Predicting recurrent glioblastoma clinical outcome to immune checkpoint inhibition and low-dose bevacizumab with tumor in situ fluid circulating tumor DNA analysis.

OBJECTIVE: Most recurrent glioblastoma (rGBM) patients do not benefit from immune checkpoint inhibition, emphasizing the necessity for response biomarkers. This study evaluates whether tumor in situ fluid (TISF) circulating tumor DNA (ctDNA) could serve as a biomarker for response to low-dose bevacizumab (Bev) plus anti-PD-1 therapy in rGBM patients, aiming to enhance systemic responses to immunotherapy. METHODS: In this phase II trial, 32 GBM patients with first recurrence after standard therapy were enrolled and then received tislelizumab plus low-dose Bev each cycle. TISF samples were analyzed for ctDNA using a 551-gene panel before each treatment. RESULTS: The median progression-free survival (mPFS) and overall survival (mOS) were 8.2 months (95% CI, 5.2-11.1) and 14.3 months (95% CI, 6.5-22.1), respectively. The 12-month OS was 43.8%, and the objective response rate was 56.3%. Patients with more than 20% reduction in the mutant allele fraction and tumor mutational burden after treatment were significantly associated with better prognosis compared to baseline TISF-ctDNA. Among detectable gene mutations, patients with MUC16 mutation, EGFR mutation & amplification, SRSF2 amplification, and H3F3B amplification were significantly associated with worse prognosis. CONCLUSIONS: Low-dose Bev plus anti-PD-1 therapy significantly improves OS in rGBM patients, offering guiding significance for future individualized treatment strategies. TISF-ctDNA can monitor rGBM patients' response to combination therapy and guide treatment. CLINICAL TRIAL REGISTRATION: This trial is registered with ClinicalTrials.gov, NCT05540275.

Daily Posture Behavior Patterns Derived From Multitime-Scale Topic Models Using Wearable Triaxial Acceleration for Assessment of Concern About Falling.

Concern about falling is prevalent in older population. This condition would cause a series of adverse physical and psychological consequences for older adults' health. Traditional assessment of concern about falling is relied on self-reported questionnaires and thus is too subjective. Therefore, we proposed a novel multi-time-scale topic modelling approach to quantitatively evaluate concern about falling by analyzing triaxial acceleration signals collected from a wearable pendent sensor. Different posture segments were firstly recognized to extract their corresponding feature subsets. Then, each selected feature related to concern about falling was clustered into discrete levels as feature letters of artificial words in different time scales. As a result, all older participants' signal recordings were converted to a collection of artificial documents, which can be processed by natural language processing methodologies. The topic modelling technique was used to discover daily posture behavior patterns from these documents as discriminants between older adults with different levels of concern about falling. The results indicated that there were significant differences in distributions of posture topics between groups of older adults with different levels of concern about falling. Additionally, the transitions of posture topics over daytime and nighttime revealed temporal regularities of posture behavior patterns of older adult's active and inactive status, which were substantially different for older adults with different levels of concern about falling. Finally, the level of concern about falling was accurately determined with accuracy of 71.2% based on the distributions of posture topics combined with the mobility performance metrics of walking behaviors and demographic information.

Csnk1a1 inhibition modulates the inflammatory secretome and enhances response to radiotherapy in glioma.

Glioblastoma multiforme (GBM), a fatal brain tumour with no available targeted therapies, has a poor prognosis. At present, radiotherapy is one of the main methods to treat glioma, but it leads to an obvious increase in inflammatory factors in the tumour microenvironment, especially IL-6 and CXCL1, which plays a role in tumour to resistance radiotherapy and tumorigenesis. Casein kinase 1 alpha 1 (CK1α) (encoded on chromosome 5q by Csnk1a1) is considered an attractive target for Tp53 wild-type acute myeloid leukaemia (AML) treatment. In this study, we evaluated the anti-tumour effect of Csnk1a1 suppression in GBM cells in vitro and in vivo. We found that down-regulation of Csnk1a1 or inhibition by D4476, a Csnk1a1 inhibitor, reduced GBM cell proliferation efficiently in both Tp53 wild-type and Tp53-mutant GBM cells. On the contrary, overexpression of Csnk1a1 promoted cell proliferation and colony formation. Csnk1a1 inhibition improved the sensitivity to radiotherapy. Furthermore, down-regulation of Csnk1a1 reduced the production and secretion of pro-inflammatory factors. In the preclinical GBM model, treatment with D4476 significantly inhibited the increase in pro-inflammatory factors caused by radiotherapy and improved radiotherapy sensitivity, thus inhibiting tumour growth and prolonging animal survival time. These results suggest targeting Csnk1a1 exert an anti-tumour role as an inhibitor of inflammatory factors, providing a new strategy for the treatment of glioma.

Clinical Evaluation a New Treatment for Infection After Ventriculoperitoneal Shunt.

ABSTRACT: To explore a new surgical treatment for infection and obstruction of ventriculoperitoneal shunt in hydrocephalus. Two cases of post-operative infection of ventriculoperitoneal shunt were analyzed retrospectively. One case was cryptococcal infection, the other case was Acinetobacter lwoffii. The number of cerebrospinal fluid cells was high, the infection of ventriculoperitoneal shunt was generally complicated with abdominal obstruction, and the hydrocephalus was aggravated again, The authors try to pull out the drainage tube at the end of abdominal cavity for external drainage, combined with intravenous antibiotics, completely control of infection, and then use the original shunt device for intraventricular jugular shunt. The authors explore that this method is simple, safe and effective, and it is an effective and feasible method for the treatment of infection after ventriculoperitoneal shunt.

Similarity Changes Analysis for Heart Rate Fluctuation Regularity as a New Screening Method for Congestive Heart Failure.

Congestive heart failure (CHF) is a chronic cardiovascular condition associated with dysfunction of the autonomic nervous system (ANS). Heart rate variability (HRV) has been widely used to assess ANS. This paper proposes a new HRV analysis method, which uses information-based similarity (IBS) transformation and fuzzy approximate entropy (fApEn) algorithm to obtain the fApEn_IBS index, which is used to observe the complexity of autonomic fluctuations in CHF within 24 h. We used 98 ECG records (54 health records and 44 CHF records) from the PhysioNet database. The fApEn_IBS index was statistically significant between the control and CHF groups (p < 0.001). Compared with the classical indices low-to-high frequency power ratio (LF/HF) and IBS, the fApEn_IBS index further utilizes the changes in the rhythm of heart rate (HR) fluctuations between RR intervals to fully extract relevant information between adjacent time intervals and significantly improves the performance of CHF screening. The CHF classification accuracy of fApEn_IBS was 84.69%, higher than LF/HF (77.55%) and IBS (83.67%). Moreover, the combination of IBS, fApEn_IBS, and LF/HF reached the highest CHF screening accuracy (98.98%) with the random forest (RF) classifier, indicating that the IBS and LF/HF had good complementarity. Therefore, fApEn_IBS effusively reflects the complexity of autonomic nerves in CHF and is a valuable CHF assessment tool.

Short-Term HRV Analysis Using Nonparametric Sample Entropy for Obstructive Sleep Apnea.

Obstructive sleep apnea (OSA) is associated with reduced heart rate variability (HRV) and autonomic nervous system dysfunction. Sample entropy (SampEn) is commonly used for regularity analysis. However, it has limitations in processing short-term segments of HRV signals due to the extreme dependence of its functional parameters. We used the nonparametric sample entropy (NPSampEn) as a novel index for short-term HRV analysis in the case of OSA. The manuscript included 60 6-h electrocardiogram recordings (20 healthy, 14 mild-moderate OSA, and 26 severe OSA) from the PhysioNet database. The NPSampEn value was compared with the SampEn value and frequency domain indices. The empirical results showed that NPSampEn could better differentiate the three groups (p < 0.01) than the ratio of low frequency power to high frequency power (LF/HF) and SampEn. Moreover, NPSampEn (83.3%) approached a higher OSA screening accuracy than the LF/HF (73.3%) and SampEn (68.3%). Compared with SampEn (|r| = 0.602, p < 0.05), NPSampEn (|r| = 0.756, p < 0.05) had a significantly stronger association with the apnea-hypopnea index (AHI). Hence, NPSampEn can fully overcome the influence of individual differences that are prevalent in biomedical signal processing, and might be useful in processing short-term segments of HRV signal.

MALT1 is a potential therapeutic target in glioblastoma and plays a crucial role in EGFR-induced NF-κB activation.

Glioblastoma multiforme (GBM) is the most common malignant tumour in the adult brain and hard to treat. Nuclear factor κB (NF-κB) signalling has a crucial role in the tumorigenesis of GBM. EGFR signalling is an important driver of NF-κB activation in GBM; however, the correlation between EGFR and the NF-κB pathway remains unclear. In this study, we investigated the role of mucosa-associated lymphoma antigen 1 (MALT1) in glioma progression and evaluated the anti-tumour activity and effectiveness of MI-2, a MALT1 inhibitor in a pre-clinical GBM model. We identified a paracaspase MALT1 that is involved in EGFR-induced NF-kB activation in GBM. MALT1 deficiency or inhibition significantly affected the proliferation, survival, migration and invasion of GBM cells both in vitro and in vivo. Moreover, MALT1 inhibition caused G1 cell cycle arrest by regulating multiple cell cycle-associated proteins. Mechanistically, MALTI inhibition blocks the degradation of IκBα and prevents the nuclear accumulation of the NF-κB p65 subunit in GBM cells. This study found that MALT1, a key signal transduction cascade, can mediate EGFR-induced NF-kB activation in GBM and may be potentially used as a novel therapeutic target for GBM.

Reversible inhibitor of CRM1 sensitizes glioblastoma cells to radiation by blocking the NF-κB signaling pathway.

BACKGROUND: Activation of nuclear factor-kappa B (NF-κΒ) through DNA damage is one of the causes of tumor cell resistance to radiotherapy. Chromosome region 1 (CRM1) regulates tumor cell proliferation, drug resistance, and radiation resistance by regulating the nuclear-cytoplasmic translocation of important tumor suppressor proteins or proto-oncoproteins. A large number of studies have reported that inhibition of CRM1 suppresses the activation of NF-κΒ. Thus, we hypothesize that the reversible CRM1 inhibitor S109 may induce radiosensitivity in glioblastoma (GBM) by regulating the NF-κΒ signaling pathway. METHODS: This study utilized the cell counting kit-8 (CCK-8), 5-ethynyl-2'-deoxyuridine (EdU), and colony formation assay to evaluate the effect of S109 combined with radiotherapy on the proliferation and survival of GBM cells. The therapeutic efficacy of S109 combined with radiotherapy was evaluated in vivo to explore the therapeutic mechanism of S109-induced GBM radiosensitization. RESULTS: We found that S109 combined with radiotherapy significantly inhibited GBM cell proliferation and colony formation. By regulating the levels of multiple cell cycle- and apoptosis-related proteins, the combination therapy induced G1 cell cycle arrest in GBM cells. In vivo studies showed that S109 combined with radiotherapy significantly inhibited the growth of intracranial GBM and prolonged survival. Importantly, we found that S109 combined with radiotherapy promoted the nuclear accumulation of IκΒα, and inhibited phosphorylation of p65 and the transcriptional activation of NF-κΒ. CONCLUSION: Our findings provide a new therapeutic regimen for improving GBM radiosensitivity as well as a scientific basis for further clinical trials to evaluate this combination therapy.

Application of the Variance Delay Fuzzy Approximate Entropy for Autonomic Nervous System Fluctuation Analysis in Obstructive Sleep Apnea Patients.

Obstructive sleep apnea (OSA) is a fatal respiratory disease occurring in sleep. OSA can induce declined heart rate variability (HRV) and was reported to have autonomic nerve system (ANS) dysfunction. Variance delay fuzzy approximate entropy (VD_fApEn) was proposed as a nonlinear index to study the fluctuation change of ANS in OSA patients. Sixty electrocardiogram (ECG) recordings of the PhysioNet database (20 normal, 14 mild-moderate OSA, and 26 severe OSA) were intercepted for 6 h and divided into 5-min segments. HRV analysis were adopted in traditional frequency domain, and nonlinear HRV indices were also calculated. Among these indices, VD_fApEn could significantly differentiate among the three groups (p < 0.05) compared with the ratio of low frequency power and high frequency power (LF/HF ratio) and fuzzy approximate entropy (fApEn). Moreover, the VD_fApEn (90%) reached a higher OSA screening accuracy compared with LF/HF ratio (80%) and fApEn (78.3%). Therefore, VD_fApEn provides a potential clinical method for ANS fluctuation analysis in OSA patients and OSA severity analysis.

SWAP-70 promotes glioblastoma cellular migration and invasion by regulating the expression of CD44s.

BACKGROUND: Switch-associated protein 70 (SWAP-70) is a guanine nucleotide exchange factor that is involved in cytoskeletal rearrangement and regulation of migration and invasion of malignant tumors. However, the mechanism by which SWAP-70 regulates the migration and invasion of glioblastoma (GB) cells has not been fully elucidated. METHODS: This study used an online database to analyze the relationship between SWAP-70 expression and prognosis in GB patients. The in vitro wound healing assay and transwell invasion assay were used to determine the role of SWAP-70 in GB cell migration and invasion as well as the underlying mechanism. RESULTS: We found that patients with high SWAP-70 expression in the GB had a poor prognosis. Downregulation of SWAP-70 inhibited GB cell migration and invasion, whereas SWAP-70 overexpression had an opposite effect. Interestingly, SWAP-70 expression was positively correlated with the expression of the standard form of CD44 (CD44s) in GB tissues. Downregulation of SWAP-70 also reduced CD44s protein expression, whereas SWAP-70 overexpression enhanced CD44s protein expression. However, downregulation of SWAP-70 expression did not affect the mRNA expression of CD44s. Reversal experiments showed that overexpressing CD44s in cell lines with downregulated SWAP-70 partially abolished the inhibitory effects of downregulated SWAP-70 on GB cell migration and invasion. CONCLUSIONS: These results suggest that SWAP-70 may promote GB cell migration and invasion by regulating the expression of CD44s. SWAP-70 may serve as a new biomarker and a potential therapeutic target for GB.

The Interaction Analysis between the Sympathetic and Parasympathetic Systems in CHF by Using Transfer Entropy Method.

Congestive heart failure (CHF) is a cardiovascular disease associated with autonomic dysfunction, where sympathovagal imbalance was reported in many studies using heart rate variability (HRV). To learn more about the dynamic interaction in the autonomic nervous system (ANS), we explored the directed interaction between the sympathetic nervous system (SNS) and the parasympathetic nervous system (PNS) with the help of transfer entropy (TE). This article included 24-h RR interval signals of 54 healthy subjects (31 males and 23 females, 61.38 ± 11.63 years old) and 44 CHF subjects (8 males and 2 females, 19 subjects' gender were unknown, 55.51 ± 11.44 years old, 4 in class I, 8 in class II and 32 in class III~IV, according to the New York Heart Association Function Classification), obtained from the PhysioNet database and then segmented into 5-min non-overlapping epochs using cubic spline interpolation. For each segment in the normal group and CHF group, frequency-domain features included low-frequency (LF) power, high-frequency (HF) power and LF/HF ratio were extracted as classical estimators of autonomic activity. In the nonlinear domain, TE between LF and HF were calculated to quantify the information exchanging between SNS and PNS. Compared with the normal group, an extreme decrease in LF/HF ratio (p = 0.000) and extreme increases in both TE(LF→HF) (p = 0.000) and TE(HF→LF) (p = 0.000) in the CHF group were observed. Moreover, both in normal and CHF groups, TE(LF→HF) was a lot greater than TE(HF→LF) (p = 0.000), revealing that TE was able to distinguish the difference in the amount of directed information transfer among ANS. Extracted features were further applied in discriminating CHF using IBM SPSS Statistics discriminant analysis. The combination of the LF/HF ratio, TE(LF→HF) and TE(HF→LF) reached the highest screening accuracy (83.7%). Our results suggested that TE could serve as a complement to traditional index LF/HF in CHF screening.

Multiscale Bidirectional Temporal Convolutional Network for Sleep Apnea Detection Based on Wearable Photoplethysmography Bracelet.

Sleep apnea syndrome (SAS), which can lead to a range of Cardiopulmonary diseases, is a common chronic sleep disorder. The unobtrusive detection based on wearable devices is helpful for early diagnosis and treatment of SAS. To this end, this paper presents a method based on a one-dimensional multi-scale bidirectional temporal convolutional neural network (1D-MsBiTCNet) and two model performance optimization techniques, i.e., regularized dropout (RD) and logit adjustment (LA). Among them, 1D-MsBiTCNet has outstanding capabilities in both feature extraction and temporal dependence representation. RD and LA play an effective role in solving the overfitting problem of model training and the class imbalance problem of the dataset, respectively. The proposed model was trained and tested on a photoplethysmography (PPG) dataset (including data from 92 subjects) collected from commercial wearable bracelets. On this dataset, our method achieved accuracy, sensitivity and specificity of 82.76%, 71.58%, 86.74% for per-segment detection, and 97.83%, 88.89%, 100.00% for per-recording severe SAS detection. For the precise quantification of apnea-hypopnea index (AHI), our method achieved a mean absolute error of 5.44 between the predicted AHI and the ground truth AHI. The experimental results show that our proposed method has an outstanding performance and can provide a methodological reference for large-scale SAS automatic detection.

Energy-Efficient Sleep Apnea Detection Using a Hyperdimensional Computing Framework Based on Wearable Bracelet Photoplethysmography.

OBJECTIVE: Sleep apnea syndrome (SAS) is a common sleep disorder, which has been shown to be an important contributor to major neurocognitive and cardiovascular sequelae. Considering current diagnostic strategies are limited with bulky medical devices and high examination expenses, a large number of cases go undiagnosed. To enable large-scale screening for SAS, wearable photoplethysmography (PPG) technologies have been used as an early detection tool. However, existing algorithms are energy-intensive and require large amounts of memory resources, which are believed to be the major drawbacks for further promotion of wearable devices for SAS detection. METHODS: In this paper, an energy-efficient method of SAS detection based on hyperdimensional computing (HDC) is proposed. Inspired by the phenomenon of chunking in cognitive psychology as a memory mechanism for improving working memory efficiency, we proposed a one-dimensional block local binary pattern (1D-BlockLBP) encoding scheme combined with HDC to preserve dominant dynamical and temporal characteristics of pulse rate signals from wearable PPG devices. RESULTS: Our method achieved 70.17 % accuracy in sleep apnea segment detection, which is comparable with traditional machine learning methods. Additionally, our method achieves up to 67× lower memory footprint, 68× latency reduction, and 93× energy saving on the ARM Cortex-M4 processor. CONCLUSION: The simplicity of hypervector operations in HDC and the novel 1D-BlockLBP encoding effectively preserve pulse rate signal characteristics with high computational efficiency. SIGNIFICANCE: This work provides a scalable solution for long-term home-based monitoring of sleep apnea, enhancing the feasibility of consistent patient care.

GOLPH3 regulates the migration and invasion of glioma cells though RhoA.

Golgi phosphoprotein 3 (GOLPH3) has been reported to be involved in the development of several human cancers. However, the biological significance of GOLPH3 in glioma progression remains largely unknown. In this study, we report, for the first time, that downregulation of GOLPH3 led to clear reductions in glioma cell migration and invasion. In addition, downregulation of GOLPH3 inhibited the expression of the small GTPase RhoA as well as cytoskeletal reorganization, which are both required for glioma cell migration. Furthermore, we found that the observed reductions in glioma cell migration and RhoA level could be rescued by RhoA overexpression. Taken together, these results show that GOLPH3 contributes to the motility of glioma cells by regulating the expression of RhoA.

MS-Net: Sleep apnea detection in PPG using multi-scale block and shadow module one-dimensional convolutional neural network.

Sleep Apnea (SA) is a respiratory disorder that affects sleep. However, the SA detection method based on polysomnography is complex and not suitable for home use. The detection approach using Photoplethysmography is low cost and convenient, which can be used to widely detect SA. This study proposed a method combining a multi-scale one-dimensional convolutional neural network and a shadow one-dimensional convolutional neural network based on dual-channel input. The time-series feature information of different segments were extracted from multi-scale temporal structure. Moreover, shadow module was adopted to make full use of the redundant information generated after multi-scale convolution operation, which improved the accuracy and ensured the portability of the model. At the same time, we introduced balanced bootstrapping and class weight, which effectively alleviated the problem of unbalanced classes. Our method achieved the result of 82.0% average accuracy, 74.4% average sensitivity and 85.1% average specificity for per-segment SA detection, and reached 93.6% average accuracy for per-recording SA detection after 5-fold cross validation. Experimental results show that this method has good robustness. It can be regarded as an effective aid in SA detection in household use.

Genomic alterations of oligodendrogliomas at distant recurrence.

BACKGROUND: Oligodendroglioma is known for its relatively better prognosis and responsiveness to radiotherapy and chemotherapy. However, little is known about the evolution of genetic changes as oligodendroglioma progresses. METHODS: In this study, we evaluated gene evolution invivo during tumor progression based on deep whole-genome sequencing data (ctDNA). We analyzed longitudinal genomic data from six patients during tumor evolution, of which five patients developed distant recurrence. RESULTS: Whole-exome sequencing demonstrated that the rate of shared mutations between the primary and recurrent samples was relatively low. In two cases, even well-known major driver mutations in CIC and FUBP1 that were detected in primary tumors were not detected in the relapse samples. Among these cases, two patients had a conversion from the IDH mutation in the originating state to the IDH1 wild state during the process of gene evolution under chemotherapy treatment, indicating that the cell phenotype and genetic characteristics of oligodendroglioma may change during tumor evolution. Two patients received long-term temozolomide (TMZ) treatment before the operation, and we found that recurrence tumors harbored mutations in the PI3K/AKT and Sonic hedgehog (SHh) signaling pathways. Hypermutation occurred with mutations in MMR genes in one patient, contributing to the rapid progression of the tumor. CONCLUSION: Oligodendroglioma displayed great spatial and temporal heterogeneity during tumor evolution. The PI3K/AKT and SHh signaling pathways may play an important role in promoting treatment resistance and distant relapse during oligodendroglioma evolution. In addition, there was a tendency to increase the degree of tumor malignancy during evolution. Distant recurrence may be a later event duringoligodendroglioma progression. CLINICALTRIALS: gov, Identifier: NCT05512325.

Molecular and clonal evolution in vivo reveal a common pathway of distant relapse gliomas.

The evolutionary trajectories of genomic alterations underlying distant recurrence in glioma remain largely unknown. To elucidate glioma evolution, we analyzed the evolutionary trajectories of matched pairs of primary tumors and relapse tumors or tumor in situ fluid (TISF) based on deep whole-genome sequencing data (ctDNA). We found that MMR gene mutations occurred in the late stage in IDH-mutant glioma during gene evolution, which activates multiple signaling pathways and significantly increases distant recurrence potential. The proneural subtype characterized by PDGFRA amplification was likely prone to hypermutation and distant recurrence following treatment. The classical and mesenchymal subtypes tended to progress locally through subclonal reconstruction, trunk genes transformation, and convergence evolution. EGFR and NOTCH signaling pathways and CDNK2A mutation play an important role in promoting tumor local progression. Glioma subtypes displayed distinct preferred evolutionary patterns. ClinicalTrials.gov, NCT05512325.

A low-cost body inertial-sensing network for practical gait discrimination of hemiplegia patients.

Gait analysis is widely used in detecting human walking disorders. Current gait analysis methods like video- or optical-based systems are expensive and cause invasion of human privacy. This article presents a self-developed low-cost body inertial-sensing network, which contains a base station, three wearable inertial measurement nodes, and the affiliated wireless communication protocol, for practical gait discrimination between hemiplegia patients and asymptomatic subjects. Every sensing node contains one three-axis accelerometer, one three-axis magnetometer, and one three-axis gyroscope. Seven hemiplegia patients (all were abnormal on the right side) and 7 asymptomatic subjects were examined. The three measurement nodes were attached on the thigh, the shank, and the dorsum of the foot, respectively (all on the right side of the body). A new method, which does not need to obtain accurate positions of the sensors, was used to calculate angles of knee flexion/extension and foot in the gait cycle. The angle amplitudes of initial contact, toe off, and knee flexion/extension were extracted. The results showed that there were significant differences between the two groups in the three angle amplitudes examined (-0.52±0.98° versus 6.94±2.63°, 28.33±11.66° versus 47.34±7.90°, and 26.85±8.6° versus 50.91±6.60°, respectively). It was concluded that the body inertial-sensing network platform provided a practical approach for wearable biomotion acquisition and was effective for discriminating gait symptoms between hemiplegia and asymptomatic subjects.

Deep Neural Network-based Video Processing to Obtain Upper-Extremity Motor Performance Toward Assessment of Cognitive and Motor Function.

Dementia is an increasing global health challenge. Motoric Cognitive Risk Syndrome (MCR) is a predementia stage that can be used to predict future occurrence of dementia. Traditionally, gait speed and subjective memory complaints are used to identify old adults with MCR. Our previous studies indicated that dual-task upper-extremity motor performance (DTUEMP) quantified by a single wrist-worn sensor was correlated with both motor and cognitive function. Therefore, the DTUEMP had a potential to be used in the diagnosis of MCR. Instead of using inertial sensors to capture kinematic data of upper-extremity movements, here we proposed a deep neural network-based video processing model to obtain DTUEMP metrics from a 20-second repetitive elbow flexion-extension test under dual-task condition. In details, we used a deep residual neural network to obtain joint coordinate set of the elbow and wrist in each frame, and then used optical flow method to correct the joint coordinates generated by the neural network. The coordinate sets of all frames in a video recording are used to generate angle sequence which represents rotation angle of the line between the wrist and elbow. Then, the DTUEMP metrics (the mean and SD of flexion and extension phase) are derived from angle sequence. Multi-task learning (MTL) is used to assess cognitive and motor function represented by MMSE and TUG scores based on DTUEMP metrics, with single-task learning (STL) linear model as a benchmark. The results showed a good agreement (r ≥ 0.80 and ICC ≥ 0.58) between the derived DTUEMP metrics from our proposed model and the ones from clinically validated sensor processing model. We also found that there were correlations with statistical significance (p < 0.05) between some of video-derived DTUEMP metrics (i.e. the mean of flexion time and extension time) and clinical cognitive scale (Mini-Mental State Examination, MMSE). Additionally, some of video-derived DTUEMP metrics (i.e. the mean and standard deviation of flexion time and extension time) was also associated with the scores of timed-up and go (TUG) which is a gold standard to measure functional mobility. Mean absolute percentage error (MAPE) of MTL surpassed that of STL (For MMSE, MTL: 18.63%, STL: 23.18%. For TUG, MTL: 17.88%, STL: 22.53%). The experiments with different light conditions and shot angles verified the robustness of our proposed video processing model to extract DTUEMP metrics in potentially various home environments (r ≥ 0.58 and ICC ≥ 0.71). This study shows possibility of replacing sensor processing model with video processing model for analyzing the DTUEMP and a promising future to adjuvant diagnosis of MCR via a mobile platform.

Fuzzy Approximate Entropy of Extrema Based on Multiple Moving Averages as a Novel Approach in Obstructive Sleep Apnea Screening.

OBJECTIVE: Obstructive sleep apnea (OSA) is a respiratory disease associated with autonomic nervous system dysfunction. As a novel method for analyzing OSA depending on heart rate variability, fuzzy approximate entropy of extrema based on multiple moving averages (Emma-fApEn) can effectively assess the sympathetic tension limits, thereby realizing a good performance in the disease severity screening. METHOD: Sixty 6-h electrocardiogram recordings (20 healthy, 16 mild/moderate OSA and 34 severe OSA) from the PhysioNet database were used in this study. The performances of minima of Emma-fApEn (fApEn-minima), maxima of Emma-fApEn (fApEn-maxima) and classic time-frequency domain indices for each recording were assessed by significance analysis, correlation analysis, parameter optimization and OSA screening. RESULTS: fApEn-minima and fApEn-maxima had significant differences between the severe OSA group and the other two groups, while the mean value (Mean) and the ratio of low-frequency power and high-frequency power (LH) could significantly differentiate OSA recordings from healthy recordings. The correlation coefficient between fApEn-minima and apnea-hypopnea index was the highest (|R| = 0.705). Machine learning methods were used to evaluate the performances of the above four indices. Random forest (RF) achieved the highest accuracy of 96.67% in OSA screening and 91.67% in severe OSA screening, with a good balance in both. CONCLUSION: Emma-fApEn may be used as a simple preliminary detection tool to assess the severity of OSA prior to polysomnography analysis.