BMC biomedical engineering最新文献

Background: CS-MRI (compressed sensing for magnetic resonance imaging) exploits image sparsity properties to reconstruct MRI from very few Fourier k-space measurements. Due to imperfect modelings in the inverse imaging, state-of-the-art CS-MRI methods tend to leave structural reconstruction errors. Compensating such errors in the reconstruction could help further improve the reconstruction quality.

Results: In this work, we propose a DECN (deep error correction network) for CS-MRI. The DECN model consists of three parts, which we refer to as modules: a guide, or template, module, an error correction module, and a data fidelity module. Existing CS-MRI algorithms can serve as the template module for guiding the reconstruction. Using this template as a guide, the error correction module learns a CNN (convolutional neural network) to map the k-space data in a way that adjusts for the reconstruction error of the template image. We propose a deep error correction network. Our experimental results show the proposed DECN CS-MRI reconstruction framework can considerably improve upon existing inversion algorithms by supplementing with an error-correcting CNN.

Conclusions: In the proposed a deep error correction framework, any off-the-shelf CS-MRI algorithm can be used as template generation. Then a deep neural network is used to compensate reconstruction errors. The promising experimental results validate the effectiveness and utility of the proposed framework.

Background: The soft tissue of the residual limb in transtibial prosthetic users encounters unique biomechanical challenges. Although not intended to tolerate high loads and deformation, it becomes a weight-bearing structure within the residuum-prosthesis-complex. Consequently, deep soft tissue layers may be damaged, resulting in Deep Tissue Injury (DTI). Whilst considerable effort has gone into DTI research on immobilised individuals, only little is known about the aetiology and population-specific risk factors in amputees. This scoping review maps out and critically appraises existing research on DTI in lower-limb prosthetic users according to (1) the population-specific aetiology, (2) risk factors, and (3) methodologies to investigate both.

Results: A systematic search within the databases Pubmed, Ovid Excerpta Medica, and Scopus identified 16 English-language studies. The results indicate that prosthetic users may be at risk for DTI during various loading scenarios. This is influenced by individual surgical, morphological, and physiological determinants, as well as the choice of prosthetic componentry. However, methodological limitations, high inter-patient variability, and small sample sizes complicate the interpretation of outcome measures. Additionally, fundamental research on cell and tissue reactions to dynamic loading and on prosthesis-induced alterations of the vascular and lymphatic supply is missing.

Conclusion: We therefore recommend increased interdisciplinary research endeavours with a focus on prosthesis-related experimental design to widen our understanding of DTI. The results have the potential to initiate much-needed clinical advances in surgical and prosthetic practice and inform future pressure ulcer classifications and guidelines.

Background: Strength and coordination of lower muscle groups typically identified in healthy subjects are two prerequisites to performing functional activities. These physical qualities can be impaired following a neurological insult. A static dynamometer apparatus that measures lower limb joint moments during directional efforts at the foot was developed to recruit different patterns of muscular activity. The objectives of the present study were to 1) validate joint moments estimated by the apparatus, and 2) to characterize lower limb joint moments and muscular activity patterns of healthy subjects during progressive static efforts. Subjects were seated in a semi-reclined position with one foot attached to a force platform interfaced with a laboratory computer. Forces and moments exerted under the foot were computed using inverse dynamics, allowing for the estimation of lower limb joint moments.To achieve the study's first objective, joint moments were validated by comparing moments of various magnitudes of force applied by turnbuckles on an instrumented leg equipped with strain gauges with those estimated by the apparatus. Concurrent validity and agreement were assessed using Pearson correlation coefficients and Bland and Altman analysis, respectively. For the second objective, joint moments and muscular activity were characterized for five healthy subjects while exerting progressive effort in eight sagittal directions. Lower limb joint moments were estimated during directional efforts using inverse dynamics. Muscular activity of eight muscles of the lower limb was recorded using surface electrodes and further analyzed using normalized root mean square data.

Results: The joint moments estimated with the instrumented leg were correlated (r > 0.999) with those measured by the dynamometer. Limits of agreement ranged between 8.5 and 19.2% of the average joint moment calculated by both devices. During progressive efforts on the apparatus, joint moments and patterns of muscular activity were specific to the direction of effort. Patterns of muscular activity in four directions were similar to activation patterns reported in the literature for specific portions of gait cycle.

Conclusion: This apparatus provides valid joint moments exerted at the lower limbs. It is suggested that this methodology be used to recruit muscular activity patterns impaired in neurological populations.

Despite advances in cancer therapeutics, particularly in the area of immuno-oncology, successful treatment of neuroblastoma (NB) remains a challenge. NB is the most common cancer in infants under 1 year of age, and accounts for approximately 10% of all pediatric cancers. Currently, children with high-risk NB exhibit a survival rate of 40-50%. The heterogeneous nature of NB makes development of effective therapeutic strategies challenging. Many preclinical models attempt to mimic the tumor phenotype and tumor microenvironment. In vivo mouse models, in the form of genetic, syngeneic, and xenograft mice, are advantageous as they replicated the complex tumor-stroma interactions and represent the gold standard for preclinical therapeutic testing. Traditional in vitro models, while high throughput, exhibit many limitations. The emergence of new tissue engineered models has the potential to bridge the gap between in vitro and in vivo models for therapeutic testing. Therapeutics continue to evolve from traditional cytotoxic chemotherapies to biologically targeted therapies. These therapeutics act on both the tumor cells and other cells within the tumor microenvironment, making development of preclinical models that accurately reflect tumor heterogeneity more important than ever. In this review, we will discuss current in vitro and in vivo preclinical testing models, and their potential applications to therapeutic development.

[This corrects the article DOI: 10.1186/s42490-019-0015-y.].

Background: Arm support devices are available to support people with Duchenne muscular dystrophy (DMD), but active trunk support devices are lacking. An active trunk support device can potentially extend the reach of the arm and stabilize the unstable trunk of people with DMD. In a previous study, we showed that healthy people were able to control an active trunk support using four different control interfaces (based on joystick, force on feet, force on sternum and surface electromyography). All four control interfaces had different advantages and disadvantages. The aim of this study was to explore which of the four inputs is detectably used by people with DMD to control an active trunk support.

Results: The results were subject-dependent in both experiments. In the active experiment, the joystick was the most promising control interface. Regarding the static experiment, surface electromyography and force on feet worked for two out of the three subjects.

Conclusions: To our knowledge, this is the first time that people with DMD have engaged in a control task using signals other than those related to their arm muscles. According to our findings, the control interfaces have to be customised to every DMD subject.

Muscles contain contractile and (visco-) elastic passive components. At the latest since Hill's classic works in the 1930s, it has been known that these elastic components affect the length and rate of change in length of the contractile component, and thus the active force capability of dynamically working muscles. In an attempt to elucidate functional properties of these muscle elastic components, scientists have introduced the notion of "series" and "parallel" elasticity. Unfortunately, this has led to much confusion and erroneous interpretations of results when the mechanical definitions of parallel and series elasticity were violated. In this review, I will focus on muscle series elasticity, by first providing the mechanical definition for series elasticity, and then provide theoretical and experimental examples of the concept of series elasticity. Of particular importance is the treatment of aponeuroses. Aponeuroses are not in series with the tendon of a muscle nor the muscle's contractile elements. The implicit and explicit treatment of aponeuroses as series elastic elements in muscle has led to incorrect conclusions about aponeuroses stiffness and Young's modulus, and has contributed to vast overestimations of the storage and release of mechanical energy in cyclic muscle contractions. Series elasticity is a defined mechanical concept that needs to be treated carefully when applied to skeletal muscle mechanics. Measuring aponeuroses mechanical properties in a muscle, and its possible contribution to the storage and release of mechanical energy is not trivial, and to my best knowledge, has not been (correctly) done yet.

Background: Hybrid exoskeletons are a recent development which combine Functional Electrical Stimulation with actuators to improve both the mental and physical rehabilitation of stroke patients. Hybrid exoskeletons have been shown capable of reducing the weight of the actuator and improving movement precision compared to Functional Electrical Stimulation alone. However little attention has been given towards the ability of hybrid exoskeletons to reduce and manage Functional Electrical Stimulation induced fatigue or towards adapting to user ability. This work details the construction and testing of a novel assist-as-need upper-extremity hybrid exoskeleton which uses model-based Functional Electrical Stimulation control to delay Functional Electrical Stimulation induced muscle fatigue. The hybrid control is compared with Functional Electrical Stimulation only control on a healthy subject.

Results: The hybrid system produced 24° less average angle error and 13.2° less Root Mean Square Error, than Functional Electrical Stimulation on its own and showed a reduction in Functional Electrical Stimulation induced fatigue.

Conclusion: As far as the authors are aware, this is the study which provides evidence of the advantages of hybrid exoskeletons compared to use of Functional Electrical Stimulation on its own with regards to the delay of Functional Electrical Stimulation induced muscle fatigue.

Background: Triple tracer meal experiments used to investigate organ glucose-insulin dynamics, such as endogenous glucose production (EGP) of the liver are labor intensive and expensive. A procedure was developed to obtain individual liver related parameters to describe EGP dynamics without the need for tracers.

Results: The development used an existing formula describing the EGP dynamics comprising 4 parameters defined from glucose, insulin and C-peptide dynamics arising from triple meal studies. The method employs a set of partial differential equations in order to estimate the parameters for EGP dynamics. Tracer-derived and simulated data sets were used to develop and test the procedure. The predicted EGP dynamics showed an overall mean R ² of 0.91.

Conclusions: In summary, a method was developed for predicting the hepatic EGP dynamics for healthy, pre-diabetic, and type 2 diabetic individuals without applying tracer experiments.

Background: Since nuclei segmentation in histopathology images can provide key information for identifying the presence or stage of a disease, the images need to be assessed carefully. However, color variation in histopathology images, and various structures of nuclei are two major obstacles in accurately segmenting and analyzing histopathology images. Several machine learning methods heavily rely on hand-crafted features which have limitations due to manual thresholding.

Results: To obtain robust results, deep learning based methods have been proposed. Deep convolutional neural networks (DCNN) used for automatically extracting features from raw image data have been proven to achieve great performance. Inspired by such achievements, we propose a nuclei segmentation method based on DCNNs. To normalize the color of histopathology images, we use a deep convolutional Gaussian mixture color normalization model which is able to cluster pixels while considering the structures of nuclei. To segment nuclei, we use Mask R-CNN which achieves state-of-the-art object segmentation performance in the field of computer vision. In addition, we perform multiple inference as a post-processing step to boost segmentation performance. We evaluate our segmentation method on two different datasets. The first dataset consists of histopathology images of various organ while the other consists histopathology images of the same organ. Performance of our segmentation method is measured in various experimental setups at the object-level and the pixel-level. In addition, we compare the performance of our method with that of existing state-of-the-art methods. The experimental results show that our nuclei segmentation method outperforms the existing methods.

Conclusions: We propose a nuclei segmentation method based on DCNNs for histopathology images. The proposed method which uses Mask R-CNN with color normalization and multiple inference post-processing provides robust nuclei segmentation results. Our method also can facilitate downstream nuclei morphological analyses as it provides high-quality features extracted from histopathology images.