Information Theory and Reliable Communication

Sickle cell disease (SCD) is a group of inherited disorders caused by mutations in HBB, which encodes haemoglobin subunit β. The incidence is estimated to be between 300,000 and 400,000 neonates globally each year, the majority in sub-Saharan Africa. Haemoglobin molecules that include mutant sickle β-globin subunits can polymerize; erythrocytes that contain mostly haemoglobin polymers assume a sickled form and are prone to haemolysis. Other pathophysiological mechanisms that contribute to the SCD phenotype are vaso-occlusion and activation of the immune system. SCD is characterized by a remarkable phenotypic complexity. Common acute complications are acute pain events, acute chest syndrome and stroke; chronic complications (including chronic kidney disease) can damage all organs. Hydroxycarbamide, blood transfusions and haematopoietic stem cell transplantation can reduce the severity of the disease. Early diagnosis is crucial to improve survival, and universal newborn screening programmes have been implemented in some countries but are challenging in low-income, high-burden settings.

Sickle cell disease

In this paper, a novel computationally intelligent-based electrocardiogram (ECG) signal classification methodology using a deep learning (DL) machine is developed. The focus is on patient screening and identifying patients with paroxysmal atrial fibrillation (PAF), which represents a life threatening cardiac arrhythmia. The proposed approach operates with a large volume of raw ECG time-series data as inputs to a deep convolutional neural networks (CNN). It autonomously learns representative and key features of the PAF to be used by a classification module. The features are therefore learned directly from the large time domain ECG signals by using a CNN with one fully connected layer. The learned features can effectively replace the traditional ad hoc and time-consuming user’s hand-crafted features. Our experimental results verify and validate the effectiveness and capabilities of the learned features for PAF patient screening. The main advantages of our proposed approach are to simplify the feature extraction process corresponding to different cardiac arrhythmias and to remove the need for using a human expert to define appropriate and critical features working with a large time-series data set. The extensive simulations and case studies conducted indicate that combining the learned features with other classifiers will significantly improve the performance of the patient screening system as compared to an end-to-end CNN classifier. The effectiveness and capabilities of our proposed ECG DL classification machine is demonstrated and quantitative comparisons with several conventional machine learning classifiers are also provided.

Deep Convolutional Neural Networks and Learning ECG Features for Screening Paroxysmal Atrial Fibrillation Patients

PhysioBank是一个大型的逐渐扩增的生理学信号和相关数据的数字化记录文档。目前包含多参数的心肺、神经和其他生物医学信号，尤以心电图（ECG）为主。信号来自健康受试者和各种疾病的患者。涉及的疾病包括心脏猝死、充血性心力衰竭、癫痫、步态不稳、睡眠呼吸暂停和衰老等。

PhysioBank

Goal: To develop a robust and real-time approach for automatic detection of atrial fibrillation (AF) in long-term electrocardiogram (ECG) recordings using deep learning (DL). Method: An end-to-end model combining the Convolutional- and Recurrent-Neural Networks (CNN and RNN) was proposed to extract high level features from segments of RR intervals (RRIs) in order to classify them as AF or normal sinus rhythm (NSR). Results: The model was trained and validated on three different databases including a total of 89 subjects. It achieved a sensitivity and specificity of 98.98% and 96.95%, respectively, validated through a 5-fold cross-validation. Additionally, the proposed model was found to be computationally efficient and it was capable of analyzing 24 h of ECG recordings in less than one second. The proposed algorithm was also tested on the unseen datasets to examine its robustness in detecting AF for new recordings which resulted in 98.96% and 86.04% for specificity and sensitivity, respectively. Conclusion: Compared to the state-of-the-art models evaluated on standard benchmark ECG datasets, the proposed model produced better performance in detecting AF. Additionally, since the model learns features directly from the data, it avoids the need for clever/cumbersome feature engineering.

https://backend.orbit.dtu.dk/ws/files/152226346/1_s2.0_S0957417418305190_main.pdf

A deep learning approach for real-time detection of atrial fibrillation

Automatic detection of atrial fibrillation using stationary wavelet transform and support vector machine

BACKGROUND: Studies from older cohorts of kidney recipients have observed that recipients with sickle cell disease (SCD) have lower patient survival compared with age- and race-matched controls. We examined whether survival has improved among SCD recipients in the current era. METHODS: Using Organ Procurement and Transplantation Network/United Network for Organ Sharing data, all black/African-American kidney recipients were stratified according to transplant year into an early (1988-99) and recent era (2000-11). Patient and allograft survival among SCD recipients and those with other diagnoses were compared (early era: SCD n = 67, others n = 20 694; recent era: SCD n = 106, others n = 34 428). A secondary-matched cohort analysis compared patient and allograft survival between SCD recipients matched to recipients with other diagnoses based on recipient and donor age, gender and donor type (deceased versus living). RESULTS: Patient survival at 6 years was lower among SCD recipients in the early era compared with other diagnoses (55.7 versus 78.0%; P < 0.001). Six-year patient survival among sickle cell recipients improved in the recent era (69.8%; P versus early era = 0.04), although still trended toward lower survival compared with other diagnoses (80.0%; P = 0.07). Multivariate Cox proportional hazard models revealed an increased mortality risk with SCD in both eras [early: hazard ratio (HR) = 3.12; 95% confidence interval (CI): 2.15-4.54; recent: HR: 2.03; 95% CI: 1.31-3.16]. Patient survival among matched SCD recipients in the recent era was comparable to diabetic recipients (SCD: 73.1%, diabetes: 74.1%; P = 0.44). CONCLUSIONS: Patient survival has improved among contemporary sickle cell recipients compared with an earlier cohort and is comparable to a matched cohort of diabetic kidney recipients. Appropriately selected SCD patients may receive kidney transplants with reasonable survival outcome.

/pdf/improved-survival-among-sickle-cell-kidney-transplant-2hho994zqt.pdf

Improved survival among sickle cell kidney transplant recipients in the recent era

Author(s): Huang, Edmund | Advisor(s): Elashoff, Robert M; Elashoff, David | Abstract: Background: Candidates may be active or temporarily inactive (status 7) on the kidney transplant waiting list. One reason candidates may be inactive is for a “weight currently inappropriate for transplantation”. We hypothesized that many of these candidates would not achieve active status. Methods: Using OPTN/UNOS data from 2006-2012, we used competing risks methods to determine the cumulative incidence of conversion to active status (activation), death, and delisting before conversion among 1679 obese adult kidney candidates designated as status 7 due to a weight inappropriate for transplantation. Fine and Gray competing risks regression was performed to characterize factors associated with conversion to active status in the overall study population and of eventual transplantation among a subgroup of activated candidates. Results: At six years, the cumulative incidence of activation was 49%, of death before conversion was 15%, and of delisting was 21%. Higher body mass index (BMI) was strongly associated with a decreased subhazard of activation (BMI ≥45 vs. 30-34.9, sHR: 0.22; 95% CI: 0.16-0.33). Female gender, diabetic end-stage renal disease, history of a previous transplant, panel reactive antibodies l80%, dialysis-dependence at listing, and UNOS region 5 were negatively associated with activation. Among activated candidates, the cumulative incidence of transplantation at six years after initial waitlisting was 61%. Conclusions: Our findings indicate that half of obese status 7 candidates with a weight inappropriate for transplantation will not achieve active waitlist status. BMI at listing had a strong association with conversion to active status; however, co-morbid factors and regional variation also impact activation.

/pdf/incidence-of-conversion-to-active-waitlist-status-among-4ny2b8ioj1.pdf

Incidence of conversion to active waitlist status among temporarily inactive obese renal transplant candidates.

In mobile ad hoc network (MANET) studies, it is imperative to use highly detailed device models as they provide high layer protocols with good prediction of underlying wireless communication performance. However, such studies often utilize abstract models for execution speed and simplicity. This paper first shows that physical layer variables including path loss, shadowing, multipath, Doppler have significant effects on the predicted overall networking performance. It then proposes an approach to simulate details of wireless propagation and radio characteristics in networking studies while still maintaining a reasonable simulation execution time. Through our runtime performance studies with detailed OFDM Simulink/MATLAB models and QualNet network simulator, it is shown that the proposed approach can improve the simulation runtime performance by three to four orders of magnitudes without compromising the fidelity of simulation results.

Detailed OFDM modeling in network simulation of mobile ad hoc networks

New insights suggest that the most efficient FIR digital filters can be created by using a scaled sequence of stages, each representing a factor of the filter's transfer function. A crucial capability for building such filters concerns finding the best FIR filter factors, then carefully scaling and sequencing them. The efficiency of the resulting structure depends heavily upon obtaining such optimal factors. We offer an algorithm to find, scale and sequence optimally factored FIR filters.

Alireza Mehrnia

Papers

Automatic detection of atrial fibrillation using stationary wavelet transform and support vector machine

Improved survival among sickle cell kidney transplant recipients in the recent era

Incidence of conversion to active waitlist status among temporarily inactive obese renal transplant candidates.

Detailed OFDM modeling in network simulation of mobile ad hoc networks

Optimal Factoring of FIR Filters