DE-AFO: A Robotic Ankle Foot Orthosis for Children with Cerebral Palsy Powered by Dielectric Elastomer Artificial Muscle.

Conventional passive ankle foot orthoses (AFOs) have not seen substantial advances or functional improvements for decades, failing to meet the demands of many stakeholders, especially the pediatric population with neurological disorders. Our objective is to develop the first comfortable and unobtrusive powered AFO for children with cerebral palsy (CP), the DE-AFO. CP is the most diagnosed neuromotor disorder in the pediatric population. The standard of care for ankle control dysfunction associated with CP, however, is an unmechanized, bulky, and uncomfortable L-shaped conventional AFO. These passive orthoses constrain the ankle's motion and often cause muscle disuse atrophy, skin damage, and adverse neural adaptations. While powered orthoses could enhance natural ankle motion, their reliance on bulky, noisy, and rigid actuators like DC motors limits their acceptability. Our innovation, the DE-AFO, emerged from insights gathered during customer discovery interviews with 185 stakeholders within the AFO ecosystem as part of the NSF I-Corps program. The DE-AFO is a biomimetic robot that employs artificial muscles made from an electro-active polymer called dielectric elastomers (DEs) to assist ankle movements in the sagittal planes. It incorporates a gait phase detection controller to synchronize the artificial muscles with natural gait cycles, mimicking the function of natural ankle muscles. This device is the first of its kind to utilize lightweight, compact, soft, and silent artificial muscles that contract longitudinally, addressing traditional actuated AFOs' limitations by enhancing the orthosis's natural feel, comfort, and acceptability. In this paper, we outline our design approach and describe the three main components of the DE-AFO: the artificial muscle technology, the finite state machine (the gait phase detection system), and its mechanical structure. To verify the feasibility of our design, we theoretically calculated if DE-AFO can provide the necessary ankle moment assistance for children with CP-aligning with moments observed in typically developing children. To this end, we calculated the ankle moment deficit in a child with CP when compared with the normative moment of seven typically developing children. Our results demonstrated that the DE-AFO can provide meaningful ankle moment assistance, providing up to 69% and 100% of the required assistive force during the pre-swing phase and swing period of gait, respectively.

vissE: a versatile tool to identify and visualise higher-order molecular phenotypes from functional enrichment analysis.

Functional analysis of high throughput experiments using pathway analysis is now ubiquitous. Though powerful, these methods often produce thousands of redundant results owing to knowledgebase redundancies upstream. This scale of results hinders extensive exploration by biologists and can lead to investigator biases due to previous knowledge and expectations. To address this issue, we present vissE, a flexible network-based analysis and visualisation tool that organises information into semantic categories and provides various visualisation modules to characterise them with respect to the underlying data, thus providing a comprehensive view of the biological system. We demonstrate vissE's versatility by applying it to three different technologies: bulk, single-cell and spatial transcriptomics. Applying vissE to a factor analysis of a breast cancer spatial transcriptomic data, we identified stromal phenotypes that support tumour dissemination. Its adaptability allows vissE to enhance all existing gene-set enrichment and pathway analysis workflows, empowering biologists during molecular discovery.

Enhancing Wearable Gait Monitoring Systems: Identifying Optimal Kinematic Inputs in Typical Adolescents.

Machine learning-based gait systems facilitate the real-time control of gait assistive technologies in neurological conditions. Improving such systems needs the identification of kinematic signals from inertial measurement unit wearables (IMUs) that are robust across different walking conditions without extensive data processing. We quantify changes in two kinematic signals, acceleration and angular velocity, from IMUs worn on the frontal plane of bilateral shanks and thighs in 30 adolescents (8-18 years) on a treadmills and outdoor overground walking at three different speeds (self-selected, slow, and fast). Primary curve-based analyses included similarity analyses such as cosine, Euclidean distance, Poincare analysis, and a newly defined bilateral symmetry dissimilarity test (BSDT). Analysis indicated that superior-inferior shank acceleration (SI shank Acc) and medial-lateral shank angular velocity (ML shank AV) demonstrated no differences to the control signal in BSDT, indicating the least variability across the different walking conditions. Both SI shank Acc and ML shank AV were also robust in Poincare analysis. Secondary parameter-based similarity analyses with conventional spatiotemporal gait parameters were also performed. This normative dataset of walking reports raw signal kinematics that demonstrate the least to most variability in switching between treadmill and outdoor walking to help guide future machine learning models to assist gait in pediatric neurological conditions.

Subthreshold electrical noise alters walking balance control in individuals with cerebral palsy.

BACKGROUND: Sensory deficits in individuals with cerebral palsy (CP) play a critical role in balance control. However, there is a lack of effective interventions that address sensory facilitation to improve walking balance. Stochastic Resonance (SR) stimulation involves delivering sub threshold noise to improve balance in patients with sensory deficits by enhancing the detection of sensory input. RESEARCH QUESTION: To investigate the immediate effects of SR on walking balance in individuals with and without CP. METHODS: Thirty-four participants (17 CP, 17 age-and sex-matched typically developing controls or TD) between 8 and 24 years of age were recruited. SR stimulation was applied to the muscles and ligaments of ankle and hip joint. An optimal SR intensity during walking was determined for each subject. Participants walked on a self-paced treadmill for three trials of two minutes each using a random order of SR stimulation (SR) and no stimulation (noSR) control conditions. Our primary outcome measure was minimum lateral margin of stability (MOS). Secondary outcome measures include anterior MOS before heelstrike and spatiotemporal gait parameters. We performed two-way mixed ANOVAs with group (CP, TD) as between-subject and condition (noSR, SR) as within subject factors. RESULTS: Compared to walking without SR, there was a small but significant increase in the lateral and anterior MOS with SR stimulation, implying that a larger impulse was needed to become unstable, in turn implying higher stability. Step width and step ength decreased with SR for the CP group with SR stimulation. There were no significant effects for other spatiotemporal variables. SIGNIFICANCE: Sub threshold electrical noise can slightly improve walking balance control in individuals with CP. SR stimulation, through enhanced proprioception, may have improved the CP group's awareness of body motion during walking, thus leading them to adopt a more conservative stability strategy to prevent a potential fall.

Clinicopathologic Dissociation: Robust Lafora Body Accumulation in Malin KO Mice Without Observable Changes in Home-cage Behavior.

Lafora Disease (LD) is a syndrome of progressive myoclonic epilepsy and cumulative neurocognitive deterioration caused by recessively inherited genetic lesions of EPM2A (laforin) or NHLRC1 (malin). Neuropsychiatric symptomatology in LD is thought to be directly downstream of neuronal and astrocytic polyglucosan aggregates, termed Lafora bodies (LBs), which faithfully accumulate in an age-dependent manner in all mouse models of LD. In this study, we applied home-cage monitoring to examine the extent of neurobehavioral deterioration in a model of malin-deficient LD, as a means to identify robust preclinical endpoints that may guide the selection of novel genetic treatments. At 6 weeks, ~6-7 months and ~12 months of age, malin deficient mice ("KO") and wild type (WT) littermates underwent a standardized home-cage behavioral assessment designed to non-obtrusively appraise features of rest/arousal, consumptive behaviors, risk aversion and voluntary wheel-running. At all timepoints, and over a range of metrics that we report transparently, WT and KO mice were essentially indistinguishable. In contrast, within WT mice compared across timepoints, we identified age-related nocturnal hypoactivity, diminished sucrose preference and reduced wheel-running. Neuropathological examinations in subsets of the same mice revealed expected age dependent LB accumulation, gliosis and microglial activation in cortical and subcortical brain regions. At 12 months of age, despite the burden of neocortical LBs, we did not identify spontaneous seizures during an electroencephalographic (EEG) survey, and KO and WT mice exhibited similar spectral EEG features. Using an in vitro assay of neocortical function, paroxysmal increases in network activity (UP states) in KO slices were more prolonged at 3 and 6 months of age, but were similar to WT at 12 months. KO mice displayed a distinct response to pentylenetetrazole, with a greater incidence of clonic seizures and a more pronounced post-ictal suppression of movement, feeding and drinking behavior. Together, these results highlight a stark clinicopathologic dissociation in a mouse model of LD, where LBs accrue substantially without clinically meaningful changes in overall wellbeing. Our findings allude to a delay between LB accumulation and neurobehavioral decline: one that may provide a window for treatment, and whose precise duration may be difficult to ascertain within the typical lifespan of a laboratory mouse.

Case report: The gait deviation index may predict neurotherapeutic effects of FES-assisted gait training in children with cerebral palsy.

Background: Children with cerebral palsy (CP) show progressive loss of ambulatory function characterized by kinematic deviations at the hip, knee, and ankle. Functional electrical stimulation (FES) can lead to more typical lower limb kinematics during walking by eliciting appropriately timed muscle contractions. FES-assisted walking interventions have shown mixed to positive results in improving lower limb kinematics through immediate correction of gait during the application of FES, or long-term, persisting effects of non-FES-assisted gait improvements following multi-week FES-assisted gait training, at the absence of stimulation, i.e., neurotherapeutic effects. It is unknown, however, if children with CP will demonstrate a neurotherapeutic response following FES-assisted gait training because of the CP population's heterogeneity in gait deviations and responses to FES. Identifying the neurotherapeutic responders is, therefore, important to optimize the training interventions to those that have higher probability of benefiting from the intervention. Objective: The purpose of this case study was to investigate the relationship between immediate and neurotherapeutic effects of FES-assisted walking to identify responders to a FES-assisted gait training protocol. Methods: The primary outcome was Gait Deviation Index (GDI) and secondary outcome was root mean squared error (RMSE) of the lower extremity joint angles in the sagittal plane between participants with CP and a typically developing (TD) dataset. Potential indicators were defined as immediate improvements from baseline during FES-assisted walking followed by neurotherapeutic improvements at the end of training. Case description: Gait analysis of two adolescent female participants with spastic diplegia (Gross Motor Function Classification System level II and III) was conducted at the start and end of a 12-week FES-assisted treadmill training protocol. Participant 1 had scissoring crouch gait, while participant 2 had jump gait. Outcomes: The GDI showed both immediate (presence of FES) and neurotherapeutic (absence of FES after training period) improvements from baseline in our two participants. Joint angle RMSE showed mixed trends between immediate and neurotherapeutic changes from baseline. The GDI warrants investigation in a larger sample to determine if it can be used to identify responders to FES-assisted gait training.

Individuals with cerebral palsy show altered responses to visual perturbations during walking.

Individuals with cerebral palsy (CP) have deficits in processing of somatosensory and proprioceptive information. To compensate for these deficits, they tend to rely on vision over proprioception in single plane upper and lower limb movements and in standing. It is not known whether this also applies to walking, an activity where the threat to balance is higher. Through this study, we used visual perturbations to understand how individuals with and without CP integrate visual input for walking balance control. Additionally, we probed the balance mechanisms driving the responses to the visual perturbations. More specifically, we investigated differences in the use of ankle roll response i.e., the use of ankle inversion, and the foot placement response, i.e., stepping in the direction of perceived fall. Thirty-four participants (17 CP, 17 age-and sex-matched typically developing controls or TD) were recruited. Participants walked on a self-paced treadmill in a virtual reality environment. Intermittently, the virtual scene was rotated in the frontal plane to induce the sensation of a sideways fall. Our results showed that compared to their TD peers, the overall body sway in response to the visual perturbations was magnified and delayed in CP group, implying that they were more affected by changes in visual cues and relied more so on visual information for walking balance control. Also, the CP group showed a lack of ankle response, through a significantly reduced ankle inversion on the affected side compared to the TD group. The CP group showed a higher foot placement response compared to the TD group immediately following the visual perturbations. Thus, individuals with CP showed a dominant proximal foot placement strategy and diminished ankle roll response, suggestive of a reliance on proximal over distal control of walking balance in individuals with CP.

Home-cage behavior in the Stargazer mutant mouse.

In many childhood-onset genetic epilepsies, seizures are accompanied by neurobehavioral impairments and motor disability. In the Stargazer mutant mouse, genetic disruptions of Cacng2 result in absence-like spike-wave seizures, cerebellar gait ataxia and vestibular dysfunction, which limit traditional approaches to behavioral phenotyping. Here, we combine videotracking and instrumented home-cage monitoring to resolve the neurobehavioral facets of the murine Stargazer syndrome. We find that despite their gait ataxia, stargazer mutants display horizontal hyperactivity and variable rates of repetitive circling behavior. While feeding rhythms, circadian or ultradian oscillations in activity are unchanged, mutants exhibit fragmented bouts of behaviorally defined "sleep", atypical licking dynamics and lowered sucrose preference. Mutants also display an attenuated response to visual and auditory home-cage perturbations, together with profound reductions in voluntary wheel-running. Our results reveal that the seizures and ataxia of Stargazer mutants occur in the context of a more pervasive behavioral syndrome with elements of encephalopathy, repetitive behavior and anhedonia. These findings expand our understanding of the function of Cacng2.

Evaluation of Tensile Bond Strength between Self-Adhesive Resin Cement and Surface-Pretreated Zirconia.

The tensile bond strength between zirconia subjected to different surface-pretreatment methods and methacryloyloxydecyl-dihydrogen-phosphate (MDP)-containing self-adhesive resin cement was evaluated herein. Eighty-eight cylindrical zirconia specimens were randomly divided into the following four groups based on the pretreatment method: (1) no treatment, (2) air abrasion, (3) HNO3/HF etching, and (4) zirconia-nanoparticle coating. The tensile bond strength of the zirconia−resin-cement complexes was investigated. One-way ANOVA and post hoc tests were performed at a 95% significance level, and the Weibull modulus was calculated. Fracture patterns were visualized by SEM. The surface roughness of the specimens without resin bonding was evaluated by AFM. The tensile bond strength of the specimens decreased as follows: Groups 3 > 4 > 2 > 1 (28.2 ± 6.6, 26.1 ± 5.7, 16.6 ± 3.3, and 13.9 ± 3.0 MPa, respectively). Groups 3 and 4 had significantly higher tensile bond strengths (p < 0.05) and lower fracture probabilities than those of Groups 1 and 2. They also showed both mixed failure and resin-cement cohesive failure, whereas Groups 1 and 2 showed mixed failure exclusively. The zirconia−resin tensile bond was stronger after HNO3/HF etching or ZrO2-nanoparticle coating than after air abrasion or no treatment. The estimated surface roughness decreased as follows: Groups 3 > 4 > 2 > 1. The combination of zirconia pretreated with HNO3/HF etching or ZrO2-nanoparticle coating and an MDP-containing self-adhesive resin cement can increase the clinical longevity of zirconia restorations by preventing their decementation.

Aerobic Responses to FES-Assisted and Volitional Cycling in Children with Cerebral Palsy.

Recumbent stationary cycling is a potential exercise modality for individuals with cerebral palsy (CP) that lack the postural control needed for upright exercises. Functional electrical stimulation (FES) of lower extremity muscles can help such individuals reach the cycling intensities that are required for aerobic benefits. The aim of this study was to examine the effect of cycling with and without FES assistance to that of a no-intervention control group on the cardiorespiratory fitness of children with CP. Thirty-nine participants were randomized to a FES group that underwent an 8-week FES-assisted cycling program, the volitional group (VOL), who cycled without FES, or a no-intervention control group (CON) (15 FES, 11 VOL, 13 CON). Cadence, peak VO2, and net rise in heart rate were assessed at baseline, end of training, and washout (8-weeks after cessation of training). Latent growth curve modeling was used for analysis. The FES group showed significantly higher cycling cadences than the VOL and CON groups at POST and WO. There were no differences in improvements in the peak VO2 and peak net HR between groups. FES-assisted cycling may help children with CP attain higher cycling cadences and to retain these gains after training cessation. Higher training intensities may be necessary to obtain improvements in peak VO2 and heart rate.

Evaluation of Individualized Functional Electrical Stimulation-Induced Acute Changes during Walking: A Case Series in Children with Cerebral Palsy.

Functional electrical stimulation (FES) walking interventions have demonstrated improvements to gait parameters; however, studies were often confined to stimulation of one or two muscle groups. Increased options such as number of muscle groups targeted, timing of stimulation delivery, and level of stimulation are needed to address subject-specific gait deviations. We aimed to demonstrate the feasibility of using a FES system with increased stimulation options during walking in children with cerebral palsy (CP). Three physical therapists designed individualized stimulation programs for six children with CP to target participant-specific gait deviations. Stimulation settings (pulse duration and current) were tuned to each participant. Participants donned our custom FES system that utilized gait phase detection to control stimulation to lower extremity muscle groups and walked on a treadmill at a self-selected speed. Motion capture data were collected during walking with and without the individualized stimulation program. Eight gait metrics and associated timing were compared between walking conditions. The prescribed participant-specific stimulation programs induced significant change towards typical gait in at least one metric for each participant with one iteration of FES-walking. FES systems with increased stimulation options have the potential to allow the physical therapist to better target the individual's gait deviations than a one size fits all device.

Foot and ankle somatosensory deficits in children with cerebral palsy: A pilot study.

PURPOSE: To investigate foot and ankle somatosensory function in children with cerebral palsy (CP). METHODS: Ten children with spastic diplegia (age 15 ± 5 y; GMFCS I-III) and 11 typically developing (TD) peers (age 15 ± 10 y) participated in the study. Light touch pressure and two-point discrimination were assessed on the plantar side of the foot by using a monofilament kit and an aesthesiometer, respectively. The duration of vibration sensation at the first metatarsal head and medial malleolus was tested by a 128 Hz tuning fork. Joint position sense and kinesthesia in the ankle joint were also assessed. RESULTS: Children with CP demonstrated significantly higher light touch pressure and two-point discrimination thresholds compared to their TD peers. Individuals with CP perceived the vibration stimulus for a longer period compared to the TD participants. Finally, the CP group demonstrated significant impairments in joint position sense but not in kinesthesia of the ankle joints. CONCLUSIONS: These findings suggest that children with CP have foot and ankle tactile and proprioceptive deficits. Assessment of lower extremity somatosensory function should be included in clinical practice as it can guide clinicians in designing more effective treatment protocols to improve functional performance in CP.

PAN: Personalized Annotation-Based Networks for the Prediction of Breast Cancer Relapse.

The classification of clinical samples based on gene expression data is an important part of precision medicine. In this manuscript, we show how transforming gene expression data into a set of personalized (sample-specific) networks can allow us to harness existing graph-based methods to improve classifier performance. Existing approaches to personalized gene networks have the limitation that they depend on other samples in the data and must get re-computed whenever a new sample is introduced. Here, we propose a novel method, called Personalized Annotation-based Networks (PAN), that avoids this limitation by using curated annotation databases to transform gene expression data into a graph. Unlike competing methods, PANs are calculated for each sample independent of the population, making it a more efficient way to obtain single-sample networks. Using three breast cancer datasets as a case study, we show that PAN classifiers not only predict cancer relapse better than gene features alone, but also outperform PPI (protein-protein interactions) and population-level graph-based classifiers. This work demonstrates the practical advantages of graph-based classification for high-dimensional genomic data, while offering a new approach to making sample-specific networks. Supplementary information: PAN and the baselines are implemented in Python. Source code and data are available at https://github.com/thinng/PAN.

The characterization of distinct populations of murine skeletal cells that have different roles in B lymphopoiesis.

Hematopoiesis is extrinsically controlled by cells of the bone marrow microenvironment, including skeletal lineage cells. The identification and subsequent studies of distinct subpopulations of maturing skeletal cells is currently limited because of a lack of methods to isolate these cells. We found that murine Lin-CD31-Sca-1-CD51+ cells can be divided into 4 subpopulations by using flow cytometry based on their expression of the platelet-derived growth factor receptors ⍺ and ß (PDGFR⍺ and PDGFRß). The use of different skeletal lineage reporters confirmed the skeletal origin of the 4 populations. Multiplex immunohistochemistry studies revealed that all 4 populations were localized near the growth plate and trabecular bone and were rarely found near cortical bone regions or in central bone marrow. Functional studies revealed differences in their abundance, colony-forming unit-fibroblast capacity, and potential to differentiate into mineralized osteoblasts or adipocytes in vitro. Furthermore, the 4 populations had distinct gene expression profiles and differential cell surface expression of leptin receptor (LEPR) and vascular cell adhesion molecule 1 (VCAM-1). Interestingly, we discovered that 1 of these 4 different skeletal populations showed the highest expression of genes involved in the extrinsic regulation of B lymphopoiesis. This cell population varied in abundance between distinct hematopoietically active skeletal sites, and significant differences in the proportions of B-lymphocyte precursors were also observed in these distinct skeletal sites. This cell population also supported pre-B lymphopoiesis in culture. Our method of isolating 4 distinct maturing skeletal populations will help elucidate the roles of distinct skeletal niche cells in regulating hematopoiesis and bone.

Characterizing Cycling Smoothness and Rhythm in Children With and Without Cerebral Palsy.

Stationary cycling is a practical exercise modality in children with cerebral palsy (CP) that lack the strength for upright exercises. However, there is a lack of robust, sensitive metrics that can quantitatively assess the motor control during cycling. The purpose of this brief report was to characterize the differences in motor control of cycling in children with CP and with typical development by developing novel metrics to quantify cycling smoothness and rhythm. Thirty one children with spastic diplegic CP and 10 children with typical development cycled on a stationary cycle. Cycling smoothness was measured by cross-correlating the crank angle with an ideal cycling pattern generated from participant-specific cadence and cycling duration. Cycling rhythmicity was assessed by evaluating the revolution-to-revolution variability in the time required to complete a revolution. Statistically significant differences (p < 0.001) using the Wilcoxon Rank Sum test were found between the two groups for both the metrics. Additionally, decision tree analysis revealed thresholds of smoothness <0.01 and rhythm <0.089-0.115 s for discriminating a less smooth, irregular cycling pattern characteristic of CP from typical cycling. In summary, the objective measures developed in this study indicate significantly less smoothness and rhythm of cycling in children with CP compared to children with typical development, suggestive of altered coordination and poor motor control. Such quantitative assessments of cycling motion in children with CP provide insights into neuromotor deficits that prevent them from cycling at intensities required for aerobic benefits and for participating in cycling related physical activities with their peers.

An Evaluation of Three Kinematic Methods for Gait Event Detection Compared to the Kinetic-Based 'Gold Standard'.

Video- and sensor-based gait analysis systems are rapidly emerging for use in 'real world' scenarios outside of typical instrumented motion analysis laboratories. Unlike laboratory systems, such systems do not use kinetic data from force plates, rather, gait events such as initial contact (IC) and terminal contact (TC) are estimated from video and sensor signals. There are, however, detection errors inherent in kinematic gait event detection methods (GEDM) and comparative study between classic laboratory and video/sensor-based systems is warranted. For this study, three kinematic methods: coordinate based treadmill algorithm (CBTA), shank angular velocity (SK), and foot velocity algorithm (FVA) were compared to 'gold standard' force plate methods (GS) for determining IC and TC in adults (n = 6), typically developing children (n = 5) and children with cerebral palsy (n = 6). The root mean square error (RMSE) values for CBTA, SK, and FVA were 27.22, 47.33, and 78.41 ms, respectively. On average, GED was detected earlier in CBTA and SK (CBTA: -9.54 ± 0.66 ms, SK: -33.41 ± 0.86 ms) and delayed in FVA (21.00 ± 1.96 ms). The statistical model demonstrated insensitivity to variations in group, side, and individuals. Out of three kinematic GEDMs, SK GEDM can best be used for sensor-based gait event detection.

Foot and Ankle Somatosensory Deficits Affect Balance and Motor Function in Children With Cerebral Palsy.

Sensory dysfunction is prevalent in cerebral palsy (CP). Evidence suggests that sensory deficits can contribute to manual ability impairments in children with CP, yet it is still unclear how they contribute to balance and motor performance. Therefore, the objective of this study was to investigate the relationship between lower extremity (LE) somatosensation and functional performance in children with CP. Ten participants with spastic diplegia (Gross Motor Function Classification Scale: I-III) and who were able to stand independently completed the study. Threshold of light touch pressure, two-point discriminatory ability of the plantar side of the foot, duration of cutaneous vibration sensation, and error in the joint position sense of the ankle were assessed to quantify somatosensory function. The balance was tested by the Balance Evaluation System Test (BESTest) and postural sway measures during a standing task. Motor performance was evaluated by using a battery of clinical assessments: (1) Gross Motor Function Measure (GMFM-66-IS) to test gross motor ability; (2) spatiotemporal gait characteristics (velocity, step length) to evaluate walking ability; (3) Timed Up and Go (TUG) and 6 Min Walk (6MWT) tests to assess functional mobility; and (4) an isokinetic dynamometer was used to test the Maximum Volitional Isometric Contraction (MVIC) of the plantar flexor muscles. The results showed that the light touch pressure measure was strongly associated only with the 6MWT. Vibration and two-point discrimination were strongly related to balance performance. Further, the vibration sensation of the first metatarsal head demonstrated a significantly strong relationship with motor performance as measured by GMFM-66-IS, spatiotemporal gait parameters, TUG, and ankle plantar flexors strength test. The joint position sense of the ankle was only related to one subdomain of the BESTest (Postural Responses). This study provides preliminary evidence that LE sensory deficits can possibly contribute to the pronounced balance and motor impairments in CP. The findings emphasize the importance of developing a thorough LE sensory test battery that can guide traditional treatment protocols toward a more holistic therapeutic approach by combining both motor and sensory rehabilitative strategies to improve motor function in CP.

DeepTRIAGE: interpretable and individualised biomarker scores using attention mechanism for the classification of breast cancer sub-types.

BACKGROUND: Breast cancer is a collection of multiple tissue pathologies, each with a distinct molecular signature that correlates with patient prognosis and response to therapy. Accurately differentiating between breast cancer sub-types is an important part of clinical decision-making. Although this problem has been addressed using machine learning methods in the past, there remains unexplained heterogeneity within the established sub-types that cannot be resolved by the commonly used classification algorithms. METHODS: In this paper, we propose a novel deep learning architecture, called DeepTRIAGE (Deep learning for the TRactable Individualised Analysis of Gene Expression), which uses an attention mechanism to obtain personalised biomarker scores that describe how important each gene is in predicting the cancer sub-type for each sample. We then perform a principal component analysis of these biomarker scores to visualise the sample heterogeneity, and use a linear model to test whether the major principal axes associate with known clinical phenotypes. RESULTS: Our model not only classifies cancer sub-types with good accuracy, but simultaneously assigns each patient their own set of interpretable and individualised biomarker scores. These personalised scores describe how important each feature is in the classification of any patient, and can be analysed post-hoc to generate new hypotheses about latent heterogeneity. CONCLUSIONS: We apply the DeepTRIAGE framework to classify the gene expression signatures of luminal A and luminal B breast cancer sub-types, and illustrate its use for genes as well as the GO and KEGG gene sets. Using DeepTRIAGE, we calculate personalised biomarker scores that describe the most important features for classifying an individual patient as luminal A or luminal B. In doing so, DeepTRIAGE simultaneously reveals heterogeneity within the luminal A biomarker scores that significantly associate with tumour stage, placing all luminal samples along a continuum of severity.

A cross-cancer metastasis signature in the microRNA-mRNA axis of paired tissue samples.

In the progression of cancer, cells acquire genetic mutations that cause uncontrolled growth. Over time, the primary tumour may undergo additional mutations that allow for the cancerous cells to spread throughout the body as metastases. Since metastatic development typically results in markedly worse patient outcomes, research into the identity and function of metastasis-associated biomarkers could eventually translate into clinical diagnostics or novel therapeutics. Although the general processes underpinning metastatic progression are understood, no clear cross-cancer biomarker profile has emerged. However, the literature suggests that some microRNAs (miRNAs) may play an important role in the metastatic progression of several cancer types. Using a subset of The Cancer Genome Atlas (TCGA) data, we performed an integrated analysis of mRNA and miRNA expression with paired metastatic and primary tumour samples to interrogate how the miRNA-mRNA regulatory axis influences metastatic progression. From this, we successfully built mRNA- and miRNA-specific classifiers that can discriminate pairs of metastatic and primary samples across 11 cancer types. In addition, we identified a number of miRNAs whose metastasis-associated dysregulation could predict mRNA metastasis-associated dysregulation. Among the most predictive miRNAs, we found several previously implicated in cancer progression, including miR-301b, miR-1296, and miR-423. Taken together, our results suggest that metastatic samples have a common cross-cancer signature when compared with their primary tumour pair, and that these miRNA biomarkers can be used to predict metastatic status as well as mRNA expression.