ParaTransCNN

This repo holds code for ParaTransCNN: Parallelized TransCNN Encoder for Medical Image Segmentation

• Overview

Requirements

We trained on NVIDIA RTX 3090, where python 3.9.10 and torch 1.12.1 on ubuntu 22.04.

We use the libraries of these versions:

• Python 3.9.10
• Torch 1.12.1+cu113
• torchvision 0.13.1+cu113
• numpy 1.21.5

You can pip the same experimental environment as us through requirements

pip install -r requirements.txt

Dataset preparation

• Synapse Dataset: please go to "./datasets/README.md" for the details about preparing preprocessed Synapse dataset or download the Synapse Dataset from here.
• AVT Dataset: please go to "./datasets/README.md" for the details about preparing preprocessed AVT dataset. The preprocessed dataset is here.

Train

• Run the following code to train ParaTransCNN on the Synapse Dataset:

python train.py --dataset Synapse --train_path <your path to Synapse train dataset> --model_name ParaTransCNN --max_epochs 150 --batch_size 4 --base_lr 0.01 

• Run the following code to train ParaTransCNN on the AVT Dataset:

python train.py --dataset AVT --train_path <your path to AVT train dataset> --model_name ParaTransCNN --max_epochs 150 --batch_size 4 --base_lr 0.01 

Test

• Run the following code to test the trained ParaTransCNN on the Synapse Dataset:

python test.py --dataset Synapse --volume_path <your path to Synapse test dataset> --model_name ParaTransCNN --max_epochs 150 --batch_size 4 --base_lr 0.01 

• Run the following code to test the trained ParaTransCNN on the AVT Dataset:

python test.py --dataset AVT --volume_path <your path to AVT test dataset> --model_name ParaTransCNN --max_epochs 150 --batch_size 4 --base_lr 0.01 

Results

• Synapse

Evaluation metrics

Methods	DSC(%)	HD	Aorta	Gallbladder	Kidney(L)	Kidney(R)	Liver	Pancreas	Spleen	Stomach
DARR	69.77	-	74.74	53.77	72.31	73.24	94.08	54.18	89.90	45.96
R50 U-Net	74.68	36.87	87.74	63.66	80.60	78.19	93.74	56.90	85.87	74.16
U-Net	76.85	39.70	89.07	69.72	77.77	68.60	93.43	53.98	86.67	75.58
R50 Att-UNet	75.57	36.97	55.92	63.91	79.20	72.71	93.56	49.37	87.19	74.95
Att-UNet	77.77	36.02	89.55	68.88	77.98	71.11	93.57	58.04	87.30	75.75
R50 ViT	71.29	32.87	73.73	55.13	75.80	72.20	91.51	45.99	81.99	73.95
TransUnet	77.48	31.69	87.23	63.13	81.87	77.02	94.08	55.86	85.08	75.62
SwinUnet	79.13	21.55	85.47	66.53	83.28	79.61	94.29	56.58	90.66	76.60
TransDeepLab	80.16	21.25	86.04	69.16	84.08	79.88	93.53	61.19	89.00	78.40
HiFormer	80.39	14.70	86.21	65.69	85.23	79.77	94.61	59.52	90.99	81.08
MISSFormer	81.96	18.20	86.99	68.65	85.21	82.00	94.41	65.67	91.92	80.81
TransCeption	82.24	20.89	87.60	71.82	86.23	80.29	95.01	65.27	91.68	80.02
DAE-Former	82.43	17.46	88.96	72.30	86.08	80.88	94.98	65.12	91.94	79.19
ParaTransCNN	83.86	15.86	88.12	68.97	87.99	83.84	95.01	69.79	92.71	84.43

Visualization on Synapse

• AVT

Visualization on AVT

• ACDC

Visualization on ACDC

• Kvasir_SEG

Evaluation metrics

Methods	Dice	Jaccard	Precision	Recall
U-Net	0.830530	0.748300	0.860328	0.858857
UNet++	0.795231	0.705255	0.825769	0.840401
Att-UNet	0.828564	0.748071	0.848016	0.863609
TransUnet	0.869120	0.799637	0.895035	0.886673
SwinUnet	0.854450	0.777262	0.890461	0.862594
TransDeepLab	0.859171	0.779644	0.881949	0.883266
HiFormer	0.859615	0.786705	0.879861	0.884120
MISSFormer	0.715535	0.611769	0.760764	0.765871
TransCeption	0.773330	0.676522	0.801368	0.813837
DAE-Former	0.779659	0.680100	0.806010	0.807978
ParaTransCNN	0.882230	0.819137	0.895940	0.900321

Visualization on Kvasir_SEG

• BUSI(benign & malignant)

Evaluation metrics

Methods	Dice	Jaccard	Precision	Recall
U-Net	0.779577	0.698415	0.795101	0.811817
UNet++	0.751396	0.665028	0.774793	0.780807
Att-UNet	0.784430	0.701824	0.820558	0.792074
TransUnet	0.791364	0.711470	0.810834	0.822213
SwinUnet	0.781966	0.694188	0.822793	0.782219
TransDeepLab	0.778597	0.693664	0.799186	0.791463
HiFormer	0.779438	0.699652	0.795937	0.803998
MISSFormer	0.731416	0.633783	0.766812	0.754716
TransCeption	0.758622	0.660874	0.803228	0.765918
DAE-Former	0.733205	0.634020	0.775104	0.745468
ParaTransCNN	0.809358	0.729952	0.850570	0.804386

Visualization on BUSI

Reference

• TransUNet
• ResNet