Deeplearning Applications In Medical Imaging

DeepLearning and Applications In Medical Imaging

In this post I’m going to show some applications of deeplerning in medical imaging and how shoulde get started in this field. topics:

 . some of  Call paperes in DeepLearning for medical imaging
 
 . deep learning applications for medical imaging
  
 . where i can find databases for medical imaging 

let’s start this!!!

### some of Call paperes in “DeepLearning for medical imaging” there are few call for paper in medical imaging and most popular toipics in them listed below :

 - theories for deep learning
 - deep learning for medical and health appliactions
 - visualization and understanding for deep learning 
 - optimizations for deep learning 
 - novel methodolugies using deeplearning for classification ,detection and segmentation in med-
   ical imaging 
 - deep learning for multimodal medical and health data(X-ray,MRI,CT,echo video ,time series da-
 	ta,text,etc)

as you can see the most topics in application of deep learning in medical imaging are about segmentation,detection,classification. CNN network is powerful tools for doing that jobs so in last part of this post i’m going create a CNN network for classification medical images.

### deep learning applications for medical imaging this document survey the use of deep learning for image classification, object detection, segmentation, registration, and other tasks. Concise overviews are provided of studies per application area: neuro, retinal, pulmonary, digital pathology, breast, cardiac, abdominal,musculoskeletal.

some methods of deep learning and examplesd in the survey listed below:

1- classification :

application DBNs and SAEs to classify patients and having Alzheimer’s disease based on brain MRI.
use mut-stream CNN to classify skin lesions.
use combination of CNNs and RNNs for grading nuclear cataracts in shit-lamp image.
use multi-stream CNN to classify ponts of interest in chest CT as nedule or non-nedule.

2- detection

identify land marks on the distal femur surface by processing three independent sets of 2D MRI slice(one for each plane ) with regular CNNs.
use a multi-stream CNN to integrate CT and positron emission tomography (PET) data.
use a 3D CNN to findmicro-bleedsin brain MRI.

3- segmentation

use aspatial clock work RNN to segment the perimysium in H&E histopathology example .
use a 3D RNN with gated recurrent units to segment gray and whitr matter in brain MRI data set.
combine bi-directional LST-RNNs with 2D unet-like_architectures to segment structurses in anisotropic 3D electron microscopy iamge.
applay pixel-wise segmentation of membranes in electron microscopy imagery in a sliding window fashion.
used 3D fCNNs to generate vertebral body like-lihood maps which drove deformable models for ver- tebral body segmentation in MR images.
In lesion segmentation we have also seen the application of U-net and similar architectures to leverage both this global and local context.
to segment white matter lesions brain brain MRI.

where i can find databases for medical imaging

the most important thing in deep learning for medical imaging is data;with enough data we can reach to reasonable result’s in training Nets or fine-tuning them .

the below list contain name and links for some popular medica image datas:

1. Medical Imaging Data

The National Library of Medicine presents MedPix®
Database of 53,000 medical images from 13,000 patients with annotations. Requires registration.
Information

ABIDE: The Autism Brain Imaging Data Exchange: towards a large-scale evaluation of the intrinsic brain architecture in autism.
Function MRI images for 539 individuals suffering from ASD and 573 typical controls. These 1112 datasets are composed of structural and resting state functional MRI data along with an extensive array of phenotypic information. Requires registration.
Paper Information Preprocessed version

Alzheimer’s Disease Neuroimaging Initiative (ADNI)
MRI database on Alzheimer’s patients and healthy controls. Also has clinical, genomic, and biomaker data. Requires registration.
Paper Access

Digital Retinal Images for Vessel Extraction (DRIVE)
The DRIVE database is for comparative studies on segmentation of blood vessels in retinal images. It consists of 40 photographs out of which 7 showing signs of mild early diabetic retinopathy.
Paper
Access

AMRG Cardiac Atlas The AMRG Cardiac MRI Atlas is a complete labelled MRI image set of a normal patient’s heart acquired with the Auckland MRI Research Group ‘s Siemens Avanto scanner. The atlas aims to provide university and school students, MR technologists, clinicians…

Congenital Heart Disease (CHD) Atlas The Congenital Heart Disease (CHD) Atlas represents MRI data sets, physiologic clinical data and computer models from adults and children with various congenital heart defects. The data have been acquired from several clinical centers including Rady…

DETERMINE Defibrillators to Reduce Risk by Magnetic Resonance Imaging Evaluation, is a prospective, multicenter, randomized clinical trials in patients with coronary artery diseases and mild-to-moderate left ventricular dysfunction. The primary objective…

MESA Multi-Ethnic Study of Atherosclerosis, is a large-scale cardiovascular population study (>6,500 participants) conducted in six centres in the USA. It aims to investigate the manifestation of subclinical to clinical cardiovascular disease before…

OASIS The Open Access Series of Imaging Studies (OASIS) is a project aimed at making MRI data sets of the brain freely available to the scientific community. Two datasets are available: a cross-sectional and a longitudinal set.

Cross-sectional MRI Data in Young, Middle Aged, Nondemented and Demented Older Adults: This set consists of a cross-sectional collection of 416 subjects aged 18 to 96. For each subject, 3 or 4 individual T1-weighted MRI scans obtained in single scan sessions are included. The subjects are all right-handed and include both men and women. 100 of the included subjects over the age of 60 have been clinically diagnosed with very mild to moderate Alzheimer’s disease (AD). Additionally, a reliability data set is included containing 20 nondemented subjects imaged on a subsequent visit within 90 days of their initial session.
Longitudinal MRI Data in Nondemented and Demented Older Adults: This set consists of a longitudinal collection of 150 subjects aged 60 to 96. Each subject was scanned on two or more visits, separated by at least one year for a total of 373 imaging sessions. For each subject, 3 or 4 individual T1-weighted MRI scans obtained in single scan sessions are included. The subjects are all right-handed and include both men and women. 72 of the subjects were characterized as nondemented throughout the study. 64 of the included subjects were characterized as demented at the time of their initial visits and remained so for subsequent scans, including 51 individuals with mild to moderate Alzheimer’s disease. Another 14 subjects were characterized as nondemented at the time of their initial visit and were subsequently characterized as demented at a later visit.

Deeplearning Applications In Medical Imaging

DeepLearning and Applications In Medical Imaging

where i can find databases for medical imaging

1. Medical Imaging Data

2. Challenges/Contest Data

3. Data derived from Electronic Health Records (EHRs)

4. National Healthcare Data

5. UCI Datasets

6. Biomedical Literature

6. TREC Precision Medicine / Clinical Decision Support Track