260,382 results match your criteria deep supervision


[Deep vein thrombosis - diagnostics and clarification].

Authors:
Lars H Lindner

Dtsch Med Wochenschr 2021 Jun 15;146(12):832-836. Epub 2021 Jun 15.

Deep vein thrombosis usually manifests as leg or pelvic vein thrombosis (DVT). The causes are either acquired or inherited and can also occur in combination. Early diagnosis and treatment of DVT can reduce the risk of pulmonary embolism and postthrombotic syndrome. Read More

View Article and Full-Text PDF

Impairment of vestibulo-collic reflex and linear vestibulo-ocular reflex in pediatric-onset multiple sclerosis patients.

Clin Neurophysiol 2021 May 24;132(8):1813-1819. Epub 2021 May 24.

Hacettepe University, Faculty of Medicine, Department of Pediatric Neurology, Ankara, Turkey.

Objectives: This study aimed to examine the vestibulo-collic reflex (VCR) and linear vestibulo-ocular reflex (lVOR) and their correlation with brain lesions in pediatric-onset multiple sclerosis (POMS).

Methods: The study group consisted of 17 patients (34 ears) with POMS (mean age 18.73 ± 2. Read More

View Article and Full-Text PDF

Abdominal Wall Thickness Predicts Surgical Site Infection in Emergency Colon Operations.

J Surg Res 2021 Jun 12;267:37-47. Epub 2021 Jun 12.

Division of Trauma, Emergency Surgery, and Surgical Critical Care, Massachusetts General Hospital, Boston, Massachusetts; Harvard Medical School, Boston, Massachusetts. Electronic address:

Background: Body mass index (BMI) does not reliably predict Surgical site infections (SSI). We hypothesize that abdominal wall thickness (AWT) would serve as a better predictor of SSI for patients undergoing emergency colon operations.

Methods: We retrospectively evaluated our Emergency Surgery Database (2007-2018). Read More

View Article and Full-Text PDF

Convolutional neural networks to identify malformations of cortical development: A feasibility study.

Seizure 2021 May 31;91:81-90. Epub 2021 May 31.

Division of Epilepsy and Clinical Neurophysiology, Department of Neurology, Boston Children's Hospital and Harvard Medical School, Boston, MA USA.

Objective: To develop and test a deep learning model to automatically detect malformations of cortical development (MCD).

Methods: We trained a deep learning model to distinguish between diffuse cortical malformation (CM), periventricular nodular heterotopia (PVNH), and normal magnetic resonance imaging (MRI). We trained 4 different convolutional neural network (CNN) architectures. Read More

View Article and Full-Text PDF

Removal of trace DNA toxic compounds using a Poly(deep eutectic solvent)@Biomass based on multi-physical interactions.

J Hazard Mater 2021 Jun 9;418:126369. Epub 2021 Jun 9.

College of Pharmaceutical Science, Institute of Life Science and Green Development, Key Laboratory of Medicinal Chemistry and Molecular Diagnosis of Ministry of Education, Hebei University, Baoding 071002, China. Electronic address:

DNA toxic compounds (DNA-T-Cs), even in trace amounts, seriously threaten human health and must be completely eliminated. However, the currently used separation media face great challenges in removing trace DNA-T-Cs. Based on the functional advantages of deep eutectic solvents (DESs) and the natural features of biomass (BioM), a series of Poly(DES)@BioMs functioning as adsorbents were prepared for the removal of aromatic/hetero-atomic DNA-T-Cs at the ppm level. Read More

View Article and Full-Text PDF

Voxel-to-voxel predictive models reveal unexpected structure in unexplained variance.

Neuroimage 2021 Jun 12:118266. Epub 2021 Jun 12.

Department of Neuroscience, Medical University of South Carolina, Charleston, SC, USA. Electronic address:

Encoding models based on deep convolutional neural networks (DCNN) predict BOLD responses to natural scenes in the human visual system more accurately than many other currently available models. However, DCNN-based encoding models fail to predict a significant amount of variance in the activity of most voxels in all visual areas. This failure could reflect limitations in the data (e. Read More

View Article and Full-Text PDF

Artificial Intelligence in Pathology: Easing the Burden of Annotation Running Title: AI Annotation Reduction.

Am J Pathol 2021 Jun 12. Epub 2021 Jun 12.

Emeritus Professor, Rutgers-NJMS; Department of Pathology and Laboratory Medicine Adjunct Professor, Perelman Medical School; Adjunct Professor, Sidney Kimmel School of Medicine, Philadelphia, PA. Electronic address:

The need for huge datasets represents a bottleneck for applications of artificial intelligence. In Pathology, an additional problem is that there are often substantially fewer annotated target lesions than normal tissues for comparison. Organic brains overcome these limitations by utilizing large numbers of specialized neural nets arranged in both linear and parallel fashion, with each solving a restricted classification problem. Read More

View Article and Full-Text PDF

Artificial Intelligence and Cellular Segmentation in Tissue Microscopy Images.

Am J Pathol 2021 Jun 12. Epub 2021 Jun 12.

Department of Radiology and Committee on Medical Physics, University of Chicago, Chicago, IL, USA, 60637. Electronic address:

With applications in object detection, image feature extraction, image classification, and image segmentation, artificial intelligence is enabling high-throughput analysis of image data in a variety of biomedical imaging disciplines, ranging from radiology and pathology to cancer biology and immunology. Specifically, a growth in research surrounding deep learning has led to widespread application of computer vision techniques to analyze and mine data from biomedical images. The availability of open-source software packages and the development of novel, trainable deep neural network architectures has led to an increase in accuracy of cell detection and segmentation algorithms. Read More

View Article and Full-Text PDF

Big and deep seated lipomatous tumours in children : results of surgical treatment.

Acta Orthop Belg 2021 Mar;87(1):197-200

The objective of the study is aimed to evaluate results of our pediatric patients with big and deep-seated lipomatous tumors Results of 32 children who underwent resection for 5 cm or larger and deep-seated lipomas were evaluated. The mean age of the patients was 9.1 years (range, 0-16 ; 11 female/21 male), and median follow-up period was 3. Read More

View Article and Full-Text PDF

Solving the enigma of posterolateral tibial plateau fractures, the clue protocol.

Acta Orthop Belg 2021 Mar;87(1):125-136

The study aim is to evaluate functional and radio- logical outcomes following a suggested protocol based on the four-column classification for management of posterolateral column tibial plateau fractures. A prospective cohort study was performed in level I academic center on 42 patients with mean age of 36 years (22-59). Eleven patients had isolated posterolateral column fractures whereas 31 patients had associated columns fractures. Read More

View Article and Full-Text PDF

Influence of outpatient total knee arthroplasty compared to inpatient surgery on medical and economic outcomes.

Acta Orthop Belg 2021 Mar;87(1):103-109

Firstly, this study compared the rate of readmission after a total knee arthroplasty between selected out- patients (no hospitalization, directly sent home after surgery) and inpatients (3 days hospitalization) at 6 weeks. Secondly, it examined the mobility and the complications in the two groups after the same period of time. The rate of readmission, complications and knee mobility of 32 outpatients (M-age : 61 years ± 10 ; 10 females), were compared against those of 32 birth- matched inpatients (M-age : 64 years ± 8. Read More

View Article and Full-Text PDF

Low blood transfusion rate after implementation of tranexamic acid for fast- track hip- and knee arthroplasty. An observational study of 5205 patients.

Acta Orthop Belg 2021 Mar;87(1):9-16

The purpose of this study was to retrospectively evaluate the efficacy of a tranexamic acid (TXA) perioperative protocol for primary hip- and knee arthroplasty, in terms of allogenic blood transfusion rates. A retrospective cohort study was conducted and included all primary hip and knee arthroplasty procedures in the period of 2014-2019. Patients who underwent surgery due to trauma or revision were excluded. Read More

View Article and Full-Text PDF

Anatomically aided PET image reconstruction using deep neural networks.

Med Phys 2021 Jun 15. Epub 2021 Jun 15.

Department of Biomedical Engineering, University of California, Davis, CA, USA.

Purpose: The developments of PET/CT and PET/MR scanners provide opportunities for improving PET image quality by using anatomical information. In this paper, we propose a novel co-learning 3D convolutional neural network (CNN) to extract modality-specific features from PET/CT image pairs and integrate complementary features into an iterative reconstruction framework to improve PET image reconstruction.

Methods: We used a pre-trained deep neural network to represent PET images. Read More

View Article and Full-Text PDF

Unraveling the Interplay of Image Formation, Data Representation and Learning in CT-based COPD Phenotyping Automation: The Need for a Meta-Strategy.

Med Phys 2021 Jun 15. Epub 2021 Jun 15.

Siemens Healthineers, CT R&D Image Analytics, Forchheim, Germany.

Purpose: In the literature on automated phenotyping of Chronic Obstructive Pulmonary Disease (COPD), there is a multitude of isolated classical machine learning and deep learning techniques, mostly investigating individual phenotypes, with small study cohorts and heterogeneous meta-parameters, e.g. different scan protocols or segmented regions. Read More

View Article and Full-Text PDF

High-speed computer-generated holography using an autoencoder-based deep neural network.

Opt Lett 2021 Jun;46(12):2908-2911

Learning-based computer-generated holography (CGH) provides a rapid hologram generation approach for holographic displays. Supervised training requires a large-scale dataset with target images and corresponding holograms. We propose an autoencoder-based neural network (holoencoder) for phase-only hologram generation. Read More

View Article and Full-Text PDF

REPAID: resolution-enhanced plenoptic all-in-focus imaging using deep neural networks.

Opt Lett 2021 Jun;46(12):2896-2899

Due to limited depth-of-focus, classical 2D images inevitably lose details of targets out of depth-of-focus, while all-in-focus images break through the limit by fusing multi-focus images, thus being able to focus on targets in extended depth-of-view. However, conventional methods can hardly obtain dynamic all-in-focus imaging in both high spatial and temporal resolutions. To solve this problem, we design REPAID, meaning resolution-enhanced plenoptic all-in-focus imaging using deep neural networks. Read More

View Article and Full-Text PDF

Anthropogenic lead pervasive in Canadian Arctic seawater.

Proc Natl Acad Sci U S A 2021 Jun;118(24)

Department of Earth Sciences, University of Toronto, Toronto, ON M5S 3B1, Canada;

Anthropogenic Pb is widespread in the environment including remote places. However, its presence in Canadian Arctic seawater is thought to be negligible based on low dissolved Pb (dPb) concentrations and proxy data. Here, we measured dPb isotopes in Arctic seawater with very low dPb concentrations (average ∼5 pmol ⋅ kg) and show that anthropogenic Pb is pervasive and often dominant in the western Arctic Ocean. Read More

View Article and Full-Text PDF

Multi-level Attention Network for Retinal Vessel Segmentation.

IEEE J Biomed Health Inform 2021 Jun 15;PP. Epub 2021 Jun 15.

Automatic vessel segmentation in the fundus images plays an important role in the screening, diagnosis, treatment, and evaluation of various cardiovascular and ophthalmologic diseases. However, due to the limited well-annotated data, varying size of vessels, and intricate vessel structures, retinal vessel segmentation has become a long-standing challenge. In this paper, a novel deep learning model called AACA-MLA-D-UNet is proposed to fully utilize the low-level detailed information and the complementary information encoded in different layers to accurately distinguish the vessels from the background with low model complexity. Read More

View Article and Full-Text PDF

A Deep Probabilistic Transfer Learning Framework for Soft Sensor Modeling With Missing Data.

IEEE Trans Neural Netw Learn Syst 2021 Jun 15;PP. Epub 2021 Jun 15.

Soft sensors have been extensively developed and applied in the process industry. One of the main challenges of the data-driven soft sensors is the lack of labeled data and the need to absorb the knowledge from a related source operating condition to enhance the soft sensing performance on the target application. This article introduces deep transfer learning to soft sensor modeling and proposes a deep probabilistic transfer regression (DPTR) framework. Read More

View Article and Full-Text PDF

SeReNe: Sensitivity-Based Regularization of Neurons for Structured Sparsity in Neural Networks.

IEEE Trans Neural Netw Learn Syst 2021 Jun 15;PP. Epub 2021 Jun 15.

Deep neural networks include millions of learnable parameters, making their deployment over resource-constrained devices problematic. Sensitivity-based regularization of neurons (SeReNe) is a method for learning sparse topologies with a structure, exploiting neural sensitivity as a regularizer. We define the sensitivity of a neuron as the variation of the network output with respect to the variation of the activity of the neuron. Read More

View Article and Full-Text PDF

Capsule Attention for Multimodal EEG-EOG Representation Learning with Application to Driver Vigilance Estimation.

IEEE Trans Neural Syst Rehabil Eng 2021 Jun 15;PP. Epub 2021 Jun 15.

Driver vigilance estimation is an important task for transportation safety. Wearable and portable brain-computer interface devices provide a powerful means for real-time monitoring of the vigilance level of drivers to help with avoiding distracted or impaired driving. In this paper, we propose a novel multimodal architecture for in-vehicle vigilance estimation from Electroencephalogram and Electrooculogram. Read More

View Article and Full-Text PDF

Medical Image Segmentation with Deep Atlas Prior.

IEEE Trans Med Imaging 2021 Jun 15;PP. Epub 2021 Jun 15.

Organ segmentation from medical images is one of the most important pre-processing steps in computer-aided diagnosis, but it is a challenging task because of limited annotated data, low-contrast and non-homogenous textures. Compared with natural images, organs in the medical images have obvious anatomical prior knowledge (e.g. Read More

View Article and Full-Text PDF

CleftNet: Augmented Deep Learning for Synaptic Cleft Detection from Brain Electron Microscopy.

Authors:
Yi Liu Shuiwang Ji

IEEE Trans Med Imaging 2021 Jun 15;PP. Epub 2021 Jun 15.

Detecting synaptic clefts is a crucial step to investigate the biological function of synapses. The volume electron microscopy (EM) allows the identification of synaptic clefts by photoing EM images with high resolution and fine details. Machine learning approaches have been employed to automatically predict synaptic clefts from EM images. Read More

View Article and Full-Text PDF

Transanal Ileal Pouch: Is It Better?

J Laparoendosc Adv Surg Tech A 2021 Jun 14. Epub 2021 Jun 14.

Department of Colon and Rectal Surgery, Cedars-Sinai Medical Center, Los Angeles, California, USA.

Restorative proctocolectomy with ileal pouch-anal anastomosis (IPAA) is the procedure of choice for patients with ulcerative colitis and select patients with Crohn's disease. Minimally invasive techniques have been increasingly adopted including the transanal approach. However there remains a dearth of comparative data assessing the technical advantages and outcomes of a transanal approach to the IPAA against other minimally invasive techniques. Read More

View Article and Full-Text PDF

Automated Brain Masking of Fetal Functional MRI with Open Data.

Neuroinformatics 2021 Jun 15. Epub 2021 Jun 15.

Department of Child and Adolescent Psychiatry, New York University School of Medicine, New York, NY, USA.

Fetal resting-state functional magnetic resonance imaging (rs-fMRI) has emerged as a critical new approach for characterizing brain development before birth. Despite the rapid and widespread growth of this approach, at present, we lack neuroimaging processing pipelines suited to address the unique challenges inherent in this data type. Here, we solve the most challenging processing step, rapid and accurate isolation of the fetal brain from surrounding tissue across thousands of non-stationary 3D brain volumes. Read More

View Article and Full-Text PDF

A deep convolutional visual encoding model of neuronal responses in the LGN.

Brain Inform 2021 Jun 15;8(1):11. Epub 2021 Jun 15.

Computer and Systems Engineering Department, Faculty of Engineering, Ain Shams University, 1 El-Sarayat St., Abbassia, Cairo, Egypt.

The Lateral Geniculate Nucleus (LGN) represents one of the major processing sites along the visual pathway. Despite its crucial role in processing visual information and its utility as one target for recently developed visual prostheses, it is much less studied compared to the retina and the visual cortex. In this paper, we introduce a deep learning encoder to predict LGN neuronal firing in response to different visual stimulation patterns. Read More

View Article and Full-Text PDF

A convolutional neural network for common coordinate registration of high-resolution histology images.

Bioinformatics 2021 Jun 15. Epub 2021 Jun 15.

Center for Computational Biology, Flatiron Institute, New York, NY, 10010, USA.

Motivation: Registration of histology images from multiple sources is a pressing problem in large-scale studies of spatial -omics data. Researchers often perform "common coordinate registration," akin to segmentation, in which samples are partitioned based on tissue type to allow for quantitative comparison of similar regions across samples. Accuracy in such registration requires both high image resolution and global awareness, which mark a difficult balancing act for contemporary deep learning architectures. Read More

View Article and Full-Text PDF

[Mutational oncology of lung cancer: molecular markers, drugs, negotiation conditions and experiences in national reference centers.]

Recenti Prog Med 2021 Jun;112(6):419-437

Drugs and Health srl, Roma.

The gradual availability of genomic profiling tests and the "agnostic approvals" from FDA and EMA have opened the oncology mutational model phase, which complements and integrates the traditional hystological approach. The non-small-cell lung cancer (NSCLC) is characterized by many molecular alterations and represents the need of a change from the traditional diagnostic, therapeutic and organisational paradigms to the "mutational" ones. From the Italian National Healthcare System point of view, access and reimbursement of drugs based on the hystological model were managed thorugh the Italian Medicines Agency's (AIFA) monitoring registries and the managed entry agreements: risk-sharing, cost-sharing and payment by results. Read More

View Article and Full-Text PDF

Focusing on long-term complications of mid-urethral slings among women with stress urinary incontinence as a patient safety improvement measure: A protocol for systematic review and meta-analysis.

Medicine (Baltimore) 2021 Jun;100(24):e26257

Department of Obstetrics and Gynecology, Taoyuan General Hospital, Ministry of Health and Welfare, Tao-Yuan City, Taiwan.

Background: There are 3 different types of mid-urethral sling, retropubic, transobturator and single incision performed for women with stress urinary incontinence. Prior studies comparing these three surgeries merely focused on the successful rate or efficacy. But nevertheless, what is more clinically important dwells upon investigating postoperative complications as a safety improvement measure. Read More

View Article and Full-Text PDF

Automation of Article Selection Process in Systematic Reviews Through Artificial Neural Network Modeling and Machine Learning: Protocol for an Article Selection Model.

JMIR Res Protoc 2021 Jun 15;10(6):e26448. Epub 2021 Jun 15.

Department of Science and Technology, Universidade Federal de São Paulo, São Paulo, Brazil.

Background: A systematic review can be defined as a summary of the evidence found in the literature via a systematic search in the available scientific databases. One of the steps involved is article selection, which is typically a laborious task. Machine learning and artificial intelligence can be important tools in automating this step, thus aiding researchers. Read More

View Article and Full-Text PDF