Using Empirical Mode Decomposition with Spatio-Temporal dynamics to classify single-trial Motor Imagery in BCI.

This paper introduces a new signal processing method called Spatio-Temporal Multivariate Empirical Mode Decomposition (ST-MEMD). It is a new variation of Empirical Mode Decomposition (EMD) that takes spatial and temporal information into account simultaneously rather than processing each signal source in isolation. The original and new methods were tested on single-trial electroencephalogram data with a two-class problem, in this case data using the Motor Imagery paradigm in brain-computer interfacing. However, whilst ST-MEMD retained the increase in sensitivity and specificity from adding spatial data, the new temporal data made no meaningful difference in terms of performance.

Novel use of Empirical Mode Decomposition in single-trial classification of motor imagery for use in brain-computer interfaces.

This paper presents a novel method, based on multi-channel Empirical Mode Decomposition (EMD), of classifying the electroencephalogram (EEG) recordings of imagined movement by a subject within a brain-computer interfacing (BCI) framework. EMD is a technique that divides any non-linear or non-stationary signal into groups of frequency harmonics, called Intrinsic Mode Functions (IMFs). As frequency is a key component of both IMFs and the µ rhythm (8-13 Hz brain activity generated during motor imagery), IMFs are then grouped by frequency. EMD is applied to the recordings from two electrodes for each trial and the resulting IMFs are grouped according to peak-frequency band via Hierarchical Clustering Analysis (HCA). The cluster containing the frequency band of the µ rhythm (8-13 Hz) is then selected and the sum-total of the IMFs from each electrode are summed together. A simple linear classifier is then sufficient to classify the motor-imagery with 89% sensitivity from a separate test set.

The plasminogen activation system in lung disease.

The importance of the plasminogen activator (PA) system in multiple pulmonary disorders has become increasingly apparent as methods to analyze its components have improved. Early investigations discovered that the pulmonary alveolar space is normally a pro-fibrinolytic environment that is diminished in a variety of lung diseases. Interest in these observations was greatly increased when animal experiments revealed that manipulations of the PA system significantly modulated the tissue fibrosis that follows many types of lung injury. In particular, enhancement of PA activity was found to consistently decrease the extent of scarring induced by lung damage. Based upon these early observations, it was hypothesized that fibrin was necessary for the pathogenesis of lung fibrosis, and that an increase in PA activity would reduce collagen accumulation by accelerating the clearance of fibrin from the provisional matrix. However, as is often the case with simple hypotheses, subsequent studies revealed that the actual role of the PA system in pulmonary disease is much more complex. Possible mechanisms beyond fibrinolysis include degradation of other matrix proteins, activation of protease cascades including those involving matrix metalloproteinases, activation and release of growth factors from sites of production and sequestration, and modulation of cell adhesion and motility. In each of these processes, the serpin plasminogen activator inhibitor-1 (PAI-1) plays a central role. For these reasons, it has become apparent that PAI-1 presents an attractive target to influence multiple disease processes within the lung, particularly those that lead to lung fibrosis.

Bleomycin-induced pulmonary fibrosis in fibrinogen-null mice.

Mice deleted for the plasminogen activator inhibitor-1 (PAI-1) gene are relatively protected from developing pulmonary fibrosis induced by bleomycin. We hypothesized that PAI-1 deficiency reduces fibrosis by promoting plasminogen activation and accelerating the clearance of fibrin matrices that accumulate within the damaged lung. In support of this hypothesis, we found that the lungs of PAI-1(-/-) mice accumulated less fibrin after injury than wild-type mice, due in part to enhanced fibrinolytic activity. To further substantiate the importance of fibrin removal as the mechanism by which PAI-1 deficiency limited bleomycin-induced fibrosis, bleomycin was administered to mice deficient in the gene for the Aalpha-chain of fibrinogen (fib). Contrary to our expectation, fib(-/-) mice developed pulmonary fibrosis to a degree similar to fib(+/-) littermate controls, which have a plasma fibrinogen level that is 70% of that of wild-type mice. Although elimination of fibrin from the lung was not in itself protective, the beneficial effect of PAI-1 deficiency was still associated with proteolytic activity of the plasminogen activation system. In particular, inhibition of plasmin activation and/or activity by tranexamic acid reversed both the accelerated fibrin clearance and the protective effect of PAI-1 deficiency. We conclude that protection from fibrosis by PAI-1 deficiency is dependent upon increased proteolytic activity of the plasminogen activation system; however, complete removal of fibrin is not sufficient to protect the lung.

Lung disease associated with the IVS8 5T allele of the CFTR gene.

Cystic fibrosis is caused by mutations in the cystic fibrosis transmembrane regulator (CFTR) gene. The 5T allele in intron 8 (IVS8) causes abnormal splicing in the CFTR gene, and is associated with lung disease when it occurs in cis with a missense mutation in the CFTR gene, R117H. However, the 5T variant alone has not been reported to cause lung disease. We describe two adult female patients with CF-like lung disease associated with the 5T allele. One patient's genotype is 5T-TG12-M470V/5T-TG12-M470V, and the other is DeltaF508/5T-TG12-M470V; full sequencing of the CFTR gene revealed no other mutation on the same allele as the 5T variant. The levels of full-length CFTR mRNA in respiratory epithelia were very low in these patients (11 and 6%, respectively, of total CFTR mRNA expression). Both patients had defective CFTR-mediated chloride conductance in the sweat ductal and/or acinar epithelia (sweat chloride, mmol/L, mean +/- SEM: 40.0 +/- 5.0 [n = 8 samples] and 80. 0 +/- 3.5 [n = 6 samples]) and airway epithelia (mV, mean +/- SEM CFTR-mediated Cl(-) conductance of 1.2 +/- 2.2 [n = 5 studies] and -6.75 +/- 8.1 [n = 4 studies]). These data suggest that the 5T polythymidine tract sequence on specific haplotype backgrounds (TG12 and M470V) may cause a low level of full-length functional CFTR protein and CF-like lung disease.

Chronic subdural hematoma in adult and elderly patients.

Chronic subdural hematomas display a wide diversity of features in adults, often simulating other neurologic and psychiatric disease processes. This is particularly problematic in elderly patients, for whom more nonspecific clinical presentations are common. Recognition of these features is the key to early diagnosis in elderly patients, who exhibit a markedly increased incidence of this disease. Prompt diagnosis prevents delays that may lead to increased morbidity in these patients.

The development of bleomycin-induced pulmonary fibrosis in mice deficient for components of the fibrinolytic system.

Acute and chronic pulmonary diseases are characterized by impaired fibrinolytic activity within the lung. To determine the role of the fibrinolytic system in regulating the pathologies associated with lung injury, we examined the effect of bleomycin, an agent that induces the development of pulmonary fibrosis, in mice deficient for plasminogen (Pg(-)(/-)), urokinase (u-PA(-)(/-)), urokinase receptor (u-PAR(-)(/-)), or tissue plasminogen activator (t-PA(-)(/-)), and in control wild-type (WT) mice. Pg(-)(/-) and t-PA(-)(/-) mice demonstrated an enhanced increase in lung collagen content relative to that observed in WT mice. Levels in u-PA(-)(/-) and u-PAR(-)(/-) mice were similar to those in WT mice. Histological analysis 14 days after lung injury confirmed enhanced interstitial fibrosis in Pg(-)(/-), u-PA(-)(/-), and t-PA(-)(/-) mice relative to WT and u-PAR(-)(/-) mice. Areas of pulmonary hemorrhage were observed in bleomycin-treated WT mice and not in Pg(-)(/-), u-PA(-)(/-), and u-PAR(-)(/-) mice or saline controls. Instead, extensive areas of fibrosis were present throughout the lungs of bleomycin-treated Pg(-)(/-) and u-PA(-)(/-) mice. A mixed phenotype (hemorrhage and fibrosis) was observed in t-PA(-)(/-) and Pg(+/-) mice. Hemosiderin-laden macrophages were abundant in the lungs of mice exhibiting hemorrhage and these mice were prone to an early death. Enhanced macrophage levels in the lungs and activation of matrix metalloelastase (MMP-12) were found in mice with a hemorrhage phenotype. The results of these studies indicate a role for the fibrinolytic system in acute lung injury and suggests that intra-alveolar hemorrhage is the result of basement membrane degradation through cell-mediated u-PA activation of Pg with possible involvement of matrix metalloproteinases. Absence of these two components of the fibrinolytic system, either urokinase or plasminogen, results in accelerated fibrosis.

Treatment of bleomycin-induced pulmonary fibrosis by transfer of urokinase-type plasminogen activator genes.

During acute and chronic inflammatory lung diseases, the normal fibrinolytic activity in the alveolar space is inhibited by increased levels of plasminogen activator inhibitor 1 (PAI-1). Transgenic mice having increased fibrinolytic activity due to genetic deficiency of PAI-1 develop less fibrosis after bleomycin-induced lung inflammation. These observations led us to hypothesize that pulmonary fibrosis could be limited through enhancement of alveolar fibrinolytic activity by adenovirus-mediated transfer of the urokinase-type plasminogen activator (uPA) gene to the lung. To investigate this hypothesis, 0.075 U of bleomycin was introduced intratracheally into mice. Twenty-one days later, the mice were treated intratracheally with phosphate-buffered saline (PBS), a control adenovirus, or adenoviruses containing murine or human uPA cDNAs. On day 28, the mice were sacrificed, and lung fibrosis was quantitated by measuring hydroxyproline content. As expected, bleomycin caused a doubling in lung hydroxyproline to 345.6+/-28.2 microg/lung (SEM) compared with mice receiving PBS (170.2+/-4.0 microg/lung). Treatment of the bleomycin-injured mice with the control adenovirus on day 21 had no impact on lung fibrosis (338.4+/-17.2 microg/lung). Importantly, the human uPA adenovirus significantly reduced (p<0.05) lung hydroxyproline (281.2+/-22.8 microg/lung), thus attenuating by 38% the bleomycin-induced increase in lung collagen. The improvement in bleomycin-induced lung fibrosis resulting from treatment with the human uPA adenovirus further supports the importance of the fibrinolytic system during inflammatory lung injury and repair.

Participation of urokinase-type plasminogen activator receptor in the clearance of fibrin from the lung.

In vitro studies have demonstrated that the binding of urokinase-type plasminogen activator (uPA) to its cell surface receptor (uPAR) greatly accelerates plasminogen activation. However, the role of uPAR in clearing abnormal fibrin deposits from the lung is uncertain. Knowing that uPA binding to uPAR is species specific, we used adenoviral vectors to transfer human or murine uPA genes into human or mouse epithelial cells in vitro and to mouse lungs in vivo. By measuring degradation of fluorescein-labeled fibrin, we found that uPA lysed fibrin matrices more efficiently when expressed in cells of the same species. A monoclonal antibody that blocks the binding of human uPA to human uPAR suppressed fibrin degradation by human cells expressing human uPA but not murine uPA. Importantly, 3 days after intratracheal delivery of the vectors, mice receiving murine uPA transgenes degraded fibrin matrices formed within their air spaces more efficiently than animals transduced with human uPA genes. These results show that uPA bound to uPAR increases the efficiency of fibrinolysis on epithelial cell surfaces in a biologically relevant fashion.

The affinity of lipid-coated microbubbles to maturing spinal cord injury sites.

OBJECTIVE: This laboratory has demonstrated that lipid-coated microbubbles (LCMs) effectively aggregate and deliver chemotherapeutic drugs into rat brain tumor cells and antigliosis agents into maturing rat brain injury sites. In this study, we report the affinity of tail vein-injected LCMs to the injured rat spinal cord by a compressive lesion to the upper thoracic region. METHODS: The accumulation of LCMs in the injured spinal cord was analyzed by labeling it with a lipid-soluble fluorescent dye, 3,3'-dioctadecyloxacarbocyanine perchlorate. Indices of glial fibrillary acidic protein were measured concomitantly with 3,3'-dioctadecyloxacarbocyanine perchlorate-labeled LCMs using confocal microscopy. RESULTS: There was no aggregation of LCMs accumulated 1 and 6 hours after injury; however, when given 2, 4, and 7 days after injury, LCMs showed a clear affinity for the injured region. LCM aggregation shifted from the central necrotic area of the injury on postinjury Day 2 and postinjury Day 4 to the white matter among glial fibrillary acidic protein-positive astrocytes by postinjury Day 7. CONCLUSION: Affinity of LCMs for spinal cord injury sites may be mediated in the early stages after injury by proliferating macrophages in the necrotic center, and then in later stages by glial fibrillary acidic protein-positive astrocytes in adjacent white matter. These findings suggest a potential for using LCMs as a delivery vehicle to concentrate lipid-soluble agents in spinal cord injury sites.

Upregulation of fibrinolysis by adenovirus-mediated transfer of urokinase-type plasminogen activator genes to lung cells in vitro and in vivo.

Impaired fibrinolytic activity within the lungs is a common manifestation of acute and chronic inflammatory lung diseases. Our previous work using transgenic mice showed that upregulation of fibrinolysis reduced pulmonary fibrosis following bleomycin-induced inflammatory lung injury. As a strategy to accelerate fibrinolysis, we generated recombinant adenoviruses containing human and mouse urokinase-type plasminogen activator (uPA) cDNAs. Both vectors induced the expression of functional uPA in human lung-derived epithelial A549 cells. A single intratracheal instillation of these uPA-containing adenoviruses into mouse lungs resulted in increased plasminogen activator activity in bronchoalveolar lavage fluid for at least 2 weeks. Plasma-derived fibrin-rich matrices overlaid on A549 cells infected with these uPA vectors were lysed efficiently in a dose-dependent fashion. Similarly, fibrin matrices formed within intact lungs that had been infected with these uPA-containing adenoviruses were also lysed more rapidly compared with noninfected and control virus-infected lungs. These results indicate that adenovirus-mediated transduction of uPA successfully upregulates fibrinolysis in vitro and in vivo. These uPA vectors can be readily used for testing the role of the fibrinolytic system in animal models of lung fibrosis, with particular attention to their therapeutic potential.

Magnetic resonance imaging of a mediastinal ependymoma.

Ependymomas typically occur in the brain and spinal cord, and represent 5% and 16% of primary intracranial and spinal cord tumors respectively. Ependymomas arising outside the central nervous systems are uncommon and when they do occur are usually seen in the sacrococcygeal region. Other reported sites include the pelvis, spinal nerve roots, and lung. An extraspinal mediastinal ependymoma is a rare occurrence, with only two previously reported cases to our knowledge. In both instances, the lesion arose in the posterior mediastinum, and were imaged with either chest radiographs alone or chest radiographs and computed tomography. In our case, plain radiography was followed by magnetic resonance imaging. This report documents the clinical, pathological, and magnetic resonance imaging findings in a case of a posterior mediastinal ependymoma. Primary mediastinal ependymomas are rare lesions that may be considered in the differential diagnosis of a posterior mediastinal mass.

The use of lipid-coated microbubbles as a delivery agent of 7beta-hydroxycholesterol in a radiofrequency lesion in the rat brain.

OBJECTIVE: This laboratory has previously described the aggregation of intravenously administered lipid-coated microbubbles (LCM) around tumors and areas of injury. 7Beta-hydroxycholesterol has been used to inhibit astrocytic proliferation in nervous system injury models. The compound has been given by direct infusion, by epidural catheter, or in liposomes (delivered stereotactically to the injury site). In this article, we report the use of LCM to deliver 7beta-hydroxycholesterol to a radiofrequency injury site in the rat cerebrum. METHODS: First, the ability of LCM to target the thermal lesion in the rat brain was characterized using a lipid-soluble fluorescent dye 3,3-dioctadecyloxacarbocyanine perchlorate. Then, the effectiveness of this delivery system in suppression of glial proliferation was measured by glial fibrillary acidic protein immunoreactivity. RESULTS: Glial fibrillary acidic protein immunoreactivity was significantly reduced when 7beta-hydroxycholesterol was administered via LCM but not alone, suggesting that astrocytic proliferation would correspondingly be diminished. CONCLUSION: LCM were assessed as a delivery vehicle for 7beta-hydroxycholesterol in a rat brain radiofrequency lesion and found to be efficient in reducing astrogliosis, as measured by glial fibrillary acidic protein immunoreactivity.

Evaluation of lipid-coated microbubbles as a delivery vehicle for Taxol in brain tumor therapy.

OBJECTIVE: This work evaluates the potential of lipid-coated microbubbles (LCM) as a delivery vehicle for lipid-soluble antineoplastic agents. We have shown, in rats, the selective affinity of intravenously administered LCM for tumor cells. They are internalized by the tumor cells both in vitro and in vivo. The specificity of LCM for tumors and subsequent incorporation into the cytoplasm could significantly reduce the systemic effects of an agent incorporated into the bubbles, such as Taxol. METHODS: The in vitro methods were as follows. C6 cells (10(5) cells) were treated with Taxol-LCM (6 micrograms/ml), Taxol-Cremophore (6 micrograms/ml), or LCM alone for 8 or 24 hours. Cell death was determined by staining the cells with nuclear staining. Abnormalities of microtubule structures were ascertained by confocal microscopy. The in vivo methods were as follows. Two rat tumor models (C6 and 9L) were used. Rats were treated with single bolus injections or with repetitive (two or three) treatment courses, with respective control animals. Each course consisted of one daily tail vein injection for 5 consecutive days and then 2 days of rest. RESULTS: When compared with either a saline control group or a group receiving Taxol in an oil vehicle, Taxol-LCM reduced tumor progression in Fischer 344 rats inoculated with 9L glioma. The most profound effect was observed with rats treated with three treatment cycles (five daily injections/cycle) separated by two rest periods (2 d/period). CONCLUSION: Both in vitro and in vivo data indicate that Taxol can be incorporated into LCM, can be delivered to the tumor site, and can exert a measurable antitumor biological effect.

The affinity of lipid-coated microbubbles for maturing brain injury sites.

The availability of a vehicle to deliver lipid soluble agents to a brain injury site may be of potential value in management of brain injury. This work describes the aggregation of intravenously administered Lipid-Coated Microbubbles (LCM) in the injury site following an experimental radiofrequency rat brain lesion. The bubbles can be identified around the region of the injury after the lesion has matured at least 48 h. The greatest bubbles density is evident after the lesion has matured for 10 days. This bubble density, reflecting "affinity," decreases to a plateau level from the second to the third week after injury. In order to investigate the potential relationship of bubble influx to posttraumatic astrocytosis and to cell turnover in the region, we utilized dual-channel laser-scanning confocal microscopy to track both bubble influx into the region and concomitant Glial Fibrillary Acidic Protein (GFAP) expressing astroctyte cell distribution. Cell turnover was assayed in separate sections using immunohistochemical staining of Proliferating Cell Nuclear Antigen (PCNA). We suggest a relationship between the LCM affinity and reactive astrocytes, but found no affinity of LCM for cells which stained positive with PCNA.

Bleomycin-induced pulmonary fibrosis in transgenic mice that either lack or overexpress the murine plasminogen activator inhibitor-1 gene.

Impaired fibrinolytic activity within the lung is a common manifestation of acute and chronic inflammatory lung diseases. Because the fibrinolytic system is active during repair processes that restore injured tissues to normal, reduced fibrinolytic activity may contribute to the subsequent development of pulmonary fibrosis. To examine the relationship between the fibrinolytic system and pulmonary fibrosis, lung inflammation was induced by bleomycin in transgenic mice that either overexpressed or were completely deficient in murine plasminogen activator inhibitor-1 (PAI-1). 2 wk after 0.075 U of bleomycin, the lungs of transgenic mice overexpressing PAI-1 contained significantly more hydroxyproline (118 +/- 8 micrograms) than littermate controls (70.5 +/- 8 micrograms, P < 0.005). 3 wk after administration of a higher dose of bleomycin (0.15 U), the lung hydroxyproline content of mice completely deficient in PAI-1 (49 +/- 8 micrograms) was not significantly different (P = 0.63) than that of control animals receiving saline (37 +/- 1 micrograms), while hydroxyproline content was significantly increased in heterozygote (77 +/- 12 micrograms, P = 0.06) and wild-type (124 +/- 19 micrograms, P < 0.001) littermates. These data demonstrate a direct correlation between the genetically determined level of PAI-1 expression and the extent of collagen accumulation that follows inflammatory lung injury. These results strongly support the hypothesis that alterations in fibrinolytic activity influence the extent of pulmonary fibrosis that occurs after inflammatory injury.

Pulmonary inflammation induced by incomplete or inactivated adenoviral particles.

One of the major obstacles to pulmonary-directed gene therapy using adenoviral vectors is the induction of inflammation. We investigated whether the adenoviral particles that constitute the initial inoculum can serve as an inflammatory stimulus, independent of their ability to express genes that they contain. Viral particles were prepared that are defective in gene expression by (i) isolating particles that have incomplete genomes by selecting those that have buoyant densities on CsCl density gradients lighter than complete viruses; and (ii) cross-linking viral DNA by exposure to ultraviolet light in the presence of 8-methoxypsoralen. The defective particles retained their icosahedral appearance when viewed by electron microscopy but lost their plaque-forming ability on 293 cells. High doses of intact, incomplete, or inactivated viral particles were instilled intratracheally into CBA/J mice, and after 6 days the amount of inflammation was quantified by counting inflammatory cells contained within lung tissue. We found that the inflammatory responses induced by the incomplete or inactivated viral vectors were quantitatively similar to those caused by intact, competent viral vectors. We conclude that high doses of adenoviral vectors that are used for gene therapy can induce pulmonary inflammation, independent of expressing the genes they contain.

Internalization of microbubbles by tumor cells in vivo and in vitro.

Lipid-coated microbubbles (LCM) administered intravenously (i.v.) to rats bearing brain tumor, specifically enhance tumor visualization by ultrasound [1]. In order to understand the basis for this observation, we have examined the interactions of LCM with glioblastoma (C6) and gliosarcoma (9L) tumor cells in vivo and in vitro. LCM and LCM labeled with the fluorescent lipophilic dye 3,3'-dioctadecyloxacarbocyanine perchlorate (diO) were administered to rats bearing brain tumor. LCM and diO-labeled LCM were found principally at the tumor site with no evidence of label in the surrounding normal brain tissue. Analysis of the tumor by confocal laser scanning microscopy revealed that labeled LCM were inside the tumor cells. Similar analysis of LCM interactions with C6 and 9L cells in culture showed that LCM first adsorb at the surface of the cells, and with time became localized inside the cells. Binding and internalization proceeded faster at 37 degrees C than at room temperature (RT). Staining of live cells with N-(3-((2,4-dinitrophenyl)amino)propyl)-N-(3-aminopropyl) methylamine dihydrochloride (DAMP), a dye that recognizes acidic compartments, showed that the majority of internalized LCM was associated with compartments containing DAMP. If the same uptake mechanism were operative in vivo, it would indicate that a portion of LCM bypasses the reticuloendothelial system and become endocytosed directly by tumor cells.

Expression of human interleukin-1 receptor antagonist in mouse lungs using a recombinant adenovirus: effects on vector-induced inflammation.

Pulmonary inflammation is a major obstacle to using adenovirus-based vectors for gene transfer to the lung. Since the pro-inflammatory cytokine, interleukin-1 (IL-1), is expressed early following adenovirus infection, we hypothesized that inhibition of IL-1 might block the inflammation caused by adenoviral vectors. To inhibit IL-1 activity at the site of infection continuously, we employed a recombinant adenovirus that contained the cDNA for human IL-1 receptor antagonist protein (IL-1ra) designated as Ad.RSVIL-1ra. When Ad.RSVIL-1ra was instilled intratracheally into CBA/J mice, human IL-1ra was recovered in lung tissue and bronchoalveolar lavage fluid for up to 30 days. Human IL-1ra is known to bind to murine IL-1 receptors and inhibit IL-1-mediated responses. To measure pulmonary inflammation, the number of inflammatory cells contained within suspensions of protease-digested lung tissue were counted 6 days after virus administration. Ad.RSVIL-1ra failed to reduce the number of inflammatory cells below that induced by a control vector that lacked an expression cassette (Ad.BgIII). Light microscopy showed that the lung tissue from Ad.RSVIL-1ra and Ad.BgIII-treated mice contained qualitatively similar amounts of inflammatory infiltrate. We conclude that adenovirus-based vectors can be used to induce high levels of IL-1ra expression within the lung, but such expression was unable to prevent adenoviral vector-induced inflammation.