Hybrid Deep Learning for OCT
Budget: ₹1,500 – ₹12,500 INR
I have a complete OCT (optical coherence tomography) image set and now need to turn it into a full, publication-ready study on age-related macular degeneration (AMD). The work has to run entirely in Google Colab and revolve around a hybrid deep-learning architecture of your choice—CNN + transformer, ensemble CNNs, or any comparable combination—as long as it meets strong SCI journal standards.
Pre-processing
Both FFT and Wavelet Transform must be applied. Please document each step in the notebook so the signal-processing pipeline is clear and reproducible.
Core modelling
• Train, validate and test the model on the OCT data.
• Track and store all metrics so they can be plotted later.
• Incorporate Explainable AI focused on feature-importance visualisation (e.g., Grad-CAM, SHAP, or an equivalent method) to highlight the retinal regions that drive the model’s AMD predictions.
Graphs required for the paper
ROC Curve, Confusion Matrix and an Accuracy-vs-Epochs plot are mandatory. Add any other standard figures—loss curves, Grad-CAM heat-maps, etc.—that strengthen the results section.
Manuscript
Deliver a 14-page SCI-style paper (single column, standard fonts) covering introduction, methods, results, discussion and references. Plagiarism must remain below 5 %, with no detectable AI-generated text. Cite recent ophthalmology and deep-learning literature to support the methodology.
Deliverables and acceptance criteria
1. Google Colab notebook with runnable code, comments and section headings.
2. High-resolution PNG/SVG copies of every figure used in the manuscript.
3. The 14-page manuscript in .docx and PDF formats, passing Turnitin (<5 %) and any AI-detection scan (0 %).
4. A short README explaining how to reproduce the results without modification.
I will review the notebook’s reproducibility, the clarity of the feature-importance explanation and the plagiarism reports before releasing the final milestone.
Pre-processing
Both FFT and Wavelet Transform must be applied. Please document each step in the notebook so the signal-processing pipeline is clear and reproducible.
Core modelling
• Train, validate and test the model on the OCT data.
• Track and store all metrics so they can be plotted later.
• Incorporate Explainable AI focused on feature-importance visualisation (e.g., Grad-CAM, SHAP, or an equivalent method) to highlight the retinal regions that drive the model’s AMD predictions.
Graphs required for the paper
ROC Curve, Confusion Matrix and an Accuracy-vs-Epochs plot are mandatory. Add any other standard figures—loss curves, Grad-CAM heat-maps, etc.—that strengthen the results section.
Manuscript
Deliver a 14-page SCI-style paper (single column, standard fonts) covering introduction, methods, results, discussion and references. Plagiarism must remain below 5 %, with no detectable AI-generated text. Cite recent ophthalmology and deep-learning literature to support the methodology.
Deliverables and acceptance criteria
1. Google Colab notebook with runnable code, comments and section headings.
2. High-resolution PNG/SVG copies of every figure used in the manuscript.
3. The 14-page manuscript in .docx and PDF formats, passing Turnitin (<5 %) and any AI-detection scan (0 %).
4. A short README explaining how to reproduce the results without modification.
I will review the notebook’s reproducibility, the clarity of the feature-importance explanation and the plagiarism reports before releasing the final milestone.