Isolated Cleft Mitral Valve with Posterior and Anterior Clefts: a Rare Cause of Congenital Valve Regurgitation

We report a case of isolated cleft mitral valve with two clefts in the posterior and one in the anterior leaflet. Our case adds to the few reports of posterior and multiple mitral valve clefts and to our knowledge is the first using real‐time transoesophageal three‐dimensional echocardiography (3DE) for assessment of isolated cleft mitral valve. (Echocardiography 2010;27:E50‐E52)     

From the Cover: Convolutional networks for fast,energy-efficient neuromorphic computing

Deep networks are now able to achieve human-level performance on a broad spectrum of recognition tasks. Independently, neuromorphic computing has now demonstrated unprecedented energy-efficiency through a new chip architecture based on spiking neurons, low precision synapses, and a scalable communication network. Here, we demonstrate that neuromorphic computing, despite its novel architectural primitives, can implement deep convolution networks that (i) approach state-of-the-art classification accuracy across eight standard datasets encompassing vision and speech, (ii) perform inference while preserving the hardware’s underlying energy-efficiency and high throughput, running on the aforementioned datasets at between 1,200 and 2,600 frames/s and using between 25 and 275 mW (effectively >6,000 frames/s per Watt), and (iii) can be specified and trained using backpropagation with the same ease-of-use as contemporary deep learning. This approach allows the algorithmic power of deep learning to be merged with the efficiency of neuromorphic processors, bringing the promise of embedded, intelligent, brain-inspired computing one step closer.The human brain is capable of remarkable acts of perception while consuming very little energy. The dream of brain-inspired computing is to build machines that do the same, requiring high-accuracy algorithms and efficient hardware to run those algorithms. On the algorithm front, building on classic work on backpropagation (1), the neocognitron (2), and convolutional networks (3), deep learning has made great strides in achieving human-level performance on a wide range of recognition tasks (4). On the hardware front, building on foundational work on silicon neural systems (5), neuromorphic computing, using novel architectural primitives, has recently demonstrated hardware capable of running 1 million neurons and 256 million synapses for extremely low power (just 70 mW at real-time operation) (6). Bringing these approaches together holds the promise of a new generation of embedded, real-time systems, but first requires reconciling key differences in the structure and operation between contemporary algorithms and hardware. Here, we introduce and demonstrate an approach we call Eedn, energy-efficient deep neuromorphic networks, which creates convolutional networks whose connections, neurons, and weights have been adapted to run inference tasks on neuromorphic hardware.For structure, typical convolutional networks place no constraints on filter sizes, whereas neuromorphic systems can take advantage of blockwise connectivity that limits filter sizes, thereby saving energy because weights can now be stored in local on-chip memory within dedicated neural cores. Here, we present a convolutional network structure that naturally maps to the efficient connection primitives used in contemporary neuromorphic systems. We enforce this connectivity constraint by partitioning filters into multiple groups and yet maintain network integration by interspersing layers whose filter support region is able to cover incoming features from many groups by using a small topographic size (7).For operation, contemporary convolutional networks typically use high precision ( ≥ 32-bit) neurons and synapses to provide continuous derivatives and support small incremental changes to network state, both formally required for backpropagation-based gradient learning. In comparison, neuromorphic designs can use one-bit spikes to provide event-based computation and communication (consuming energy only when necessary) and can use low-precision synapses to colocate memory with computation (keeping data movement local and avoiding off-chip memory bottlenecks). Here, we demonstrate that by introducing two constraints into the learning rule—binary-valued neurons with approximate derivatives and trinary-valued ({−1,0,1}) synapses—it is possible to adapt backpropagation to create networks directly implementable using energy efficient neuromorphic dynamics. This approach draws inspiration from the spiking neurons and low-precision synapses of the brain (8) and builds on work showing that deep learning can create networks with constrained connectivity (9), low-precision synapses (10, 11), low-precision neurons (12–14), or both low-precision synapses and neurons (15, 16). For input data, we use a first layer to transform multivalued, multichannel input into binary channels using convolution filters that are learned via backpropagation (12, 16) and whose output can be sent on chip in the form of spikes. These binary channels, intuitively akin to independent components (17) learned with supervision, provide a parallel distributed representation to carry out high-fidelity computation without the need for high-precision representation.Critically, we demonstrate that bringing the above innovations together allows us to create networks that approach state-of-the-art accuracy performing inference on eight standard datasets, running on a neuromorphic chip at between 1,200 and 2,600 frames/s (FPS), using between 25 and 275 mW. We further explore how our approach scales by simulating multichip configurations. Ease-of-use is achieved using training tools built from existing, optimized deep learning frameworks (18), with learned parameters mapped to hardware using a high-level deployment language (19). Although we choose the IBM TrueNorth chip (6) for our example deployment platform, the essence of our constructions can apply to other emerging neuromorphic approaches (20–23) and may lead to new architectures that incorporate deep learning and efficient hardware primitives from the ground up.     

Predictors of venous obstruction following pacemaker or implantable cardioverter-defibrillator implantation: a contrast venographic study on 100 patients admitted for generator change, lead revision, or device upgrade.

AIM: Venous obstruction following transvenous device implantation rarely cause immediate clinical problems. When lead revision or device upgrade is indicated, venous obstruction become a significant challenge. The aim of this study was to determine the predictors of venous obstruction after transvenous device implantation, and to asess likely effects of antiplatelet/anticoagulant drugs in preventing venous thrombosis. METHODS AND RESULTS: Between March 2005 and July 2006, contrast venography was performed in 100 patients who were candidates for generator change, lead revision, or device upgrade. Vessel patency was graded as either completely obstructed, partially obstructed (>70%), or patent. The incidence of venous obstruction was 26%, with 9% of patients having total obstruction and 17% of patients exhibiting partial obstruction. No statistically significant differences between obstructed and non-obstructed patients were seen for age, sex, indication for device implantation, atrial fibrillation, cardiothoracic ratio, insulation material, operative technique, device type, and manufacturer (all Ps > 0.05). In a univariate analysis, multiple leads (P = 0.033), and presence of dilated cardiomyopathy (P = 0.036) were associated with higher risk of venous obstruction, whereas anticoagulant/antiplatelet therapy (P = 0.047) significantly reduced incidence of venous obstruction. Multivariate logistic regression analysis showed that only number of the leads (P = 0.039, OR: 2.22, and 95% CI: 1.03-4.76) and antiplatelet/anticoagulant therapy (P = 0.044, OR: 2.79, and 95% CI: 0.98-7.96) were predictors of venous obstruction. CONCLUSION: Total or partial obstruction of the access veins occurs relatively frequently after pacemaker or ICD implantation. Multiple pacing or ICD leads are associated with an increased risk of venous obstruction, whereas antiplatelet/anticoagulant therapy appears to have a preventive effect on development of access vein thrombosis.     

Electrocardiographic and electrophysiologic characteristics of anteroseptal, midseptal, and posteroseptal accessory pathways.

BACKGROUND: Approximately 30% of all accessory pathways (APs) are located in the septal area, and understanding the electrocardiographic and electrophysiologic of these APs is crucial for safe and effective ablation of these pathways. OBJECTIVE: In this study, the electrocardiographic and electrophysiologic characteristics of anteroseptal, midseptal, and posteroseptal APs were investigated in detail to elucidate unique electrical properties of APs in each location. METHODS: From April 2002 to October 2006, a total of 120 patients with a septal AP-mediated tachycardia were enrolled in the study. A detailed examination including electrocardiographic analysis and electrophysiologic study was performed in all patients. RESULTS: A total of 120 patients, including 98 patients with posteroseptal APs, 14 patients with anteroseptal APs, and 8 patients with midseptal APs, were studied. The anteroseptal APs could be differentiated from the midseptal APs by the 2 or more positive delta waves in inferior leads, whereas there is significant overlap in electrocardiographic features of midseptal and posteroseptal APs. The mean tachycardia cycle length was significantly shorter in patients with midseptal AP compared with those with anteroseptal and posteroseptal APs (284 +/- 49 ms vs 342 +/- 46 ms vs 350 +/- 68 ms, P = .03). The AH interval during tachycardia was also shorter in patients with midseptal APs (149 +/- 16 ms vs 200 +/- 51 ms vs 168 +/- 48 ms, P = .04). The patients with posteroseptal AP had a significantly higher incidence of atrial fibrillation (35%) than those with either midseptal (12%) or anteroseptal (14%) APs (P = .04). The patients with posteroseptal APs also had a significantly shorter antegrade effective refractory period of the AP (276 +/- 54 ms) than those with either midseptal (313 +/- 71 ms) or anteroseptal (325 +/- 61) APs (P = .036). CONCLUSION: Electrocardiographic analysis is a reliable method for differentiation of the anteroseptal from the midseptal APs, whereas the same is not true for the midseptal and posteroseptal APs. Midseptal APs were characterized by faster orthodromic tachycardia, whereas posteroseptal APs had a higher inducibility of atrial fibrillation.     

Spectral Analysis of the High Resolution QRS Complex During Exercise-Induced Ischemia

Background: The effect of acute myocardial infarction and regional ischemia on the frequency content of the ECG signal has been described by several investigators. In the present study, the feasibility of assessing changes in the QRS spectrum during exercise testing, and whether these changes are related to the occurrence of ischemia were examined. Methods: Spectral analysis of the high resolution ECGs from leads V₃, V₄, V₅, and V₆ were performed in two groups of male subjects before, during, and following treadmill exercise testing. Group A included 32 coronary artery disease (CAD) patients, with arteriographically proven >75% obstruction of at least two main coronary arteries, and group B included 30 healthy subjects, without history or symptoms of CAD. Signal averaging and filtering techniques were used in order to enhance the signal-to-noise ratio of the recorded ECGs. The power spectrum of the averaged QRS waveform for the different stages of the exercise testing was computed using a Fast Fourier Transform, and the slope of the linear regression line was found in the frequency range 7.81–249.92 Hz on the plot of log((amplitude)²) versus log(frequency). Results: Regression line slopes immediately after peak exercise were significantly lower for the CAD group than for the healthy subjects in 3 of the 4 examined leads. No significant changes in slopes were found between the two groups at rest or during late recovery. Comparing the differences between slopes at different stages of the test revealed that the difference between postexercise slope and rest slope has lower mean values for the CAD group in all four leads, with a significant difference in lead V6, and for the difference between postexercise slope and recovery slope, lower mean values were found for the CAD group in all four leads, with a significant difference in V5 and V6. Conclusions: These findings indicate that ischemic changes affect the power spectrum of the QRS complex, and result in a steeper regression line on a log-log scale.     

Vapor condensation with daytime radiative cooling

A radiative vapor condenser sheds heat in the form of infrared radiation and cools itself to below the ambient air temperature to produce liquid water from vapor. This effect has been known for centuries, and is exploited by some insects to survive in dry deserts. Humans have also been using radiative condensation for dew collection. However, all existing radiative vapor condensers must operate during the nighttime. Here, we develop daytime radiative condensers that continue to operate 24 h a day. These daytime radiative condensers can produce water from vapor under direct sunlight, without active consumption of energy. Combined with traditional passive cooling via convection and conduction, radiative cooling can substantially increase the performance of passive vapor condensation, which can be used for passive water extraction and purification technologies.

Energy and clean water are global challenges that are intertwined in an unfavorable way: even in areas where water is abundant, energy may not be available to purify it for human use (1, 2). There has been strong interest in developing passive technologies to purify or harvest water without using fuel or electricity. In this context, passive vapor condensation becomes particularly important because many passive water technologies go through the vapor phase of water in their harvesting or purification processes.Traditional vapor condensation technique is based on convective and conductive heat exchange with ambient environments. This technique is widely used in systems with hot vapors (3–6). However, with ever-increasing emphasis on passive systems, there are many situations in which warm- or even room-temperature vapor needs to be effectively condensed, such as extracting water from atmosphere (7–9) and warm vapor generated from high-efficiency solar evaporation (10). For vapor at such temperatures, most traditional condensers fail. For this reason, there is a clear need for a condensation technique to complement traditional condensers.A different technique is based on radiative vapor condensation. Darkling beetles in the Namib desert (11) use this technique to collect water. Their bodies function as a cooling surface by shedding thermal energy through midinfrared (mid-IR) radiation toward a clear nighttime sky, generating dew from humid air. This mechanism is also used by commercial radiative dew condensers (7–9). However, neither Namib beetle nor existing dew condensers can operate in the daytime (7). Those nighttime radiative condensers are incompatible with many emerging water technologies that require 24 h operation or direct access to sunlight.Recently, Fan et al. showed that passive radiative cooling to subambient temperatures can be realized even during the daytime, by integrating a high-efficiency solar reflector with a high-emissivity thermal emitter in the mid-IR atmospheric transparency window (12). Using this work as a basis, here we demonstrate a daytime radiative condenser. Compared to existing radiative vapor condensers (7–9), our condenser can function even in the presence of sunlight, which is essential for integration into passive water-harvesting systems that mainly operate during daytime.     

Clinical evidence for utilization of the A3 adenosine receptor as a target to treat rheumatoid arthritis: data from a phase II clinical trial

OBJECTIVE: Adenosine exerts antiinflammatory effects via activation of the A3 adenosine receptor (A3AR), a Gi protein-associated cell-surface receptor, overexpressed in synovial tissue and peripheral blood mononuclear cells (PBMC) in patients with active rheumatoid arthritis (RA). CF101 is a highly specific orally bioavailable A3AR agonist. METHODS: This was a multicenter study, blinded to dose, designed to assess the clinical activity and safety of CF101 in active RA. Seventy-four patients were randomized to receive 0.1, 1.0, or 4.0 mg CF101 bid for 12 weeks. The primary efficacy endpoint was American College of Rheumatology 20% response (ACR20) at Week 12. A3AR expression levels were analyzed in PBMC from 18 patients. RESULTS:. Maximal responses were observed with 1.0 mg bid, lower at 0.1 and 4.0 mg bid. At 12 weeks, 55.6%, 33.3%, and 11.5% of the patients receiving 1.0 mg CF101 achieved ACR20%, 50%, and 70% responses, respectively. CF101 was generally well tolerated, with mild headache (4.1%), nausea (2.7%), and rash (2.7%) being the most common treatment-related adverse events. Statistically significant correlations between A3AR overexpression at baseline and ACR50 and ACR70 responses were observed. CONCLUSION: CF101 administered bid for 12 weeks resulted in improvement in signs and symptoms of RA that did not achieve statistical significance, and was safe and well tolerated. The expression level of A3AR was directly correlated with patient responses to CF101, suggesting its utilization as a biomarker for the pharmacodynamic and therapeutic effects of this novel agent. These findings require confirmation in a double-blind randomized placebo-controlled trial, currently under way.     

Lymphocytes proliferate in blood and lymph nodes following interleukin-2 therapy in addition to highly active antiretroviral therapy.

BACKGROUND: Substantial redistribution of lymphocytes occurs upon the initiation of highly active antiretroviral therapy (HAART) and immune-based HIV therapies. OBJECTIVE: To evaluate the relative contribution of apoptosis and proliferation to changes in lymphocyte populations in peripheral blood and lymph node resulting from interleukin-2 (IL-2) therapy in patients receiving stable HAART. METHODS: Lymphocyte apoptosis was analyzed on various subtypes using fluorescence activated cell sorting with an annexin-V antibody in peripheral blood and by the TUNEL (terminal uridine nucleotide end labelling) method in corresponding lymph node sections. Lymphocyte proliferation was evaluated using an antibody against the cell cycle-associated marker Ki-67 (MIB-1) in peripheral blood and lymph nodes. RESULTS: A transient increase in apoptosis was seen in peripheral blood and lymph nodes during a cycle of subcutaneous IL-2. A pronounced proliferative effect of IL-2 (from 6.4% of total lymphocytes in patients only treated with HAART to 23.4% in those treated with HAART + IL-2) was detected in peripheral blood, affecting the CD4, CD8 and CD16/56 subsets to a similar extent. Remarkably, the proliferative effect also occurred in lymphoid tissues. While the lymph node structure gradually disintegrated over 24 months in some individuals, the amount of proliferating lymphocytes, including CD4 cells, B cells and follicular dendritic cells, greatly increased upon IL-2, while HIV RNA load in lymph nodes remained unaffected. CONCLUSION: These results show that IL-2 leads to lymphocyte proliferation in peripheral blood and lymph nodes without an impact on viral load in lymphoid tissue. These results have important implications for attempts to reconstitute the immune system in HIV disease.