Automatic Assessment of Upper Extremity Function and Mobile Application for Self-Administered Stroke Rehabilitation.

Rehabilitation training is essential for a successful recovery of upper extremity function after stroke. Training programs are typically conducted in hospitals or rehabilitation centers, supervised by specialized medical professionals. However, frequent visits to hospitals can be burdensome for stroke patients with limited mobility. We consider a self-administered rehabilitation system based on a mobile application in which patients can periodically upload videos of themselves performing reach-to-grasp tasks to receive recommendations for self-managed exercises or progress reports. Sensing equipment aside from cameras is typically unavailable in the home environment. A key contribution of our work is to propose a deep learning-based assessment model trained only with video data. As all patients carry out identical tasks, a fine-grained assessment of task execution is required. Our model addresses this difficulty by learning RGB and optical flow data in a complementary manner. The correlation between the RGB and optical flow data is captured by a novel module for modality fusion using cross-attention with Transformers. Experiments showed that our model achieved higher accuracy in movement assessment than existing methods for action recognition. Based on the assessment model, we developed a patient-centered, solution-based mobile application for upper extremity exercises for hemiplegia, which can recommend 57 exercises with three levels of difficulty. A prototype of our application was evaluated by potential end-users and achieved a good quality score on the Mobile Application Rating Scale (MARS).

Application of deep learning technology for temporal analysis of videofluoroscopic swallowing studies.

Temporal parameters during swallowing are analyzed for objective and quantitative evaluation of videofluoroscopic swallowing studies (VFSS). Manual analysis by clinicians is time-consuming, complicated and prone to human error during interpretation; therefore, automated analysis using deep learning has been attempted. We aimed to develop a model for the automatic measurement of various temporal parameters of swallowing using deep learning. Overall, 547 VFSS video clips were included. Seven temporal parameters were manually measured by two physiatrists as ground-truth data: oral phase duration, pharyngeal delay time, pharyngeal response time, pharyngeal transit time, laryngeal vestibule closure reaction time, laryngeal vestibule closure duration, and upper esophageal sphincter opening duration. ResNet3D was selected as the base model for the deep learning of temporal parameters. The performances of ResNet3D variants were compared with those of the VGG and I3D models used previously. The average accuracy of the proposed ResNet3D variants was from 0.901 to 0.981. The F1 scores and average precision were 0.794 to 0.941 and 0.714 to 0.899, respectively. Compared to the VGG and I3D models, our model achieved the best results in terms of accuracy, F1 score, and average precision values. Through the clinical application of this automatic model, temporal analysis of VFSS will be easier and more accurate.

Quantitative Analysis of Temporal Parameters Correlated with Aspiration and Lesion Location in Stroke Patients.

The purpose of this study was to identify differences in temporal parameters correlating to the presence of aspiration and the severity of penetration-aspiration scale (PAS) in patients with dysphagia after stroke. We also investigated whether there was a significant difference in temporal parameters based on the location of the stroke lesion. A total of 91 patient videofluoroscopic swallowing study (VFSS) videos of stroke patients with dysphagia were retrospectively analyzed. Various temporal parameters including oral phase duration, pharyngeal delay time, pharyngeal response time, pharyngeal transit time, laryngeal vestibule closure reaction time, laryngeal vestibule closure duration, upper esophageal sphincter opening duration and upper esophageal sphincter reaction time were measured. Subjects were grouped by the presence of aspiration, PAS score, and location of the stroke lesion. Pharyngeal response time, laryngeal vestibule closure duration, and upper esophageal sphincter opening duration were significantly prolonged in the aspiration group. These three factors showed positive correlation with PAS. In terms of stroke lesion, oral phase duration was significantly prolonged in the supratentorial lesion group, while upper esophageal sphincter opening duration was significantly prolonged in the infratentorial lesion group. We have demonstrated that quantitative temporal analysis of VFSS can be a clinically valuable tool identifying dysphagia pattern associated with stroke lesion or aspiration risk.

The reliability and validity of the Korean version of the reaching performance scale for stroke after translation and cross-cultural adaptation.

The reaching performance scale for stroke evaluates reach-to grasp movement of the upper extremity and compensatory movements. The objective of the study was to translate and cross-culturally adapt the reaching performance scale for stroke into Korean and assess its reliability and validity. An expert committee supervised the forward and backward translation and the final translated Korean version of reaching performance scale for stroke (K-RPSS). Ninety-eight individuals (mean age 63.63 ± 13.13 years), with hemiplegia after stroke (97.82 ± 191.50 weeks) were evaluated independently by physiatrist and occupational therapist using K-RPSS. Degree of the motor impairment and functional status of patients were evaluated with the Fugl-Meyer assessment for upper limb and manual function test. The K-RPSS demonstrated excellent intra-rater (intraclass correlation coefficient = 0.9865), inter-rater (intraclass correlation coefficient 0.9866) reliability, and internal consistency (Cronbach α = 0.977 for total score). The validity of K-RPSS was confirmed with strong correlation with Fugl-Meyer assessment and manual function test. The K-RPSS had ceiling effect and the 3 points is the score change needed to indicate a change in a patient. K-RPSS has been shown to be a reliable and valid tool for assessing the compensation and movement quality of the upper limb and trunk when performing a reach-to grasp task. The K-RPSS is expected to be used widely by Korean speaking clinicians and researchers.

Fabrication and characterization of white polymer light emitting diodes using PFO:MDMO-PPV.

White polymer light emitting diodes (WPLEDs) with a glass/ITO/PEDOT:PSS/PFO:MDMO-PPV/ TPBI/LIF/Al structure were fabricated in order to investigate the optimum doping concentration of the emission materials. PEDOT:PSS was introduced as the hole transport material. The PFO and MDMO-PPV were used as the light emitting host and the guest materials, respectively. The PFO:MDMO-PPV mixed solution was spin-coated onto the PEDOT:PSS/ITO substrate. TPBI, LiF and Al were deposited by thermal evaporation as the hole blocking, electron injection, and cathode materials, respectively. As a result, the current density and luminance of the WPLED with the 20.0 wt% MDMO-PPV concentration in the PFO host material were found to be about 365 mA/cm2 and 4315 cd/m2, respectively. The maximum external quantum efficiency (EQE) of the same sample was found to be 11.26%, which may be ascribed to the efficient energy transfer from the PFO host to the MDMO-PPV guest material.

Properties of flexible phosphorescence polymer light emitting diodes coated on polyethylenenaphthalate plastic substrates.

Flexible phosphorescence polymer light emitting diodes (PhPLEDs) with PEN/ITO/PEDOT:PSS/ PVK:Ir(ppy)3/TPBI/LiF/Al structure were fabricated to investigate the effects of Ir(ppy)3 doping concentrations on the optical and electrical properties of the devices. PVK and Ir(ppy)3 conjugated polymers as host and guest materials in the emission layer were spun coated at various concentrations of Ir(ppy)3 ranging from 2.0 to 8.0 vol%. As the concentration of Ir(ppy)3 increased from 2.0 to 6.0 vol%, the electrical and optical properties of the flexible PhPLEDs were improved clearly. Maximum luminance and current density were obtained for a PhPLED with an Ir(ppy)3 concentration of 6.0 vol%, with 6815 cd/m2 and 393 mA/cm2 at 9 V. The current efficiency tends to increase with the Ir(ppy)3 concentration, because of the formation of the excitons required to emit light.

Retrieval of the extreme values under deadline constraints in wireless sensor networks.

We consider a problem of retrieving the extreme value among sensed data under deadline constraints in wireless sensor networks with potential applications to alarm systems. The sensed data is mapped to a score which we adopt as a unified measure of the relative urgency of the data. The objective is to retrieve the data with the maximum score. We propose fully distributed schemes for contention based medium access and data combining. The proposed medium access scheme uses a randomized back-off which is controlled based on the score of the data to be transmitted. Data combining techniques are proposed to further suppress unnecessary traffic and reduce contention. The key observation is that one should aggressively prioritize packets with high score, up to an extent that does not incur excessive contention in channel access. Designed to capture such aspect, the proposed scheme is shown to substantially decrease the latency of the retrieval.