Generalizable Retinal Disease Detection

₹600-1500 INR

Sürüyor

İlan edilme: 3 ay önce

₹600-1500 INR

Teslimde ödenir

I need a deep-learning foundation model that can reliably spot Diabetic Retinopathy and Glaucoma in fundus photographs, even when the images come from clinics or cameras it has never seen before. I will rely on publicly available datasets, so part of the job is to help me choose, download, and harmonise the best open-source collections, handle class imbalance, and set up robust cross-dataset evaluation. Once the data pipeline is in place, build and train a modern architecture—CNN, Vision Transformer or a hybrid—optimised for both accuracy and domain generalisation. Strong data-augmentation, colour normalisation and adversarial or contrastive techniques are welcome if they improve out-of-distribution performance. Deliverables • Cleaned and documented dataset splits with the code to reproduce them • Training code (Python, PyTorch or TensorFlow) with clear README • Pre-trained model weights and an inference script that runs from the command line and outputs disease probabilities for each eye image • Evaluation report comparing performance on at least two separate public datasets and a small unseen hold-out set • Brief note on future extensibility to Macular Degeneration I will test the model on my own images; acceptance depends on ≥ 85 % AUC for each disease and no major drop when the source dataset changes.

Deep Learning

Computer Vision

Convolutional Neural Network

Data Augmentation

Python

Machine Learning (ML)

Data Mining

Statistical Analysis

Proje No: 40077283

Proje hakkında

4 teklifler

Uzaktan proje

Son aktiviteden bu yana geçen zaman 3 ay önce

Biraz para mı kazanmak istiyorsunuz?

E-posta adresi

Freelancer'da teklif vermenin faydaları

Bütçenizi ve zaman çerçevenizi belirleyin

Çalışmanız için ödeme alın

Teklifinizin ana hatlarını belirleyin

Kaydolmak ve işlere teklif vermek ücretsizdir

Seçilen:

@vishnunetaran

Hi — I’d love to work on this. You’re looking for more than a classifier; you want a clinically reliable foundation model for Diabetic Retinopathy + Glaucoma that generalises across datasets, devices and clinics. That’s exactly the kind of applied deep learning work I specialize in. My Approach • Help shortlist and harmonise top public datasets (EyePACS / Messidor / APTOS for DR, REFUGE / RIM-ONE / DRISHTI for Glaucoma) • Clean + standardise images, handle class imbalance, analyse bias • Build strong baselines (EfficientNet / ResNeXt) and benchmark against ViT / hybrid models • Use robust augmentation, colour normalisation, weighted sampling, mixup/cutmix; consider domain-generalisation strategies if needed • Cross-dataset evaluation + unseen hold-out test Deliverables • Clean reproducible dataset splits + code • Training script / notebook with README • Trained model weights + CLI inference script • Report with ROC-AUC, confusion matrix, generalisation analysis • Notes on extending to AMD • Target ≥ 85% AUC with minimal drop across datasets I can collaborate closely, iterate quickly and ensure the model performs in realistic settings. Ready to start immediately.

₹800 INR 1 gün içinde

0,0

(0 değerlendirme)

0,0

4 freelancer bu proje için ortalama ₹775 INR teklif veriyor

@PatoPro

As a seasoned Data Scientist with a focus on real-world applications of Machine Learning, I am excited about the possibility of collaborating with you to solve this critical health issue. My extensive experience in data mining and ML gives me a distinct advantage in managing complex datasets and implementing robust solutions. I've also successfully handled similar projects involving classification and generalization tasks, and thus understand the challenges involved. My grasp of Python, TensorFlow, and PyTorch will ensure I build and train an effective, optimised model for your unique diagnostic needs. Furthermore, I am well-versed in handling imbalanced data to ensure comprehensive understanding and detection. My adeptness with advanced techniques like data augmentation, domain adaptation, and adversarial training is particularly valuable for tackling the out-of-distribution performance requirement. I propose a meticulously documented pipeline that will allow us to harmonize the best publicly available datasets, evaluate their performance on various disease types including the unseen Macular Degeneration. I don't just intend to meet your expectations but to exceed them consistently. Thank you for considering my candidacy; I look forward to discussing how we can revolutionize retinal disease detection together!

₹1.050 INR 2 gün içinde

5,0

(1 değerlendirme)

0,1

@m0h1t12

I saw you are looking for someone to handle. Because I am building my profile here, I am willing to go the extra mile to ensure you are 100% satisfied and earn a 5-star review. You will get my full attention and unlimited revisions until the work is perfect. I can have a sample ready for you .

₹600 INR 1 gün içinde

0,0

(0 değerlendirme)

0,0

@patelaryan77

Hello, I have hands-on experience building and deploying medical imaging and predictive machine learning systems, with a strong focus on robustness and cross-dataset generalisation. I understand that the main challenge here is maintaining performance when images come from unseen clinics or cameras, not just achieving high accuracy on one dataset. I will start by selecting and harmonising high-quality public fundus datasets for Diabetic Retinopathy and Glaucoma, applying consistent preprocessing, colour normalisation, and clinically meaningful dataset splits. Class imbalance will be handled using stratified sampling, loss weighting, and targeted data augmentation. For modelling, I will work with modern architectures such as EfficientNet and Vision Transformer based hybrids, combined with strong augmentation and domain generalisation techniques to reduce performance drop on unseen data. Cross-dataset evaluation will be built into the pipeline from the beginning. Deliverables will include reproducible dataset splits, clean and well-documented training code, pretrained weights, a command-line inference script outputting disease probabilities, and a concise evaluation report comparing performance across datasets and a held-out unseen set. Happy to clarify any details before starting.

₹650 INR 1 gün içinde