39 results match your criteria vae improved

Integrated multi-omics analysis of ovarian cancer using variational autoencoders.

Sci Rep 2021 Mar 18;11(1):6265. Epub 2021 Mar 18.

School of Health and Life Sciences, Teesside University, Middlesbrough, TS4 3BX, UK.

Cancer is a complex disease that deregulates cellular functions at various molecular levels (e.g., DNA, RNA, and proteins). Read More

View Article and Full-Text PDF

A deep survival interpretable radiomics model of hepatocellular carcinoma patients.

Phys Med 2021 Feb 10;82:295-305. Epub 2021 Mar 10.

Department of Machine Learning, Moffitt Cancer Center, Tampa, FL, USA.

This work aims to identify a new radiomics signature using imaging phenotypes and clinical variables for risk prediction of overall survival (OS) in hepatocellular carcinoma (HCC) patients treated with stereotactic body radiation therapy (SBRT). 167 patients were retrospectively analyzed with repeated nested cross-validation to mitigate overfitting issues. 56 radiomic features were extracted from pre-treatment contrast-enhanced (CE) CT images. Read More

View Article and Full-Text PDF
February 2021

Long-Term Survival of a Patient with Recurrent Dedifferentiated High-Grade Liposarcoma of the Retroperitoneum Under Adjuvant Treatment with L. Extract: A Case Report.

Integr Cancer Ther 2021 Jan-Dec;20:1534735421995258

Institute for Applied Epistemology and Medical Methodology at the University of Witten/Herdecke, Freiburg, Germany.

Case Presentation: A 68-year-old male patient underwent a resection of a DDL (T2 N0 M0, FNCLCC grade 2, stage IIIA) in the retroperitoneum. Three months after this first surgery, a recurrence occurred, and was treated with neoadjuvant and adjuvant doxorubicin plus ifosfamide and surgery (resection). A second recurrence-11 months after the second surgery-was treated with surgery and radiotherapy. Read More

View Article and Full-Text PDF
February 2021

End-to-End Learnt Image Compression via Non-Local Attention Optimization and Improved Context Modeling.

IEEE Trans Image Process 2021 25;30:3179-3191. Epub 2021 Feb 25.

This article proposes an end-to-end learnt lossy image compression approach, which is built on top of the deep nerual network (DNN)-based variational auto-encoder (VAE) structure with Non-Local Attention optimization and Improved Context modeling (NLAIC). Our NLAIC 1) embeds non-local network operations as non-linear transforms in both main and hyper coders for deriving respective latent features and hyperpriors by exploiting both local and global correlations, 2) applies attention mechanism to generate implicit masks that are used to weigh the features for adaptive bit allocation, and 3) implements the improved conditional entropy modeling of latent features using joint 3D convolutional neural network (CNN)-based autoregressive contexts and hyperpriors. Towards the practical application, additional enhancements are also introduced to speed up the computational processing (e. Read More

View Article and Full-Text PDF
February 2021

Enhancing vigilance for cerebral air embolism after pneumonectomy: a case report.

BMC Pulm Med 2021 Jan 7;21(1):16. Epub 2021 Jan 7.

Department of Thoracic Surgery, Shenzhen Hospital, Southern Medical University, No.1333 Xinhu Road, Baoan District, Shenzhen, 518101, Guangdong, China.

Background: Vascular air embolism (VAE) is a rare but important complication that has not been paid enough attention to in the medical process such as surgery and anesthesia.

Case Presentation: We report for the first time that a 54-year-old male patient with central lung cancer developed severe complications of CAE after right pneumonectomy. After targeted first-aid measures such as assisted breathing, mannitol dehydration and antibiotic treatment, the patient gradually improved. Read More

View Article and Full-Text PDF
January 2021

Conditional out-of-distribution generation for unpaired data using transfer VAE.

Bioinformatics 2020 12;36(Suppl_2):i610-i617

Institute of Computational Biology, Helmholtz Center Munich, Neuherberg, Germany.

Motivation: While generative models have shown great success in sampling high-dimensional samples conditional on low-dimensional descriptors (stroke thickness in MNIST, hair color in CelebA, speaker identity in WaveNet), their generation out-of-distribution poses fundamental problems due to the difficulty of learning compact joint distribution across conditions. The canonical example of the conditional variational autoencoder (CVAE), for instance, does not explicitly relate conditions during training and, hence, has no explicit incentive of learning such a compact representation.

Results: We overcome the limitation of the CVAE by matching distributions across conditions using maximum mean discrepancy in the decoder layer that follows the bottleneck. Read More

View Article and Full-Text PDF
December 2020

Crash data augmentation using variational autoencoder.

Accid Anal Prev 2021 Mar 25;151:105950. Epub 2020 Dec 25.

Department of Civil, Environmental and Construction Engineering, University of Central Florida, Orlando, FL, 32816, USA. Electronic address:

In this paper, we present a data augmentation technique to reproduce crash data. The dataset comprising crash and non-crash events are extremely imbalanced. For instance, the dataset used in this paper consists of only 625 crash events for over 6. Read More

View Article and Full-Text PDF

Annotation-Efficient Learning for Medical Image Segmentation based on Noisy Pseudo Labels and Adversarial Learning.

IEEE Trans Med Imaging 2020 Dec 28;PP. Epub 2020 Dec 28.

Despite that deep learning has achieved state-of-the-art performance for medical image segmentation, its success relies on a large set of manually annotated images for training that are expensive to acquire. In this paper, we propose an annotation-efficient learning framework for segmentation tasks that avoids annotations of training images, where we use an improved Cycle-Consistent Generative Adversarial Network (GAN) to learn from a set of unpaired medical images and auxiliary masks obtained either from a shape model or public datasets. We first use the GAN to generate pseudo labels for our training images under the implicit high-level shape constraint represented by a Variational Auto-encoder (VAE)-based discriminator with the help of the auxiliary masks, and build a Discriminator-guided Generator Channel Calibration (DGCC) module which employs our discriminator's feedback to calibrate the generator for better pseudo labels. Read More

View Article and Full-Text PDF
December 2020

Beneficial effects of extract in osteoporosis and osteoarthritis.

Food Sci Nutr 2020 Dec 20;8(12):6550-6556. Epub 2020 Oct 20.

Immunoregulatory Material Research Center Korea Research Institute of Bioscience and Biotechnology Jeongeup-si Korea.

In Asia, (azuki bean) has been used as a traditional medicine to treat various diseases because of its biological properties. Osteoarthritis (OA) and osteoporosis (OP) are common regenerative bone diseases that are characterized by deterioration of joint and bone structure. In this study, we evaluated the effects of extract (VAE) on monosodium iodoacetate (MIA)-induced OA and ovariectomy (OVX)-induced OP models. Read More

View Article and Full-Text PDF
December 2020

fNIRS Assessment of Speech Comprehension in Children with Normal Hearing and Children with Hearing Aids in Virtual Acoustic Environments: Pilot Data and Practical Recommendations.

Children (Basel) 2020 Nov 7;7(11). Epub 2020 Nov 7.

Child Neuropsychology Section, Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, Medical Faculty, RWTH Aachen University, 52074 Aachen, Germany.

The integration of virtual acoustic environments (VAEs) with functional near-infrared spectroscopy (fNIRS) offers novel avenues to investigate behavioral and neural processes of speech-in-noise (SIN) comprehension in complex auditory scenes. Particularly in children with hearing aids (HAs), the combined application might offer new insights into the neural mechanism of SIN perception in simulated real-life acoustic scenarios. Here, we present first pilot data from six children with normal hearing (NH) and three children with bilateral HAs to explore the potential applicability of this novel approach. Read More

View Article and Full-Text PDF
November 2020

Collective dynamics of repeated inference in variational autoencoder rapidly find cluster structure.

Sci Rep 2020 09 29;10(1):16001. Epub 2020 Sep 29.

Department of Complexity Science and Engineering, The University of Tokyo, Chiba, 277-8561, Japan.

Deep neural networks are good at extracting low-dimensional subspaces (latent spaces) that represent the essential features inside a high-dimensional dataset. Deep generative models represented by variational autoencoders (VAEs) can generate and infer high-quality datasets, such as images. In particular, VAEs can eliminate the noise contained in an image by repeating the mapping between latent and data space. Read More

View Article and Full-Text PDF
September 2020

Continuous Molecular Representations of Ionic Liquids.

J Phys Chem B 2020 09 9;124(38):8347-8357. Epub 2020 Sep 9.

Department of Chemical Engineering, University of Washington, Seattle, Washington 98105, United States.

Designing new ionic liquids (ILs) is of crucial importance for various industrial applications. However, this always leads to a daunting challenge, as the number of possible combinations of cation and anion are very high and it is impossible to experimentally propose and screen a wide pool of potential candidates. However, recent applications of machine learning (ML) models have greatly improved the overall chemical discovery pipeline. Read More

View Article and Full-Text PDF
September 2020

Effects of External Environments on the Fixed Elongation and Tensile Properties of the VAE Emulsion-Cement Composite Joint Sealant.

Materials (Basel) 2020 Jul 21;13(14). Epub 2020 Jul 21.

Aeronautics Engineering College, Air Force Engineering University, Xi'an 710038, China.

Joint sealant is affected by various environmental factors in service, such as different temperatures, water soaking, ultraviolet and so on. In this paper, the VAE emulsion-cement compositejoint sealant was pretreated under multiple simulation environments. Thereafter, the degradation rules of fixed elongation and tensile properties of joint sealants at different mix proportions were systemically investigated under the action of external environments (temperature, water soaking and ultraviolet), and the influence mechanisms of diverse environmental factors were analyzed. Read More

View Article and Full-Text PDF

Using the VQ-VAE to improve the recognition of abnormalities in short-duration 12-lead electrocardiogram records.

Comput Methods Programs Biomed 2020 Nov 4;196:105639. Epub 2020 Jul 4.

Department of Cardiology, The Second Affiliated Hospital of Chongqing Medical University, 74 Linjiang Road, Chongqing, 400010, China. Electronic address:

Background And Objective: Morphological diagnosis is a basic clinical task of the short-duration 12-lead electrocardiogram (ECG). Due to the scarcity of positive samples and other factors, there is currently no algorithm that is comparable to human experts in ECG morphological recognition. Our objective is to develop an ECG specialist-level deep learning method that can accurately identify ten ECG morphological abnormalities in real scene data. Read More

View Article and Full-Text PDF
November 2020

Improving the Structure-Function Relationship in Glaucomatous Visual Fields by Using a Deep Learning-Based Noise Reduction Approach.

Ophthalmol Glaucoma 2020 May - Jun;3(3):210-217. Epub 2020 Jan 11.

Department of Ophthalmology, Kagoshima University Graduate School of Medical and Dental Sciences, Kagoshima, Japan.

Purpose: To investigate whether processing visual field (VF) measurements using a variational autoencoder (VAE) improves the structure-function relationship in glaucoma.

Design: Cross-sectional study.

Participants: The training data consisted of 82 433 VF measurements from 16 836 eyes. Read More

View Article and Full-Text PDF
January 2020

Outcomes in single-level posterior cervical spine surgeries performed in the sitting and prone positions.

J Neurosurg Spine 2020 Jul 3:1-7. Epub 2020 Jul 3.

Departments of1Neurologic Surgery and.

Objective: The sitting or semisitting position in neurosurgery allows for several technical advantages, including improved visualization of the surgical field. However, it has also been associated with an increased risk of venous air embolisms and positioning-related complications that limit its commonplace adoption. The authors report a large, single-center series of cervical spine procedures performed with patients in the sitting or prone position in order to assess the perceived risk of intraoperative and postoperative complications associated with the sitting position. Read More

View Article and Full-Text PDF

Embedding high-dimensional Bayesian optimization via generative modeling: Parameter personalization of cardiac electrophysiological models.

Med Image Anal 2020 05 27;62:101670. Epub 2020 Feb 27.

Rochester Institute of Technology, Rochester, NY, USA. Electronic address:

The estimation of patient-specific tissue properties in the form of model parameters is important for personalized physiological models. Because tissue properties are spatially varying across the underlying geometrical model, it presents a significant challenge of high-dimensional (HD) optimization at the presence of limited measurement data. A common solution to reduce the dimension of the parameter space is to explicitly partition the geometrical mesh. Read More

View Article and Full-Text PDF

Feedback Modulates Audio-Visual Spatial Recalibration.

Front Integr Neurosci 2019 17;13:74. Epub 2020 Jan 17.

Biological Psychology and Neuropsychology, University of Hamburg, Hamburg, Germany.

In an ever-changing environment, crossmodal recalibration is crucial to maintain precise and coherent spatial estimates across different sensory modalities. Accordingly, it has been found that perceived auditory space is recalibrated toward vision after consistent exposure to spatially misaligned audio-visual stimuli (VS). While this so-called ventriloquism aftereffect (VAE) yields internal consistency between vision and audition, it does not necessarily lead to consistency between the perceptual representation of space and the actual environment. Read More

View Article and Full-Text PDF
January 2020

Long-term survival of a patient with an inoperable thymic neuroendocrine tumor stage IIIa under sole treatment with Viscum album extract: A CARE compliant clinical case report.

Medicine (Baltimore) 2020 Jan;99(5):e18990

Institute for Applied Epistemology and Medical Methodology at the University of Witten/Herdecke.

Rationale: Thymic neuroendocrine tumor (TNET) is very rare and characterized by a tendency to invade adjacent structures, frequent metastasis, resistance to therapy, and a poor prognosis. Viscum album extracts (VAE) have shown immunological, apoptogenic, and cytotoxic properties.

Patient Concerns: A 54-year-old Peruvian man was suffering from constant fatigue, cough, dyspnea, and fever for a couple of months. Read More

View Article and Full-Text PDF
January 2020

Estimating Rates of Progression and Predicting Future Visual Fields in Glaucoma Using a Deep Variational Autoencoder.

Sci Rep 2019 12 2;9(1):18113. Epub 2019 Dec 2.

Duke Eye Center and Department of Ophthalmology, Duke University, Durham, North Carolina, USA.

In this manuscript we develop a deep learning algorithm to improve estimation of rates of progression and prediction of future patterns of visual field loss in glaucoma. A generalized variational auto-encoder (VAE) was trained to learn a low-dimensional representation of standard automated perimetry (SAP) visual fields using 29,161 fields from 3,832 patients. The VAE was trained on a 90% sample of the data, with randomization at the patient level. Read More

View Article and Full-Text PDF
December 2019

Management of Congenital Aniridia-Associated Keratopathy: Long-Term Outcomes from a Tertiary Referral Center.

Am J Ophthalmol 2020 02 12;210:8-18. Epub 2019 Nov 12.

Department of Ophthalmology and Visual Sciences, Illinois Eye and Ear Infirmary, University of Illinois at Chicago, Chicago, Illinois, USA. Electronic address:

Purpose: To report the outcomes of medical and surgical management for congenital aniridia-associated keratopathy (AAK) over a long-term follow-up period.

Design: Retrospective, comparative case series.

Methods: Medical records of patients diagnosed with congenital aniridia were retrospectively reviewed. Read More

View Article and Full-Text PDF
February 2020

Bowman's topography for improved detection of early ectasia.

J Biophotonics 2019 10 27;12(10):e201900126. Epub 2019 Jun 27.

Imaging, Biomechanics and Mathematical Modeling Solutions, Narayana Nethralaya Foundation, Bangalore, India.

The aim of this study was to evaluate whether OCT topography of the Bowman's layer and artificial intelligence (AI) can result in better diagnosis of forme fruste (FFKC) and clinical keratoconus (KC). Normal (n = 221), FFKC (n = 72) and KC (n = 116) corneas were included. Some of the FFKC and KC patients had the fellow eye (VAE-NT) with normal topography (n = 30). Read More

View Article and Full-Text PDF
October 2019

Long-term survival of a patient with advanced pancreatic cancer under adjunct treatment with extracts: A case report.

World J Gastroenterol 2019 Mar;25(12):1524-1530

Department of Methodology, Institute for Applied Epistemology and Medical Methodology (IFAEMM), Freiburg 79111, Germany.

Background: Advanced pancreatic cancer (aPC) has a poor prognosis with limited survival benefit from current standard treatment. extracts (VAE) are used by many cancer patients, showing immune-stimulating effects, improved quality of life, and a survival benefit in patients with aPC.

Case Summary: A 59-year-old architect developed epigastric pain. Read More

View Article and Full-Text PDF

Exploring semi-supervised variational autoencoders for biomedical relation extraction.

Methods 2019 08 27;166:112-119. Epub 2019 Feb 27.

National Center for Biotechnology Information (NCBI), National Library of Medicine (NLM), National Institutes of Health (NIH), Bethesda, MD 20894, USA. Electronic address:

The biomedical literature provides a rich source of knowledge such as protein-protein interactions (PPIs), drug-drug interactions (DDIs) and chemical-protein interactions (CPIs). Biomedical relation extraction aims to automatically extract biomedical relations from biomedical text for various biomedical research. State-of-the-art methods for biomedical relation extraction are primarily based on supervised machine learning and therefore depend on (sufficient) labeled data. Read More

View Article and Full-Text PDF

A Nurse-Driven Oral Care Protocol to Reduce Hospital-Acquired Pneumonia.

Am J Nurs 2019 02;119(2):44-51

Chastity Warren is an assistant professor at the Michigan State University College of Nursing and a clinical nurse specialist in critical care at Sparrow Health System, Lansing, MI, where Mary Kathryn Medei and Brooke Wood are RNs in the surgical specialties department. Debra Schutte is an associate professor at the Wayne State University College of Nursing in Detroit and a nurse research consultant in the nursing education and practice department at Sparrow Health System. The following funding sources contributed to the completion of this project: the Sparrow Hospital Department of Nursing Evidence-Based Practice Fellowship Program, Sparrow Foundation Sue Tadgerson Nursing Research Fund, and Delta Dental Foundation, an affiliate of Delta Dental of Michigan, Ohio, and Indiana, which provided an unrestricted educational grant. Contact author: Chastity Warren, The authors have disclosed no potential conflicts of interest, financial or otherwise.

: Purpose: The purpose of this evidence-based quality improvement (QI) project was to implement an oral care protocol in the adult in-patient care areas of a level 1 trauma hospital and to evaluate its impact on the incidence of hospital-acquired pneumonia (HAP).

Methods: A standardized, evidence-based oral care protocol was implemented depending on the level of care required by each ventilated, at-risk, or short-term care patient. The QI project included the introduction of a new suction toothbrush kit for at-risk patients and, for short-term patients, a new short-term oral care kit that featured a more ergonomically appropriate toothbrush, a baking soda toothpaste, and an alcohol-free antiseptic mouthwash; the project also supported the continued use of the then current suction toothbrush kit for patients receiving mechanical ventilation. Read More

View Article and Full-Text PDF
February 2019

Long-Term Tumor-Free Survival in a Patient with Stage IV Epithelial Ovarian Cancer Undergoing High-Dose Chemotherapy and Viscum album Extract Treatment: A Case Report.

Perm J 2018 ;23:18-025

Institute for Applied Epistemology and Medical Methodology, University of Witten/Herdecke, Freiburg, Germany.

Introduction: Epithelial ovarian cancer (EOC) has a poor prognosis in advanced stages. High-dose chemotherapy (HDC) was pursued in the 1990s but was not found to improve survival of patients with EOC in larger studies. Many patients with cancer use Viscum album extracts (VAE). Read More

View Article and Full-Text PDF

Ventilator-associated events versus ventilator-associated respiratory infections-moving into a new paradigm or merging both concepts, instead?

Ann Transl Med 2018 Nov;6(21):425

Vall d'Hebron Institut of Research, Barcelona, Spain.

Despite ventilator-associated respiratory infections (VARI) are reported as the most common and fatal complications related to mechanical ventilation (MV), they are not the unique occurrences. The new classification of ventilator-associated events (VAE) proposed by the centers for disease control and prevention (CDC) enhance the spectra of complications due to MV including both infection-related and non-infectious events. Both VAEs and VARIs are associated with prolonged duration of MV, longer stay in hospital and in the intensive care unit (ICU) and more antibiotic consumption, nonetheless patients with VAEs have worst outcomes. Read More

View Article and Full-Text PDF
November 2018

Complete Remission and Long-term Survival of a Patient with a Diffuse Large B-cell Lymphoma Under Extracts After Resistance to R-CHOP: A Case Report.

Anticancer Res 2018 Sep;38(9):5363-5369

Institute for Applied Epistemology and Medical Methodology, University of Witten/Herdecke, Freiburg, Germany.

Background: A nodular lymphocyte predominant Hodgkin's lymphoma (NLPHL) is a lymphoproliferative neoplasm with a fair prognosis, but the possibility of a malignant transformation into a diffuse large B-cell lymphoma (DLBCL) is high. DLBCL progresses aggressively. Introduction of rituximab into therapy had led to improved outcomes. Read More

View Article and Full-Text PDF
September 2018

Novel deep generative simultaneous recurrent model for efficient representation learning.

Neural Netw 2018 Nov 9;107:12-22. Epub 2018 Aug 9.

The Vision Lab in Department of Electrical and Computer Engineering, Old Dominion University, Norfolk, VA 23529, United States. Electronic address:

Representation learning plays an important role for building effective deep neural network models. Deep generative probabilistic models have shown to be efficient in the data representation learning task which is usually carried out in an unsupervised fashion. Throughout the past decade, there has been almost exclusive focus on the learning algorithms to improve representation capability of the generative models. Read More

View Article and Full-Text PDF
November 2018

Contemporary analysis of the intraoperative and perioperative complications of neurosurgical procedures performed in the sitting position.

J Neurosurg 2017 Jul 5;127(1):182-188. Epub 2016 Aug 5.

Departments of 1 Neurologic Surgery and.

OBJECTIVE Historically, performing neurosurgery with the patient in the sitting position offered advantages such as improved visualization and gravity-assisted retraction. However, this position fell out of favor at many centers due to the perceived risk of venous air embolism (VAE) and other position-related complications. Some neurosurgical centers continue to perform sitting-position cases in select patients, often using modern monitoring techniques that may improve procedural safety. Read More

View Article and Full-Text PDF