Dual ultra-wideband(UWB)radar-based sleep posture recognition system:Towards ubiquitous sleep monitoring

Sleep posture monitoring is an essential assessment for obstructive sleep apnea(OSA)patients.The objective of this study is to develop a machine learning-based sleep posture recognition system using a dual ultra-wideband radar system.We collected radiofrequency data from two radars positioned over and at the side of the bed for 16 patients performing four sleep postures(supine,left and right lateral,and prone).We proposed and evaluated deep learning approaches that streamlined feature extraction and classification,and the traditional machine learning approaches that involved different combinations of feature extractors and classifiers.Our results showed that the dual radar system performed better than either single radar.Predetermined statistical features with random forest classifier yielded the best accuracy(0.887),which could be further improved via an ablation study(0.938).Deep learning approach using transformer yielded accuracy of 0.713.

X-reality for phantom limb management for amputees:A systematic review and meta-analysis

Phantom limb is a disabling neuropsychiatric condition among amputees resulting in pain and disturbance that impact their functions,quality of life,and autonomy.While pharmacological approaches appeared to be ineffec-tive,the emergence and integration of X-reality,including virtual reality,augmented reality,and mixed reality,might elevate the effectiveness of mirror therapy in managing phantom limb.The objective of this study is to review X-reality for managing phantom pain.A systematic search was conducted on PubMed,Scopus,Web of Science,PsycINFO,Embase,and CINAHL.Sixteen(n=16)studies containing 66 lower-limb and 53 upper-limb amputees were included for the review over the thematic framework of amputee characteristics and intervention designs,while thirteen(n=13)studies were further proceeded for the meta-analysis.We found eleven studies on virtual reality(n=11),four studies on marker-based augmented reality(n=4)and one study on mixed reality(n=1)with a total of 40 game/task themes involving,motor skills,motor control,and stimulus-sensing.Regardless,all these interventions adopted the movement representation strategies with different techniques.Overall,the X-reality interventions reduced the pain level of the amputees(mean difference:-2.30,95%CI,-3.38 to-1.22),especially the virtual reality subgroup(mean difference:-2.83,95%CI,-4.43 to-1.22).However,there were substantial heterogeneity and partially explained by the subgroup analysis on publication year.The strength of evidence was limited by case reports and case series in this review.

Corrigendum to‘Dual ultra-wideband(UWB)radar-based sleep posture recognition system:Towards ubiquitous sleep monitoring’[Engineered Regeneration 4(2023)36-43]

The authors regret that the name of the ethic committee that approved the study and the reference number was omitted from the published paper.In this research,all participants signed an informed consent after receiving an oral and written description of the experiment before the start of the experiment.The study was approved by the Human Subjects Ethics HSEARS20210127007.

Augmented reality(AR)and fracture mapping model on middle-aged femoral neck fracture:A proof-of-concept towards interactive visualization

A thorough understanding of the fracture characteristics can assist the decision-making process for surgery.This study aimed to characterize the femoral neck fractures among middle-aged patients and illustrated a biomedical visualization method using a fracture mapping model and augmented reality.We collected plain radiography and computed tomography(CT)data from 156 adult patients with a femoral neck fracture.The descriptive study showed that Type I and Type II fractures accounted for 8(5%)and 64(41%)cases.In comparison,Type IV fractures accounted for 44(28%)and 40(25%)cases according to the Garden classification.Comminuted fractures and cortical defects were identified in 14.74%and 29.49%of the cases.A fracture mapping model was reconstructed based on the CT data and demonstrated the location and distribution of the major fracture lines surrounding the head-neck junction.We also illustrated the application of augmented reality technology to visualize and interact with the patient-specific fracture model and the fracture mapping model that facilitated education,training,and surgical planning.Future studies may consider mapping other biomechanical data,such as joint loading and stress distribution,and exploring artificial intelligence via deep reinforcement learning for computer-aided fracture reduction and procedure planning.

Virtual reality based multiple life skill training for intellectual disability:A multicenter randomized controlled trial

Life skill-based training is essential for intellectual disabled individuals to regain autonomy and social inclusion.Virtual reality(VR)could deliver a nascent application that is more engaging and secure.The objective of this study was to evaluate the training effects of the VR-based multiple life skill training program on life skill perfor-mance,self-efficacy,memory,cognitive and behavioral functions via a multicenter randomized controlled trial.A total of 145 intellectual disabled participants were recruited for a randomized controlled trial with three in-tervention arms:VR(n=42);traditional(n=53);and control(n=50).The life skill tasks for the interventions included 1)grocery shopping,2)cooking,and 3)kitchen cleaning.Outcome measures were performance scores for the three tasks,self-efficacy scale,digit span score,and Frontal Assessment Battery score,graded by blinded assessors.Before-after effects of each group were evaluated using separated Wilcoxon Signed-rank tests.VR sig-nificantly improved cooking,cleaning performance,and memory span,while traditional training significantly improved shopping,cooking,and cleaning tasks.The generalized estimating equation evaluated effects between groups adjusted for age,gender and intelligence quotient(IQ).VR had significantly larger improvement effects in cooking and cleaning than the control group and memory span compared to traditional training and controls groups.IQ appeared to be a significant confounder on the training effect.Future work may consider developing artificial intelligence to customize programs depending on IQ level.