Scribd
Upload
14 views2 pages

New 1

The document outlines a Python project titled 'Retinal Image Analysis for multi-Disease Detection' aimed at addressing preventable visual impairments through automated multi-disease detection models. It details a pipeline utilizing ensemble learning and various deep learning techniques for analyzing retinal images, including methods for class imbalance and real-time image augmentation. The project leverages the AUCMEDI framework for efficient medical image classification and supports a wide range of retinal diseases, enhancing diagnostic capabilities.

Uploaded by

Meenachi Sundaram
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
14 views2 pages

New 1

The document outlines a Python project titled 'Retinal Image Analysis for multi-Disease Detection' aimed at addressing preventable visual impairments through automated multi-disease detection models. It details a pipeline utilizing ensemble learning and various deep learning techniques for analyzing retinal images, including methods for class imbalance and real-time image augmentation. The project leverages the AUCMEDI framework for efficient medical image classification and supports a wide range of retinal diseases, enhancing diagnostic capabilities.

Uploaded by

Meenachi Sundaram
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 2

I have developed a python project titles

"Retinal Image Analysis for multi-Disease Detection" using Python 3.11.

The details are


"
Preventable or undiagnosed visual impairment and blindness affects billion of
people worldwide. Automated multi-disease detection models offer great potential to
address this problem via clinical decision support in diagnosis. In this work, we
proposed an innovative multi-disease detection pipeline for retinal imaging which
utilizes ensemble learning to combine the predictive power of several heterogeneous
deep convolutional neural network models. Our pipeline includes state-of-the-art
strategies like transfer learning, class weighting, real-time image augmentation
and focal loss utilization. Furthermore, we integrated ensemble learning techniques
like heterogeneous deep learning models, bagging via 5-fold cross-validation and
stacked logistic regression models.

# Train detector models


python scripts/detector_DenseNet201.py
python scripts/detector_EfficientNetB4.py

# Train classifier models


python scripts/classifier_DenseNet201.py
python scripts/classifier_InceptionV3.py
python scripts/classifier_ResNet152.py
python scripts/classifier_EfficientNetB4.py

# Train Logistic Regression models


python scripts/ensemble_training.py

# Run Inference
python scripts/inference.py
python scripts/ensemble.py

# Perform result evaluation


python scripts/evaluation.py

The open-source software AUCMEDI allows fast setup of medical image classification
pipelines with state-of-the-art methods via an intuitive, high-level Python API or
via an AutoML deployment through Docker/CLI.

This pipeline was based on AUCMEDI, which is an in-house developed open-source


framework to setup complete medical image classification pipelines with deep
learning models on top of Tensorflow/Keras. The framework supports extensive
preprocessing, image augmentation, class imbalance strategies, state-of-the-art
deep learning models and ensemble learning techniques.

This dataset consists of diseases/abnormalities (diabetic retinopathy (DR), age-


related macular degeneration (ARMD), media haze (MZ), drusen (DN), myopia (MYA),
branch retinal vein occlusion (BRVO), tessellation (TSLN), epiretinal membrane
(ERM), laser scar (LS), macular scar (MS), central serous retinopathy (CSR), optic
disc cupping (ODC), central retinal vein occlusion (CRVO), tortuous vessels (TV),
asteroid hyalosis (AH), optic disc pallor (ODP), optic disc edema (ODE), shunt
(ST), anterior ischemic optic neuropathy (AION), parafoveal telangiectasia (PT),
retinal traction (RT), retinitis (RS), chorioretinitis (CRS), exudation (EDN),
retinal pigment epithelium changes (RPEC), macular hole (MHL), retinitis pigmentosa
(RP), cotton wool spots (CWS), coloboma (CB), optic disc pit maculopathy (ODPM),
preretinal hemorrhage (PRH), myelinated nerve fibers (MNF), hemorrhagic retinopathy
(HR), central retinal artery occlusion (CRAO), tilted disc (TD), cystoid macular
edema (CME), post traumatic choroidal rupture (PTCR), choroidal folds (CF),
vitreous hemorrhage (VH), macroaneurysm (MCA), vasculitis (VS), branch retinal
artery occlusion (BRAO), plaque (PLQ), hemorrhagic pigment epithelial detachment
(HPED) and collateral (CL)) based on their visual characteristics.

## Methods

The implemented medical image classification pipeline can be summarized in the


following core steps:
- Class Weighted Focal Loss and Upsampling to conquer Class Imbalance
- Stratified Multi-label 5-fold Cross-Validation
- Extensive real-time image augmentation
- Multiple Deep Learning Model Training
- Distinct Training for Multi-Disease Labels and Disease Risk Detection
- Ensemble Learning Strategy: Bagging & Stacking
- Stacked Binary Logistic Regression Models for Distinct Classification
"

I need five para abstract "(Existing System - NBC and KNN, Proposed ABOVE METHODS"

You might also like