from age and gender-matched healthy controls (HC). Specifically, we use the FDCCM networks to compute the betweenness centrality of the network nodes, which act as features for the classification models. Result: The analysis on simulated data showed that FDCCM is resilient to additive Gaussian noise, making it suitable for real-world applications. Our proposed method also decodes scalp-EEG signals to classify the PD and HC groups with approximately 97% leave-one-subject-out cross-validation accuracy. We compared decoders from six cortical regions to find that features derived from the left temporal lobe lead to a higher classification accuracy of 84.5% compared to other regions. Moreover, when the classifier trained using FDCCM networks from one dataset was tested on an independent out-of-sample dataset, it attained an accuracy of 84%. This accuracy is significantly higher than correlational networks (45.2%) and CCM networks (54.84%). Significance: These findings suggest that our spectral-based causality measure can improve classification performance and reveal useful network biomarkers of Parkinson's disease. 

   more » « less
5. Dyskinesia estimation during activities of daily living using wearable motion sensors and deep recurrent networks

   https://doi.org/10.1038/s41598-021-86705-1  

   Hssayeni, Murtadha D. ; Jimenez-Shahed, Joohi ; Burack, Michelle A. ; Ghoraani, Behnaz ( December 2021 , Scientific Reports)

   null (Ed.)

   Abstract Levodopa-induced dyskinesias are abnormal involuntary movements experienced by the majority of persons with Parkinson’s disease (PwP) at some point over the course of the disease. Choreiform as the most common phenomenology of levodopa-induced dyskinesias can be managed by adjusting the dose/frequency of PD medication(s) based on a PwP’s motor fluctuations over a typical day. We developed a sensor-based assessment system to provide such information. We used movement data collected from the upper and lower extremities of 15 PwPs along with a deep recurrent model to estimate dyskinesia severity as they perform different activities of daily living (ADL). Subjects performed a variety of ADLs during a 4-h period while their dyskinesia severity was rated by the movement disorder experts. The estimated dyskinesia severity scores from our model correlated highly with the expert-rated scores ( r = 0.87 ( p < 0.001)), which was higher than the performance of linear regression that is commonly used for dyskinesia estimation ( r = 0.81 ( p < 0.001)). Our model provided consistent performance at different ADLs with minimum r = 0.70 (during walking) to maximum r = 0.84 (drinking) in comparison to linear regression with r = 0.00 (walking) to r = 0.76 (cutting food). These findings suggest that when our model is applied to at-home sensor data, it can provide an accurate picture of changes of dyskinesia severity facilitating effective medication adjustments. 

   more » « less