ci-ber commited on
Commit
d28872d
·
1 Parent(s): d4ffb74

submitted versions

Browse files
Files changed (2) hide show
  1. croissant.json +46 -54
  2. data/nova-v1.parquet +2 -2
croissant.json CHANGED
@@ -3,7 +3,7 @@
3
  "@language": "en",
4
  "@vocab": "https://schema.org/",
5
  "cr": "http://mlcommons.org/croissant/",
6
- "sc": "https://schema.org/",
7
  "recordSet": "cr:recordSet",
8
  "field": "cr:field",
9
  "fileSet": "cr:fileSet",
@@ -11,35 +11,32 @@
11
  "dataType": { "@id": "cr:dataType", "@type": "@vocab" },
12
  "source": "cr:source",
13
  "extract": "cr:extract",
14
- "transform": "cr:transform",
15
  "subField": "cr:subField",
16
  "repeated": "cr:repeated",
17
  "key": "cr:key",
 
18
  "column": "cr:column",
19
- "format": "cr:format",
20
- "jsonPath": "cr:jsonPath"
21
  },
22
- "@type": ["Dataset", "sc:Dataset"],
23
  "name": "NOVA",
24
- "description": "Annotated brain MRI cases with radiology captions, gold + multi-rater bounding boxes (as JSON), and clinical metadata (as JSON).",
25
  "url": "https://huggingface.co/datasets/c-i-ber/Nova",
26
  "license": "https://creativecommons.org/licenses/by/4.0/",
27
  "version": "1.0.0",
 
28
  "conformsTo": "http://mlcommons.org/croissant/1.1",
29
- "sc:citation": "Bercea C. et al., NOVA: A Benchmark for Anomaly Localization and Clinical Reasoning in Brain MRI, 2025.",
30
- "sc:datePublished": "2025-10-23",
31
 
32
- "distribution": [
33
- {
34
- "@type": [
35
- "https://schema.org/FileObject"
36
- ],
37
- "@id": "parquet",
38
- "name": "novav1parquet",
39
- "contentUrl": "https://huggingface.co/datasets/c-i-ber/Nova/tree/main/data/nova-v1.parquet",
40
- "encodingFormat": "application/x-parquet",
41
- "sha256": "1fd1b2fd3a4f4ed4a2aa670ed13baaaec6e32b8b291c401b9d5df118a8aa7ba3"
42
- }
43
  ],
44
 
45
  "recordSet": [
@@ -54,82 +51,77 @@
54
  "@type": "cr:Field",
55
  "@id": "records/image_path",
56
  "dataType": "Text",
57
- "source": { "fileSet": { "@id": "parquet" }, "extract": { "column": "image_path" } }
58
  },
59
  {
60
  "@type": "cr:Field",
61
  "@id": "records/filename",
62
  "dataType": "Text",
63
- "source": { "fileSet": { "@id": "parquet" }, "extract": { "column": "filename" } }
64
  },
65
  {
66
  "@type": "cr:Field",
67
  "@id": "records/split",
68
  "dataType": "Text",
69
- "source": { "fileSet": { "@id": "parquet" }, "extract": { "column": "split" } }
70
  },
71
  {
72
  "@type": "cr:Field",
73
  "@id": "records/case_id",
74
  "dataType": "Text",
75
- "source": { "fileSet": { "@id": "parquet" }, "extract": { "column": "case_id" } }
76
  },
77
  {
78
  "@type": "cr:Field",
79
  "@id": "records/scan_id",
80
  "dataType": "Text",
81
- "source": { "fileSet": { "@id": "parquet" }, "extract": { "column": "scan_id" } }
82
  },
83
  {
84
  "@type": "cr:Field",
85
  "@id": "records/caption_text",
86
  "dataType": "Text",
87
- "source": { "fileSet": { "@id": "parquet" }, "extract": { "column": "caption_text" } }
 
 
 
 
 
 
 
88
  },
89
-
90
  {
91
  "@type": "cr:Field",
92
  "@id": "records/bboxes",
93
  "dataType": "StructuredValue",
94
- "description": "Bounding boxes stored as a JSON string in the 'bboxes' column.",
95
  "repeated": true,
96
- "source": {
97
- "fileSet": { "@id": "parquet" },
98
- "extract": { "column": "bboxes" },
99
- "format": "application/json",
100
- "transform": { "jsonPath": "$[*]" }
101
- },
102
  "subField": [
103
- { "@type": "cr:Field", "@id": "records/bboxes/x", "dataType": "cr:Float", "source": { "transform": { "jsonPath": "$.x" } } },
104
- { "@type": "cr:Field", "@id": "records/bboxes/y", "dataType": "cr:Float", "source": { "transform": { "jsonPath": "$.y" } } },
105
- { "@type": "cr:Field", "@id": "records/bboxes/width", "dataType": "cr:Float", "source": { "transform": { "jsonPath": "$.width" } } },
106
- { "@type": "cr:Field", "@id": "records/bboxes/height", "dataType": "cr:Float", "source": { "transform": { "jsonPath": "$.height" } } },
107
- { "@type": "cr:Field", "@id": "records/bboxes/source", "dataType": "Text", "source": { "transform": { "jsonPath": "$.source" } } },
108
- { "@type": "cr:Field", "@id": "records/bboxes/label", "dataType": "Text", "source": { "transform": { "jsonPath": "$.label" } } }
109
  ]
110
  },
111
-
112
  {
113
  "@type": "cr:Field",
114
  "@id": "records/meta",
115
  "dataType": "StructuredValue",
116
- "description": "Per-case metadata stored as a JSON string in the 'meta' column.",
117
- "source": {
118
- "fileSet": { "@id": "parquet" },
119
- "extract": { "column": "meta" },
120
- "format": "application/json",
121
- "transform": { "jsonPath": "$" }
122
- },
123
  "subField": [
124
- { "@type": "cr:Field", "@id": "records/meta/title", "dataType": "Text", "source": { "transform": { "jsonPath": "$.title" } } },
125
- { "@type": "cr:Field", "@id": "records/meta/publication_date", "dataType": "Text", "source": { "transform": { "jsonPath": "$.publication_date" } } },
126
- { "@type": "cr:Field", "@id": "records/meta/clinical_history", "dataType": "Text", "source": { "transform": { "jsonPath": "$.clinical_history" } } },
127
- { "@type": "cr:Field", "@id": "records/meta/differential_diagnosis", "dataType": "Text", "source": { "transform": { "jsonPath": "$.differential_diagnosis" } } },
128
- { "@type": "cr:Field", "@id": "records/meta/final_diagnosis", "dataType": "Text", "source": { "transform": { "jsonPath": "$.final_diagnosis" } } },
129
- { "@type": "cr:Field", "@id": "records/meta/link", "dataType": "Text", "source": { "transform": { "jsonPath": "$.link" } } }
130
  ]
131
  }
132
  ]
133
  }
134
  ]
135
- }
 
3
  "@language": "en",
4
  "@vocab": "https://schema.org/",
5
  "cr": "http://mlcommons.org/croissant/",
6
+ "dct": "http://purl.org/dc/terms/",
7
  "recordSet": "cr:recordSet",
8
  "field": "cr:field",
9
  "fileSet": "cr:fileSet",
 
11
  "dataType": { "@id": "cr:dataType", "@type": "@vocab" },
12
  "source": "cr:source",
13
  "extract": "cr:extract",
 
14
  "subField": "cr:subField",
15
  "repeated": "cr:repeated",
16
  "key": "cr:key",
17
+ "path": "cr:path",
18
  "column": "cr:column",
19
+ "conformsTo": "dct:conformsTo",
20
+ "citeAs": "cr:citeAs"
21
  },
22
+ "@type": ["Dataset"],
23
  "name": "NOVA",
24
+ "description": "Annotated brain MRI cases with radiology captions, gold + multi-rater bounding boxes, and clinical metadata.",
25
  "url": "https://huggingface.co/datasets/c-i-ber/Nova",
26
  "license": "https://creativecommons.org/licenses/by/4.0/",
27
  "version": "1.0.0",
28
+ "citeAs": "Bercea C. et al., NOVA: A Benchmark for Anomaly Localization and Clinical Reasoning in Brain MRI, 2025.",
29
  "conformsTo": "http://mlcommons.org/croissant/1.1",
 
 
30
 
31
+ "distribution": [
32
+ {
33
+ "@type": "cr:FileObject",
34
+ "@id": "local-parquet",
35
+ "name": "NOVA Parquet",
36
+ "contentUrl": "https://huggingface.co/datasets/c-i-ber/Nova/resolve/main/data/nova-v1.parquet",
37
+ "encodingFormat": "application/x-parquet",
38
+ "sha256": "71880c12b710b8c82100da51d574f13312cc4811fb5b20073ef4f5d98d22a319"
39
+ }
 
 
40
  ],
41
 
42
  "recordSet": [
 
51
  "@type": "cr:Field",
52
  "@id": "records/image_path",
53
  "dataType": "Text",
54
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "image_path" } }
55
  },
56
  {
57
  "@type": "cr:Field",
58
  "@id": "records/filename",
59
  "dataType": "Text",
60
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "filename" } }
61
  },
62
  {
63
  "@type": "cr:Field",
64
  "@id": "records/split",
65
  "dataType": "Text",
66
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "split" } }
67
  },
68
  {
69
  "@type": "cr:Field",
70
  "@id": "records/case_id",
71
  "dataType": "Text",
72
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "case_id" } }
73
  },
74
  {
75
  "@type": "cr:Field",
76
  "@id": "records/scan_id",
77
  "dataType": "Text",
78
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "scan_id" } }
79
  },
80
  {
81
  "@type": "cr:Field",
82
  "@id": "records/caption_text",
83
  "dataType": "Text",
84
+ "description": "Radiology caption.",
85
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "caption_text" } }
86
+ },
87
+ {
88
+ "@type": "cr:Field",
89
+ "@id": "records/image",
90
+ "dataType": "ImageObject",
91
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "image" } }
92
  },
 
93
  {
94
  "@type": "cr:Field",
95
  "@id": "records/bboxes",
96
  "dataType": "StructuredValue",
97
+ "description": "Bounding boxes in pixel coords with source (gold/raters).",
98
  "repeated": true,
99
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "bboxes" } },
 
 
 
 
 
100
  "subField": [
101
+ { "@type": "cr:Field", "@id": "records/bboxes/x", "dataType": "Float" },
102
+ { "@type": "cr:Field", "@id": "records/bboxes/y", "dataType": "Float" },
103
+ { "@type": "cr:Field", "@id": "records/bboxes/width", "dataType": "Float" },
104
+ { "@type": "cr:Field", "@id": "records/bboxes/height", "dataType": "Float" },
105
+ { "@type": "cr:Field", "@id": "records/bboxes/source", "dataType": "Text" },
106
+ { "@type": "cr:Field", "@id": "records/bboxes/label", "dataType": "Text" }
107
  ]
108
  },
 
109
  {
110
  "@type": "cr:Field",
111
  "@id": "records/meta",
112
  "dataType": "StructuredValue",
113
+ "description": "Per-case metadata (Eurorad).",
114
+ "source": { "fileSet": { "@id": "local-parquet" }, "extract": { "column": "meta" } },
 
 
 
 
 
115
  "subField": [
116
+ { "@type": "cr:Field", "@id": "records/meta/title", "dataType": "Text" },
117
+ { "@type": "cr:Field", "@id": "records/meta/publication_date", "dataType": "Text" },
118
+ { "@type": "cr:Field", "@id": "records/meta/clinical_history", "dataType": "Text" },
119
+ { "@type": "cr:Field", "@id": "records/meta/differential_diagnosis", "dataType": "Text" },
120
+ { "@type": "cr:Field", "@id": "records/meta/final_diagnosis", "dataType": "Text" },
121
+ { "@type": "cr:Field", "@id": "records/meta/link", "dataType": "URL" }
122
  ]
123
  }
124
  ]
125
  }
126
  ]
127
+ }
data/nova-v1.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1fd1b2fd3a4f4ed4a2aa670ed13baaaec6e32b8b291c401b9d5df118a8aa7ba3
3
- size 126910752
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71880c12b710b8c82100da51d574f13312cc4811fb5b20073ef4f5d98d22a319
3
+ size 199088