Please enable JavaScript.
Coggle requires JavaScript to display documents.
Deep Learning - Coggle Diagram
Deep Learning
Parkinson’s disease
accelerometer
sleep behavior disorder (RBD
rapid eye movement (REM)
tools
finger tapping
walking
activities
data collected
characteristic
healthy controls (HCs
changes in gait pattern
tremors
freezing of gait
lab-recorded
longitudinal measurements
fluc- tuations
heterogeneity
multifaceted nature
cross-sectional
delay diagnosis
vascular parkinsonism
corticobasal degeneration
progressive supranuclear palsy
multiple system atrophy
mischaracterized
undetected
Unified Parkinson’s Disease Rating Scale (UPDRS)
objective biomarkers
non-motor
mood
cognition
bradykinesia
motor
neurological disorders
used human activity recognition (HAR
unlabeled data
unconstrained
Verily Study Watches
inertial sensor data
unsupervised
wearable sensors
commonplace
human activity recognition
commercialized devices
Fitbit
Apple Watch
Verily Study Watch
algorithm
a deep convolutional neural network (DCNN)
extreme classification
NIH chest X-ray dataset (ChestX-ray8)
preprocessing
an EfficientNetV2-S backbone
classification with deep metric learning
evaluated
two public datasets
two clinical chest X-ray image datasets
extractor
characteristic
the PadChest dataset
malpractice
human errors
convolutional neural network
classify
monitor breathing
classification method
pre-activation residual network (Pre-ResNet)
two-dimensional (2D) images
one-dimensional
classify psychological
a depth camera
storage
computational complexity
Emergency Departments (ED
A QRS detection algorithm
ECG strips
annotation
grayscale images
ONNX (Open Neural Network Exchange Format
TensorRT engine
artificial neural network (ANN)
binary models
multilayer perceptrons
one-dimensional convolutional neural networks (1D-CNNs
recurrent neural network (RNN
long short-term memory (LSTM) model
logistic regression (LR
extra tree classifier (ETC
stochastic gradient descent (SGD)
random forest (RF
Gaussian Naive Bayes (GNB
Rain optimization algorithm (ROA)
the YOLOv4 convolutional neural network
optical flow (OF) algorithm
Cardiovascular disease (CVD)
heart
blood
Coronary Heart Disease
Cardiac Valvular Diseases
Pericardial Diseases
Cardiac Muscular Diseases
Congenital Heart Diseases
heart attack
angina
abnormal rhythms (Cardiac Arrhythmias
coronary artery disease
recording electrophysiological signals
electrocardiogram (ECG
photoplethysmogram (PPG)
as time series signals
electrical activities
classifiers
vector machines
logistic regression
K nearest neighbor
Naive Bayes
decision tree
random forest
ECG Lead II signals
arrhythmia database
PhysioNet
segmented into heartbeats
ANSI/AAMI EC57 standard
normal pulse (N)
supraventricular ectopic pulse (S
ventricular ectopic pulse (V
a fusion of ventricular and normal pulse (F
paced pulse
a fusion of paced and normal pulse
pulse that cannot be classified
Cardiovascular disease prediction
electrocardiograms
K-fold cross-validation
assistive technology
wearable devices
smart footwear
detecting walking behaviour
increase in the proportion of elderly
custodial care
scalable
easily modulated
detect different user behaviours/events
sitting
standing without imbalance
standing with imbalance
walking
running
tripping
sensor measurements
pressure sensors
temperature sensor
a hierarchical structure
ageing population
falling fertility rates
increasing life expectancy
advances in science
technology
medicine and public health
nutrition
physical
mental
sensory impairments
smart home systems
(Internet of Things, IoT)
concerns
privacy
ease of use
design
aesthetics
unobtrusiveness of the device
integration of sensors
pre-processing
analysing
miniaturisation of sensors
data storage
transmission systems
AT prototype
incorporation of sensors
clothing
furniture
footwear
older adults easier
more independent
fashionable
comfortable
smart footwear prototype
clinical images
Biological fingerprints
picture archiving
communication systems
methods
clinical
performance
variability
improve the performance
A novel method
automatically identify
patients
arrhythmia
ECG
continuous monitoring; edge computing; deep learning
cardiac arrhythmias
preventing
complications
stroke
diagnostic process
cardiologists
hospital emergency departments
privacy
portability
constant operation
MIT- BIH arrhythmia database
continuous real-time (RT) monitoring scenarios
camera-based vital sign measurement
infrared thermography
camera-based skin temperature
object detection
ICU monitoring
medical practice
record data
intensive care unit
the most vital hospital wards
critical health conditions
basic parameters monitored
body temperature (BT
hypo
hyperthermia
electrodes
medical knowledge for operation
Biometrics
X-ray image
posteroanterior (PA)
scout images
anteroposterior (AP)
chest X-ray images
Patient identification
misidentification
healthcare industry
radiography
hospital
adverse events databases
relating to wrong side or site
wrong procedure
wrong patient occurred annually
operating room
the primary risk of patient misidentification
clinical X-ray
radiation exposure
serious incidents
accidents
operating
Risk management
clinics
standard process
forensic
potential risk
Patient verification
patient-specific information
full name
date of birth
two patient identifiers
reliability of patient identification
human errors
resulting in misfiling
misidentificatio
minimize the risk of human error
improve healthcare workflow
utility
fingerprints and face recognition
workload of healthcare providers
variabilities
view positions
image processing
patient conditions
regarding view positions
healthy subjects
domain-shift problem
imaging geometry
similarity index
Lung involvement diagnosis
intelligent lung tumor diagnostic system
image pro- cessing techniques
Lung CT images
gray-level
co-occurrence matrix technique
whale opti- mization algorithm (WOA
radial bias function (RBF
mobile health application
mobile sensor data
obesity
body mass index (BMI)
weight
height
the movements of the ankle joint speed
peak extensor knee moments
motion sensors of a smartphone
mHealth
monitor individual health
the National Heart, Lung, and Blood Institute
heart disease
type 2 diabetes (high blood sugar)
high blood pressure
died
children with obesity
obese adults
to influence an individual’s manner of walking
joint and walking velocities
knee joint loads
effectiveness
efficiency
human balance
sensorimotor
musculoskeletal systems
control vision
proprioception
vestibular function
muscle contraction
diagnose disorders
diseases related to the nervous system
ataxia
cognitive deficits
Parkinson’s disease
vision problems
Alzheimer
The walking balance
good method
visual
vestibular
experienced a stroke
lower limb
back injury
the health-belief model
Information theory
information
communication
Artificial Intelligence
Big Data
entropy
great contributions
respiratory disease
COVID-19
the global coronavirus pandemic
WHO
artery disease
worldwide pandemic
lockdown
limited medical personnel
highly contagious infectious
increased risk
healthcare providers
efficient treatments
devices
thermometers
pulse oximeters
smartwatches
rings
monitor the symptoms
automated monitoring
infected individuals
prevent
infectious disease
, chest radiography
blood
sputum tests
breathing problems
heart infections
VGG16 architecture
principal component analysis (PCA
unified framework
as the modest appearance
moderate
severe instances
interpretability
high- and low-dimen- sional sample size (HDLSS) data
class imbalance
chest hierarchical segmentation
data augmentation
multitask multimodal SVM (MM-SVM
medical staff
personal protective equipment donning detection
Personal protective equipment (PPE)
headcover
medical goggles
inner medical mask
outer medical mask
medical protective clothing
medical protective gloves
medical protective foot covers
4G network transmission
safety guarantee
datasets
under the epidemic
health care personnel
self-protection safety
infectious respiratory diseases
microorganisms
viruses
fungi
parasites
bacteria
tuberculosis
diphtheria
bacterial pneumonia
viral pneumonia
influenza
methods
radar
CT scans
X-rays
ultrasounds
high costs
regarding accessibility
patient movement
data acquisition
near-infrared spectroscopy (NIRS)
simple
low-cost
tissue hemodynamic
oxygenation responses
hemoglobin
microvasculature
hypoxia
pulse oximetry
oxygenated hemoglobin (O2 Hb
deoxygenated hemoglobin (HHb
total hemoglobin (THb
saturation index (TSI)
O2 Hb signals
O2 Hb change
breathing interval
breathing depth
sensors
biological signals
poor circulation
skin thickness
skin color
monitor breathing patterns
real- time
hemodynamic responses
sternal manubrium
near-infrared spectroscopy (NIRS)
Common features
cough
rapid and shallow breathing
fever
headache
fatigue
lethargy
machine learning model
(Random Forest classification)
baseline (rest
rapid/shallow
loaded breathing
three breathing patterns
feature selection
decision tree construction
Deep metric learning
lesion detection
estimation of individual attributes
gender
age
clinical applications
Digital transformation
technologies
streamline operations
patient safety
edge computing
Ant colony optimization (ACO)
extreme learning machine (ELM
online sequential ELM
support vector machine (SVM
intelligent monitoring