ABSTRACT
In blurry images, the degree of image blurs may vary drastically due to different factors, such as varying speeds of shaking cameras and moving objects, as well as defects of the camera lens. However, current end-to-end models failed to explicitly take into account such diversity of blurs. This unawareness compromises the specialization at each blur level, yielding sub-optimal deblurred images as well as redundant post-processing. Therefore, how to specialize one model simultaneously at different blur levels, while still ensuring coverage and generalization, becomes an emerging challenge. In this work, we propose Ada-Deblur, a super-network that can be applied to a "broad spectrum" of blur levels with no re-training on novel blurs. To balance between individual blur level specialization and wide-range blur levels coverage, the key idea is to dynamically adapt the network architectures from a single well-trained super-network structure, targeting flexible image processing with different deblurring capacities at test time. Extensive experiments demonstrate that our work outperforms strong baselines by demonstrating better reconstruction accuracy while incurring minimal computational overhead. Besides, we show that our method is effective for both synthetic and realistic blurs compared to these baselines. The performance gap between our model and the state-of-the-art becomes more prominent when testing with unseen and strong blur levels. Specifically, our model demonstrates surprising deblurring performance on these images with PSNR improvements of around 1 dB. Our code is publicly available at https://github.com/wuqiuche/Ada-Deblur.
ABSTRACT
Petabytes of health data are collected annually across the globe in electronic health records (EHR), including significant information stored as unstructured free text. However, the lack of effective mechanisms to securely share clinical text has inhibited its full utilization. We propose a new method, DataSifterText, to generate partially synthetic clinical free-text that can be safely shared between stakeholders (e.g., clinicians, STEM researchers, engineers, analysts, and healthcare providers), limiting the re-identification risk while providing significantly better utility preservation than suppressing or generalizing sensitive tokens. The method creates partially synthetic free-text data, which inherits the joint population distribution of the original data, and disguises the location of true and obfuscated words. Under certain obfuscation levels, the resulting synthetic text was sufficiently altered with different choices, orders, and frequencies of words compared to the original records. The differences were comparable to machine-generated (fully synthetic) text reported in previous studies. We applied DataSifterText to two medical case studies. In the CDC work injury application, using privacy protection, 60.9-86.5% of the synthetic descriptions belong to the same cluster as the original descriptions, demonstrating better utility preservation than the naïve content suppressing method (45.8-85.7%). In the MIMIC III application, the generated synthetic data maintained over 80% of the original information regarding patients' overall health conditions. The reported DataSifterText statistical obfuscation results indicate that the technique provides sufficient privacy protection (low identification risk) while preserving population-level information (high utility).
Subject(s)
Electronic Health Records , Privacy , HumansABSTRACT
Health advances are contingent on continuous development of new methods and approaches to foster data-driven discovery in the biomedical and clinical sciences. Open-science and team-based scientific discovery offer hope for tackling some of the difficult challenges associated with managing, modeling, and interpreting of large, complex, and multisource data. Translating raw observations into useful information and actionable knowledge depends on effective domain-independent reproducibility, area-specific replicability, data curation, analysis protocols, organization, management and sharing of health-related digital objects. This study expands the functionality and utility of an ensemble semi-supervised machine learning technique called Compressive Big Data Analytics (CBDA). Applied to high-dimensional data, CBDA (1) identifies salient features and key biomarkers enabling reliable and reproducible forecasting of binary, multinomial and continuous outcomes (i.e., feature mining); and (2) suggests the most accurate algorithms/models for predictive analytics of the observed data (i.e., model mining). The method relies on iterative subsampling, combines function optimization and statistical inference, and generates ensemble predictions for observed univariate outcomes. The novelty of this study is highlighted by a new and expanded set of CBDA features including (1) efficiently handling extremely large datasets (>100,000 cases and >1,000 features); (2) generalizing the internal and external validation steps; (3) expanding the set of base-learners for joint ensemble prediction; (4) introducing an automated selection of CBDA specifications; and (5) providing mechanisms to assess CBDA convergence, evaluate the prediction accuracy, and measure result consistency. To ground the mathematical model and the corresponding computational algorithm, CBDA 2.0 validation utilizes synthetic datasets as well as a population-wide census-like study. Specifically, an empirical validation of the CBDA technique is based on a translational health research using a large-scale clinical study (UK Biobank), which includes imaging, cognitive, and clinical assessment data. The UK Biobank archive presents several difficult challenges related to the aggregation, harmonization, modeling, and interrogation of the information. These problems are related to the complex longitudinal structure, variable heterogeneity, feature multicollinearity, incongruency, and missingness, as well as violations of classical parametric assumptions. Our results show the scalability, efficiency, and usability of CBDA to interrogate complex data into structural information leading to derived knowledge and translational action. Applying CBDA 2.0 to the UK Biobank case-study allows predicting various outcomes of interest, e.g., mood disorders and irritability, and suggests new and exciting avenues of evidence-based research in the context of identifying, tracking, and treating mental health and aging-related diseases. Following open-science principles, we share the entire end-to-end protocol, source-code, and results. This facilitates independent validation, result reproducibility, and team-based collaborative discovery.
Subject(s)
Data Mining/methods , Data Science/methods , Algorithms , Big Data , Data Compression , Humans , Machine Learning , Meta-Analysis as Topic , Models, Theoretical , Physical Phenomena , Prognosis , Reproducibility of Results , SoftwareABSTRACT
There are no practical and effective mechanisms to share high-dimensional data including sensitive information in various fields like health financial intelligence or socioeconomics without compromising either the utility of the data or exposing private personal or secure organizational information. Excessive scrambling or encoding of the information makes it less useful for modelling or analytical processing. Insufficient preprocessing may compromise sensitive information and introduce a substantial risk for re-identification of individuals by various stratification techniques. To address this problem, we developed a novel statistical obfuscation method (DataSifter) for on-the-fly de-identification of structured and unstructured sensitive high-dimensional data such as clinical data from electronic health records (EHR). DataSifter provides complete administrative control over the balance between risk of data re-identification and preservation of the data information. Simulation results suggest that DataSifter can provide privacy protection while maintaining data utility for different types of outcomes of interest. The application of DataSifter on a large autism dataset provides a realistic demonstration of its promise practical applications.
ABSTRACT
Chinese herbal medicine (CHM) has been used for treating insomnia for centuries. The most used CHM for insomnia was Polygonum multiflorum. However, the molecular mechanism for CHM preventing insomnia is unknown. Stilbene glucoside (THSG), an important active component of P. multiflorum, may play an important role for treating insomnia. To test the hypothesis, Kunming mice were treated with different dosages of THSG. To examine the sleep duration, a computer-controlled sleep-wake detection system was implemented. Electroencephalogram (EEG) and electromyogram (EMG) electrodes were implanted to determine sleep-wake state. RT-PCR and Western blot was used to measure the levels of lactate dehydrogenase (LDH) and saliva alpha amylase. Spearman's rank correlation coefficient was used to identify the strength of correlation between the variables. The results showed that THSG significantly prolonged the sleep time of the mice (p<0.01). THSG changed sleep profile by reducing wake and rapid eye movement (REM) period, and increasing non-REM period. RT-PCR and Western blot analysis showed that THSG could down-regulate the levels of LDH and saliva alpha amylase (p<0.05). The level of lactate and glucose was positively related with the activity of LDH and saliva alpha amylase (p<0.05), respectively. On the other hand, the activities of LDH and amylase were negatively associated with sleep duration (p<0.05). The levels of lactate and glucose affect sleep homeostasis. Thus, THSG may prevent insomnia by regulating sleep duration via LDH and salivary alpha amylase.
Subject(s)
Glucosides/pharmacology , L-Lactate Dehydrogenase/metabolism , Polygonum/chemistry , Salivary alpha-Amylases/metabolism , Sleep/drug effects , Stilbenes/pharmacology , Animals , Dose-Response Relationship, Drug , Electrodes , Glucosides/chemistry , Glucosides/isolation & purification , Homeostasis/drug effects , L-Lactate Dehydrogenase/genetics , Mice , Mice, Inbred Strains , Models, Molecular , Molecular Structure , Salivary alpha-Amylases/genetics , Stilbenes/chemistry , Stilbenes/isolation & purification , Structure-Activity RelationshipABSTRACT
A 56-year-old Asian woman was admitted to hospital for the consideration of hemodialysis (HD). A right femoral dialysis catheter was inserted for HD. Three months after removal of catheter, she was admitted because of right inguinal swelling. A thrill and bruit were felt and heard at the inguinal area. Color Doppler detected a fistula between right superficial femoral artery and right common femoral vein and subsequently confirmed by contrast enhanced computed tomography scan and 3-dimensional reconstruction with computed tomography. At surgery, a 4-mm-diameter fistula was found between the right superficial femoral artery and right common femoral vein. A primary closure of both defects in the artery and vein was then carried out. A follow-up digital vascular study 3 months after surgical repair was normal. In conclusion, nephrologist should have a heightened awareness to the potential of this complication and should at least document a normal exam following the removal of femoral catheters.
Subject(s)
Arteriovenous Fistula/surgery , Catheterization/adverse effects , Kidney Failure, Chronic/complications , Renal Dialysis/methods , Female , Femoral Vein/surgery , Humans , Middle AgedABSTRACT
The aim of this study is to evaluate Chinese herbs' efficacy on adhesive properties of Escherichia coli (E. coli). The effects of Chinese herbal solution on the hemagglutination and adhesion by E. coli strain were studied. E. coli C16 was isolated from a patient with urinary tract infection. The MIC value of herbal solution for the E. coli C16 was 0.1g/ml. The MBC value was 0.2g/ml. The effects of herbal solution on the hemagglutination abilities of E. coli C16 were dependent on the herbal solution used. The strain C16 lost half of its hemagglutination abilities when the herbal solution concentration was at MIC (0.05g/ml). Herbal solution decreased the adherence of strain C16 in a dose-dependent way. The numbers of adherent bacteria were reduced to 45% of the control values after growth with herbal solution at MIC. The results show that anti-adhesion is one mode of action for Chinese herbs used against pathogens.