File size: 1,772 Bytes
c20d956
 
 
 
 
 
 
 
 
 
 
 
 
 
 
0fb3e9d
c20d956
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c246f4d
c20d956
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
{
    "model_name": "DenseNet3DSmooth",
    "preprocessor_name": null,
    "model_params": {
        "in_channels_ct": 1,
        "in_channels_dose": 1,
        "in_channels_mask": 1,
        "drop_rate": 0.3,
        "use_skip_connections": true,
        "arch": [4, 4, 4],
        "stem_out_channels_total": 96,
        "growth_rate": 32,
        "batch_norm_dose": true
    },
    "model_preprocessing": {
        "type_order": ["dose", "mask", "ct"],
        "modality": {
            "dose": {
                "extract": false,
                "normalization_value": 70.0,
                "description": "Physical dose, normalized by the prescription."
            },
            "ct": {
                "window": [-1000, 1000],
                "description": "CT in Hounsfield units, windowed to [-1000, 1000]."
            },
            "mask": {
                "collapse": true,
                "description": "Binary structure mask (collapsed to a single channel)."
            }
        },
        "input_dimensions": [100, 100, 40],
        "input_spacing": [3.0, 3.0, 2.5],
        "center": "target"
    },
    "metadata": {
        "description": "DenseNet3DSmooth NTCP outcome model for the TG119 proton phantom.",
        "training": {
            "dataset": "tg119",
            "modality": "protons",
            "input_resolution_mm": [3.0, 3.0, 2.5],
            "notes": "Dose-preserving DenseNet3D (BatchNorm, anti-aliased pooling); dose normalized by 70 Gy prescription."
        },
        "model_input": {
            "type_order": ["dose", "mask", "ct"],
            "tensor_shapes": {
                "dose": [1, 100, 100, 40],
                "ct": [1, 100, 100, 40],
                "mask": [1, 100, 100, 40]
            }
        }
    }
}