Cross-Domain Unpaired Learning for Low-Dose CT Imaging.

Supervised deep-learning techniques with paired training datasets have been widely studied for low-dose computed tomography (LDCT) imaging with excellent performance. However, the paired training datasets are usually difficult to obtain in clinical routine, which restricts the wide adoption of supervised deep-learning techniques in clinical practices. To address this issue, a general idea is to construct a pseudo paired training dataset based on the widely available unpaired data, after which, supervised deep-learning techniques can be adopted for improving the LDCT imaging performance by training on the pseudo paired training dataset. However, due to the complexity of noise properties in CT imaging, the LDCT data are difficult to generate in order to construct the pseudo paired training dataset. In this article, we propose a simple yet effective cross-domain unpaired learning framework for pseudo LDCT data generation and LDCT image reconstruction, which is denoted as CrossDuL. Specifically, a dedicated pseudo LDCT sinogram generative module is constructed based on a data-dependent noise model in the sinogram domain, and then instead of in the sinogram domain, a pseudo paired dataset is constructed in the image domain to train an LDCT image restoration module. To validate the effectiveness of the proposed framework, clinical datasets are adopted. Experimental results demonstrate that the CrossDuL framework can obtain promising LDCT imaging performance in both quantitative and qualitative measurements.

Enhancing genomic mutation data storage optimization based on the compression of asymmetry of sparsity.

Background: With the rapid development of high-throughput sequencing technology and the explosive growth of genomic data, storing, transmitting and processing massive amounts of data has become a new challenge. How to achieve fast lossless compression and decompression according to the characteristics of the data to speed up data transmission and processing requires research on relevant compression algorithms. Methods: In this paper, a compression algorithm for sparse asymmetric gene mutations (CA_SAGM) based on the characteristics of sparse genomic mutation data was proposed. The data was first sorted on a row-first basis so that neighboring non-zero elements were as close as possible to each other. The data were then renumbered using the reverse Cuthill-Mckee sorting technique. Finally the data were compressed into sparse row format (CSR) and stored. We had analyzed and compared the results of the CA_SAGM, coordinate format (COO) and compressed sparse column format (CSC) algorithms for sparse asymmetric genomic data. Nine types of single-nucleotide variation (SNV) data and six types of copy number variation (CNV) data from the TCGA database were used as the subjects of this study. Compression and decompression time, compression and decompression rate, compression memory and compression ratio were used as evaluation metrics. The correlation between each metric and the basic characteristics of the original data was further investigated. Results: The experimental results showed that the COO method had the shortest compression time, the fastest compression rate and the largest compression ratio, and had the best compression performance. CSC compression performance was the worst, and CA_SAGM compression performance was between the two. When decompressing the data, CA_SAGM performed the best, with the shortest decompression time and the fastest decompression rate. COO decompression performance was the worst. With increasing sparsity, the COO, CSC and CA_SAGM algorithms all exhibited longer compression and decompression times, lower compression and decompression rates, larger compression memory and lower compression ratios. When the sparsity was large, the compression memory and compression ratio of the three algorithms showed no difference characteristics, but the rest of the indexes were still different. Conclusion: CA_SAGM was an efficient compression algorithm that combines compression and decompression performance for sparse genomic mutation data.

Estimating the amplitude spectrum area of ventricular fibrillation during cardiopulmonary resuscitation using only ECG waveform.

BACKGROUND: Amplitude spectrum area (AMSA) calculated from ventricular fibrillation (VF) can be used to monitor the effectiveness of chest compression (CC) and optimize the timing of defibrillation. However, reliable AMSA can only be obtained during CC pause because of artifacts. In this study, we sought to develop a method for estimating AMSA during cardiopulmonary resuscitation (CPR) using only the electrocardiogram (ECG) waveform. METHODS: Intervals of 8 seconds ECG and CC-related references, including 4 seconds during CC and an adjacent 4 seconds without CC, were collected before 1,008 defibrillation shocks from 512 out-of-hospital cardiac arrest patients. Signal quality was analyzed based on the irregularity of autocorrelation of VF. If signal quality index (SQI) was high, AMSA would be calculated from the original signal. Otherwise, CC-related artifacts would be constructed and suppressed using the least mean square filter from VF before calculation of AMSA. The algorithm was optimized using 480 training shocks and evaluated using 528 independent testing shocks. RESULTS: Overall, CC resulted in lower SQI [0.15 (0.04-0.61) with CC vs. 0.75 (0.61-0.83) without CC, P<0.01] and higher AMSA [11.2 (7.7-16.2) with CC vs. 7.2 (4.9-10.6) mVHz without CC, P<0.01] values. The predictive accuracy (49.2% vs. 66.5%, P<0.01) and area under the receiver operating characteristic curve (AUC) (0.647 vs. 0.734, P<0.01) were significantly decreased during CC. Using the proposed method, the estimated AMSA was 7.1 (5.0-15.2) mVHz, the predictive accuracy was 67.0% and the AUC was 0.713, which were all comparable with those calculated without CC. CONCLUSIONS: Using the signal quality-based artifact suppression method, AMSA can be reliably estimated and continuously monitored during CPR.

[Establishment of porcine model of prolonged cardiac arrest and cardiopulmonary resuscitation electrically induced by ventricular fibrillation].

OBJECTIVE: To investigate the optimal injury time point of cardiac arrest (CA) induced electrically, and establish a reproducible prolonged CA and cardiopulmonary resuscitation (CPR) model in pigs. METHODS: Forty healthy domestic male pigs were randomly divided into four groups, which were ventricular fibrillation (VF) 8, 10, 11, and 12 minutes groups, each group for 10 animals. In these groups, VF was induced by alternating current delivered to right ventricular endocardium and untreated for 8, 10, 11, and 12 minutes, respectively, followed by 6 minutes of CPR procedure. The resuscitation and survival outcomes were recorded. Hemodynamic parameters and arterial blood gases of animals after successful resuscitation were measured and recorded for 6 hours. Those successful resuscitation animals were regularly evaluated for the neurological deficit score (NDS) and survival outcomes every 24 hours till 96 hours after resuscitation. RESULTS: The shortest duration of CPR (minute: 6.9±1.3) and the highest successful ratio of the first defibrillation (7/10) were observed in group VF 8 minutes, and the ratio of successful resuscitation was 100%. The best coronary perfusion pressure (CPP) during the CPR, less neurological impairment, longer survival time, more stable hemodynamics, and shorter time for arterial pH and lactate level restoring to the original state after CPR were also observed in group VF 8 minutes, and no severe damage was found in those animals. The longest duration of CPR (minute: 10.3±2.9) and the lowest successful ratio of the first defibrillation (1/10) were observed in group VF 12 minutes, and only 4 animals achieved restoration of spontaneous circulation (ROSC), and no animal survived to CPR 96 hours. The worst CPP during CPR and the highest NDS after resuscitation were also found in VF 12 minutes animals compared to those animals in the other groups. The injuries caused by ischemia and hypoxia in groups VF 10 minutes and VF 11 minutes were in between those of the groups VF 8 minutes and VF 12 minutes, and the duration of CPR were (7.0±2.1) minutes and (8.2±2.6) minutes. There were 9 and 7 animals achieved ROSC in groups VF 10 minutes and VF 11 minutes correspondingly, and 6 and 4 animals survived to 96 hours respectively. Obviously unstable hemodynamics was observed during the period of CPR 2 hours in the two groups. At CPR 1 hour, the heart rates (HR, beats/min) in groups VF 10 minutes and VF 11 minutes increased to 172 (155, 201) and 168 (136, 196) respectively, and the mean arterial pressures (MAP, mmHg, 1 mmHg = 0.133 kPa) declined to 97 (92, 100) and 81 (77, 100), the cardiac output (CO, L/min) decreased to 5.0 (4.0, 5.8), 3.7 (3.0, 5.4) correspondingly. Distinct injuries were found in the two groups [CPR 24-96 hours NDS in groups VF 10 minutes and VF 11 minutes: 180 (110, 255)-20 (0, 400) and 275 (223, 350)-240 (110, 400)], and the arterial pH of the two group decreased to 7.26±0.09 and 7.23±0.09 respectively, and the level of lactate (mmol/L) increased to 9.17±1.48 and 12.80±2.71 correspondingly at CPR 0.5 hour. Significantly lower pH was observed in group VF 11 minutes compared to group VF 8 minutes at CPR 0.5 hour (7.23±0.09 vs. 7.33±0.04, P < 0.05). The highest level of lactate (mmol/L) was also found at the same time point in group VF 11 minutes, which recovered to normal slowly, and was still significantly higher than groups VF 8, 10, 12 minutes (7.58±3.99 vs. 2.55±1.53, 2.13±2.00, 3.40±2.30, all P < 0.05) at CPR 4 hours. CONCLUSIONS: The longer duration of CA was, the more severe damage would be, the longer CPR time would be required, and the harder of the animals to achieve ROSC. In this prolonged CA and CPR porcine model, 10-11 minutes for untreated VF, was an optimal time point with appropriate successful rate of resuscitation, survival outcomes, and post-resuscitation injuries. Therefore, we recommended 10-11 minutes might be the rational length of no-flow time in this model.

Short Duration Combined Mild Hypothermia Improves Resuscitation Outcomes in a Porcine Model of Prolonged Cardiac Arrest.

OBJECTIVE: In this study, our aim was to investigate the effects of combined hypothermia with short duration maintenance on the resuscitation outcomes in a porcine model of ventricular fibrillation (VF). METHODS: Fourteen porcine models were electrically induced with VF and untreated for 11 mins. All animals were successfully resuscitated manually and then randomized into two groups: combined mild hypothermia (CH group) and normothermia group (NT group). A combined hypothermia of ice cold saline infusion and surface cooling was implemented in the animals of the CH group and maintained for 4 hours. The survival outcomes and neurological function were evaluated every 24 hours until a maximum of 96 hours. Neuron apoptosis in hippocampus was analyzed. RESULTS: There were no significant differences in baseline physiologies and primary resuscitation outcomes between both groups. Obvious improvements of cardiac output were observed in the CH group at 120, 180, and 240 mins following resuscitation. The animals demonstrated better survival at 96 hours in the CH group when compared to the NT group. In comparison with the NT group, favorable neurological functions were observed in the CH group. CONCLUSION: Short duration combined cooling initiated after resuscitation improves survival and neurological outcomes in a porcine model of prolonged VF.

A calibrated method for blood pressure measurement based on volume pulse wave / 中国医疗器械杂志

Physiology parameters measurement based on volume pulse wave is suitable for the monitoring blood pressure continuously. This paper described that the systolic blood pressure (SBP) and diastolic blood pressure (DBP) can be calibrated by measuring the pulse propagation time, just on one point of finger tip. The volume pulse wave was acquired by lighting the red and infrared LED alternately, and after signal processing, an accelerated pulse wave was obtained. Then by measuring the pulse wave propagation time between the progressive wave and reflected wave, we can find the relationship of the time and the blood pressure, and establish the related systolic blood pressure measurement equation. At the same time, based on the relationship between alternating current and direct current components in the volume pulse waveforms and through regression analysising, the relevant diastolic blood pressure measurement equation can be established. 33 clinical experimentation cases have been worked by dividing them into two groups: training group (18 cases) and control group (15 cases), by comparing with the measuring results of the OMRON electronic sphygmomanometer. The results indicated that the two methods had good coherence. The measurement described is simple and reliable, and may be served as a new method for noninvasively and continuously measurement of blood pressure.