fernandogd commited on
Commit
706a4a8
·
verified ·
1 Parent(s): bf72c20

Release CardioBERTa-based clinical concept disambiguation model

Browse files
README.md ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ library_name: transformers
5
+ pipeline_tag: feature-extraction
6
+ base_model: DT4H/CardioBERTa.en
7
+ tags:
8
+ - biomedical
9
+ - clinical
10
+ - cardiology
11
+ - entity-linking
12
+ - concept-normalization
13
+ - umls
14
+ - metric-learning
15
+ ---
16
+
17
+ # DT4H_CardioBERTa_grandparents_en_enriched
18
+
19
+ `DT4H_CardioBERTa_grandparents_en_enriched` is a English biomedical terminology encoder for **clinical concept normalization and entity linking**. It is initialized from [`DT4H/CardioBERTa.en`] and specialized using CUI-supervised terminology pairs and metric learning.
20
+
21
+ ## Backbone
22
+
23
+ The backbone belongs to the **CardioBERTa** family from *CardioLM - a multilingual suite of small language models for the cardiology domain*. CardioBERTa comprises language-specific encoder models adapted to cardiology through continued pretraining on monolingual biomedical and cardiology-related corpora using Masked Language Modeling (MLM). The family covers Czech, Dutch, English, Italian, Romanian, Spanish and Swedish.
24
+
25
+ ## Training
26
+
27
+ | | |
28
+ |---|---|
29
+ | Language | English (`en`) |
30
+ | Triplet collection | `enriched` |
31
+ | Strategy | `grandparents` |
32
+ | Objective | Multi-Similarity Loss |
33
+ | Mining | All triplets, margin 0.2 |
34
+ | Pooling | CLS |
35
+ | Epochs | 1 |
36
+ | Batch size | 256 |
37
+ | Learning rate | 2e-5 |
38
+ | Max. length | 25 |
39
+
40
+ CUI-supervised terminology pairs enriched with grandparent-level ontology relations.
41
+
42
+ ### Terminology statistics
43
+
44
+ | Strategy | Triplets | CUIs | Unique terms | Unique positives | Terms/CUI | Δ terms |
45
+ |---|---:|---:|---:|---:|---:|---:|
46
+ | synonyms | 83,914 | 83,914 | 165,661 | 83,472 | 2.00 | 0 |
47
+ | parents | 1,699,553 | 477,290 | 550,651 | 432,552 | 4.03 | +384,990 |
48
+ | grandparents | 4,952,020 | 477,293 | 550,651 | 485,607 | 10.06 | +384,990 |
49
+
50
+ This model uses **4,952,020 triplets**, covering **477,293 CUIs** and **550,651 unique normalized terms**.
51
+
52
+ The training terminology is **not distributed** with this repository because it contains resources subject to UMLS licensing conditions. Only aggregate statistics are released.
53
+
54
+ ## Intended use
55
+
56
+ The model is intended for terminology embedding, biomedical candidate retrieval, concept normalization and entity linking, particularly in cardiology and clinical NLP pipelines. It is not intended for direct clinical decision-making.
57
+
58
+ ## Usage
59
+
60
+ ```python
61
+ import torch
62
+ import torch.nn.functional as F
63
+ from transformers import AutoModel, AutoTokenizer
64
+
65
+ model_id = "DT4H/DT4H_CardioBERTa_grandparents_en_enriched"
66
+
67
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
68
+ model = AutoModel.from_pretrained(model_id)
69
+
70
+ inputs = tokenizer(
71
+ "clinical concept",
72
+ return_tensors="pt",
73
+ truncation=True,
74
+ max_length=25,
75
+ )
76
+
77
+ with torch.no_grad():
78
+ output = model(**inputs)
79
+
80
+ embedding = F.normalize(
81
+ output.last_hidden_state[:, 0, :],
82
+ p=2,
83
+ dim=1,
84
+ )
85
+ ```
86
+
87
+ ## Reference
88
+
89
+ **Danu et al.** *CardioLM - a multilingual suite of small language models for the cardiology domain*.
90
+
91
+ Developed within the **DataTools4Heart (DT4H)** project, Grant Agreement 101057849.
config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/mnt/home/users/gaias_001_uma/fgd/SapBERT/models/baseline/CardioBERTa.en",
3
+ "architectures": [
4
+ "RobertaModel"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "roberta",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 12,
20
+ "pad_token_id": 1,
21
+ "position_embedding_type": "absolute",
22
+ "torch_dtype": "float32",
23
+ "transformers_version": "4.48.0",
24
+ "type_vocab_size": 1,
25
+ "use_cache": true,
26
+ "vocab_size": 50265
27
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f26a53fd540e02f4c19a152dd6b30199a7c020dc69d585be949a07d08e68781
3
+ size 498604904
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": false,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "extra_special_tokens": {},
51
+ "mask_token": "<mask>",
52
+ "max_len": 512,
53
+ "model_max_length": 512,
54
+ "pad_token": "<pad>",
55
+ "sep_token": "</s>",
56
+ "tokenizer_class": "RobertaTokenizer",
57
+ "trim_offsets": true,
58
+ "unk_token": "<unk>"
59
+ }
training_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "language": "en",
3
+ "strategy": "grandparents",
4
+ "triplets_type": "enriched",
5
+ "backbone": "DT4H/CardioBERTa.en",
6
+ "training": {
7
+ "objective": "Multi-Similarity Loss",
8
+ "pairwise": true,
9
+ "miner": {
10
+ "enabled": true,
11
+ "type_of_triplets": "all",
12
+ "margin": 0.2
13
+ },
14
+ "pooling": "cls",
15
+ "epochs": 1,
16
+ "batch_size": 256,
17
+ "learning_rate": 2e-05,
18
+ "max_length": 25,
19
+ "random_seed": 33
20
+ },
21
+ "training_data_availability": "Training terminology is not redistributed because it contains resources subject to UMLS licensing conditions."
22
+ }
training_data_stats.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "triplets": 4952020,
3
+ "malformed_rows": 0,
4
+ "unique_triplets": 4951542,
5
+ "duplicate_triplets": 478,
6
+ "unique_cuis": 477293,
7
+ "unique_anchors": 467097,
8
+ "unique_positives": 485607,
9
+ "unique_terms": 550651,
10
+ "unique_anchor_positive_pairs": 4915749,
11
+ "ambiguous_terms_across_cuis": 199127,
12
+ "self_pairs_exact": 0,
13
+ "self_pairs_normalized": 40,
14
+ "cuis_with_multiple_terms": 477293,
15
+ "terms_per_cui_mean": 10.059801840798,
16
+ "terms_per_cui_median": 7,
17
+ "terms_per_cui_p95": 28.0,
18
+ "terms_per_cui_max": 37,
19
+ "triplets_per_cui_mean": 10.375220252549273,
20
+ "triplets_per_cui_median": 7,
21
+ "triplets_per_cui_p95": 33.0,
22
+ "triplets_per_cui_max": 35,
23
+ "anchor_words_mean": 4.16947488095767,
24
+ "positive_words_mean": 4.407634056405265,
25
+ "cuis_common_with_synonyms": 83914,
26
+ "cuis_added_vs_synonyms": 393379,
27
+ "cuis_missing_vs_synonyms": 0,
28
+ "terms_common_with_synonyms": 165661,
29
+ "terms_added_vs_synonyms": 384990,
30
+ "terms_missing_vs_synonyms": 0
31
+ }
triplet_comparison.csv ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ strategy,triplets,unique_triplets,unique_cuis,unique_anchors,unique_positives,unique_terms,unique_anchor_positive_pairs,terms_per_cui_mean,terms_per_cui_median,cuis_added_vs_synonyms,cuis_missing_vs_synonyms,terms_added_vs_synonyms,terms_missing_vs_synonyms
2
+ synonyms,83914,83914,83914,83292,83472,165661,83798,1.9995709893462355,2.0,0,0,0,0
3
+ parents,1699553,1699259,477290,397180,432552,550651,1687484,4.028630392423893,3.0,393376,0,384990,0
4
+ grandparents,4952020,4951542,477293,467097,485607,550651,4915749,10.059801840798,7,393379,0,384990,0
vocab.json ADDED
The diff for this file is too large to render. See raw diff