You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

TripleNetFusion — OCT Multiclass Classification

Feature-fusion ensemble of Xception, EfficientNetV2S, and DenseNet121 for multiclass classification of OCT (optical coherence tomography) images.

This is a research/educational model. It is NOT a diagnostic tool and must not be used for clinical decision-making.

Architecture

Backbone 1: Xception (2048-d features)
Backbone 2: EfficientNetV2S (1280-d features)
Backbone 3: DenseNet121 (1024-d features)
Fusion classifier: concatenated 4352-d -> Dropout -> Linear(1024) -> BatchNorm -> ReLU -> Dropout -> Linear(num_classes)

Classes

AMD
CNV
CSR
DME
DR
DRUSEN
ERM
MH
NO
NORMAL
RVO

Usage

from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
import torch
from modeling import TripleNetFusion

path = hf_hub_download(repo_id="adityadebnathtirtha/oct-triple-fusion", filename="model.safetensors")
state_dict = load_file(path)

model = TripleNetFusion(num_classes=11)
model.load_state_dict(state_dict)
model.eval()

Input images should be resized to 224x224 and normalized with mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225].

Downloads last month: -

Safetensors

Model size

52.7M params

Tensor type

F32