Datasets:

ziq
/

RSNA-ATD2023

patient_id int64 1.52k 65.5k	series_id int64 137 64.5k	frame_id int64 0 1.04k	bowel int16 0 1	aortic_hu int16 87 736	bowel_healthy int16 0 1	extravasation_injury int16 0 1	kidney_low int16 0 1	liver_healthy int16 0 1	spleen_high int16 0 1	any_injury int16 1 1
10,004	21,057	0	1	146	1	1	1	1	1	1
10,004	21,057	1	1	146	1	1	1	1	1	1
10,004	21,057	10	1	146	1	1	1	1	1	1
10,004	21,057	100	1	146	1	1	1	1	1	1
10,004	21,057	1,000	0	146	1	1	1	1	1	1
10,004	21,057	1,001	0	146	1	1	1	1	1	1
10,004	21,057	1,002	0	146	1	1	1	1	1	1
10,004	21,057	1,003	0	146	1	1	1	1	1	1
10,004	21,057	1,004	0	146	1	1	1	1	1	1
10,004	21,057	1,005	0	146	1	1	1	1	1	1
10,004	21,057	1,006	0	146	1	1	1	1	1	1
10,004	21,057	1,007	0	146	1	1	1	1	1	1
10,004	21,057	1,008	0	146	1	1	1	1	1	1
10,004	21,057	1,009	0	146	1	1	1	1	1	1
10,004	21,057	101	1	146	1	1	1	1	1	1
10,004	21,057	1,010	0	146	1	1	1	1	1	1
10,004	21,057	1,011	0	146	1	1	1	1	1	1
10,004	21,057	1,012	0	146	1	1	1	1	1	1
10,004	21,057	1,013	0	146	1	1	1	1	1	1
10,004	21,057	1,014	0	146	1	1	1	1	1	1
10,004	21,057	1,015	0	146	1	1	1	1	1	1
10,004	21,057	1,016	0	146	1	1	1	1	1	1
10,004	21,057	1,017	0	146	1	1	1	1	1	1
10,004	21,057	1,018	0	146	1	1	1	1	1	1
10,004	21,057	1,019	0	146	1	1	1	1	1	1
10,004	21,057	102	1	146	1	1	1	1	1	1
10,004	21,057	1,020	0	146	1	1	1	1	1	1
10,004	21,057	1,021	0	146	1	1	1	1	1	1
10,004	21,057	103	1	146	1	1	1	1	1	1
10,004	21,057	104	1	146	1	1	1	1	1	1
10,004	21,057	105	1	146	1	1	1	1	1	1
10,004	21,057	106	1	146	1	1	1	1	1	1
10,004	21,057	107	1	146	1	1	1	1	1	1
10,004	21,057	108	1	146	1	1	1	1	1	1
10,004	21,057	109	1	146	1	1	1	1	1	1
10,004	21,057	11	1	146	1	1	1	1	1	1
10,004	21,057	110	1	146	1	1	1	1	1	1
10,004	21,057	111	1	146	1	1	1	1	1	1
10,004	21,057	112	1	146	1	1	1	1	1	1
10,004	21,057	113	1	146	1	1	1	1	1	1
10,004	21,057	114	1	146	1	1	1	1	1	1
10,004	21,057	115	1	146	1	1	1	1	1	1
10,004	21,057	116	1	146	1	1	1	1	1	1
10,004	21,057	117	1	146	1	1	1	1	1	1
10,004	21,057	118	1	146	1	1	1	1	1	1
10,004	21,057	119	1	146	1	1	1	1	1	1
10,004	21,057	12	1	146	1	1	1	1	1	1
10,004	21,057	120	1	146	1	1	1	1	1	1
10,004	21,057	121	1	146	1	1	1	1	1	1
10,004	21,057	122	1	146	1	1	1	1	1	1
10,004	21,057	123	1	146	1	1	1	1	1	1
10,004	21,057	124	1	146	1	1	1	1	1	1
10,004	21,057	125	1	146	1	1	1	1	1	1
10,004	21,057	126	1	146	1	1	1	1	1	1
10,004	21,057	127	1	146	1	1	1	1	1	1
10,004	21,057	128	1	146	1	1	1	1	1	1
10,004	21,057	129	1	146	1	1	1	1	1	1
10,004	21,057	13	1	146	1	1	1	1	1	1
10,004	21,057	130	1	146	1	1	1	1	1	1
10,004	21,057	131	1	146	1	1	1	1	1	1
10,004	21,057	132	1	146	1	1	1	1	1	1
10,004	21,057	133	1	146	1	1	1	1	1	1
10,004	21,057	134	1	146	1	1	1	1	1	1
10,004	21,057	135	1	146	1	1	1	1	1	1
10,004	21,057	136	1	146	1	1	1	1	1	1
10,004	21,057	137	1	146	1	1	1	1	1	1
10,004	21,057	138	1	146	1	1	1	1	1	1
10,004	21,057	139	1	146	1	1	1	1	1	1
10,004	21,057	14	1	146	1	1	1	1	1	1
10,004	21,057	140	1	146	1	1	1	1	1	1
10,004	21,057	141	1	146	1	1	1	1	1	1
10,004	21,057	142	1	146	1	1	1	1	1	1
10,004	21,057	143	1	146	1	1	1	1	1	1
10,004	21,057	144	1	146	1	1	1	1	1	1
10,004	21,057	145	1	146	1	1	1	1	1	1
10,004	21,057	146	1	146	1	1	1	1	1	1
10,004	21,057	147	1	146	1	1	1	1	1	1
10,004	21,057	148	1	146	1	1	1	1	1	1
10,004	21,057	149	1	146	1	1	1	1	1	1
10,004	21,057	15	1	146	1	1	1	1	1	1
10,004	21,057	150	1	146	1	1	1	1	1	1
10,004	21,057	151	1	146	1	1	1	1	1	1
10,004	21,057	152	1	146	1	1	1	1	1	1
10,004	21,057	153	1	146	1	1	1	1	1	1
10,004	21,057	154	1	146	1	1	1	1	1	1
10,004	21,057	155	1	146	1	1	1	1	1	1
10,004	21,057	156	1	146	1	1	1	1	1	1
10,004	21,057	157	1	146	1	1	1	1	1	1
10,004	21,057	158	1	146	1	1	1	1	1	1
10,004	21,057	159	1	146	1	1	1	1	1	1
10,004	21,057	16	1	146	1	1	1	1	1	1
10,004	21,057	160	1	146	1	1	1	1	1	1
10,004	21,057	161	1	146	1	1	1	1	1	1
10,004	21,057	162	1	146	1	1	1	1	1	1
10,004	21,057	163	1	146	1	1	1	1	1	1
10,004	21,057	164	1	146	1	1	1	1	1	1
10,004	21,057	165	1	146	1	1	1	1	1	1
10,004	21,057	166	1	146	1	1	1	1	1	1
10,004	21,057	167	1	146	1	1	1	1	1	1
10,004	21,057	168	1	146	1	1	1	1	1	1

End of preview. Expand in Data Studio

📁 Dataset

This dataset only comprised of 205 series of CT scans in .png file with raw images and raw mask.

Data source: Kaggle RSNA 2023 Abdominal Trauma Detection

🚀 Setup

pip install datasets

🤩 Feel the Magic

Load Dataset

from datasets import load_dataset

data = load_dataset('ziq/RSNA-ATD2023')
print(data)

DatasetDict({
    train: Dataset({
        features: ['patient_id', 'series_id', 'frame_id', 'image', 'mask'],
        num_rows: 70291
    })
})

Set Labels

labels = ["background", "liver", "spleen", "right_kidney", "left_kidney", "bowel"]

Train Test Split

data = data['train'].train_test_split(test_size=0.2)

train, test = data['train'], data['test']

# train[0]['patient_id']
# train[0]['image'] -> PIL Image
# train[0]['mask'] -> PIL Image

Get Image & Segmentation Mask

ids = 3
image, mask = train[ids]['image'], \ # shape: (512, 512)
                  train[ids]['mask']  # shape: (512, 512)

Convert mask into np.ndarray

mask = np.array(mask)

Visualize Image & Mask

fig = plt.figure(figsize=(16,16))
ax1 = fig.add_subplot(131)
plt.axis('off')
ax1.imshow(image, cmap='gray')
ax2 = fig.add_subplot(132)
plt.axis('off')
ax2.imshow(mask, cmap='gray')
ax3 = fig.add_subplot(133)
ax3.imshow(image*np.where(mask>0,1,0), cmap='gray')
plt.axis('off')
plt.show()

Write Custom Plotting Function

from matplotlib.colors import ListedColormap, BoundaryNorm

colors = ['#02020e', '#520e6d', '#c13a50', '#f57d15', '#fac62c', '#f4f88e'] # inferno
bounds = range(0, len(colors) + 1)

# Define the boundaries for each class in the colormap
cmap, norm = ListedColormap(colors), BoundaryNorm(bounds, len(colors))

# Plot the segmentation mask with the custom colormap
def plot_mask(mask, alpha=1.0):
    _, ax = plt.subplots()
    cax = ax.imshow(mask, cmap=cmap, norm=norm, alpha=alpha)
    cbar = plt.colorbar(cax, cmap=cmap, norm=norm, boundaries=bounds, ticks=bounds)
    cbar.set_ticks([])
    _labels = [""] + labels
    for i in range(1, len(_labels)):
        cbar.ax.text(2, -0.5 + i, _labels[i], ha='left', color=colors[i - 1], fontsize=8)
    plt.axis('off')
    plt.show()

Custom Color

plot_mask(mask)

Plot only one class (e.g. liver)

liver, spleen, right_kidney, left_kidney, bowel = [(mask == i,1,0)[0] * i for i in range(1, len(labels))]
plot_mask(liver)

Downloads last month: 29

Size of downloaded dataset files:

3.95 GB

Size of the auto-converted Parquet files:

3.99 GB

Number of rows:

70,291