Dmitry43243242/icd10-ru-subgroup-e

ICD-10 subgroup classifier - group E (distilled specialist)

Multi-label classifier over 3-character ICD-10 subgroups inside chapter E. This specialist was distilled from local BERT teacher models into alexyalunin/RuBioBERT. Teacher weights are not uploaded to Hugging Face.

Intended use / Назначение

• EN: Decision-support signal for suggesting candidate ICD-10 subgroups from Russian clinical notes. Not a substitute for clinician judgment; not validated for autonomous diagnosis.
• RU: Вспомогательный сигнал для предложения кандидатных 3-символьных кодов МКБ-10 по русскому клиническому тексту. Не заменяет врача и не предназначен для автономных клинических решений.

Training data / Обучающие данные

• Source CSV: datasets/subgroups/group_E.csv
• SHA-256: 7bd98fc0eea937b8edf1391e86ca15afd2aed5c98996951f822684805713ed0b
• Splits: train=919 · val=200 · test=199
• Labels: 45; rare/interface-only ids are listed in label_map.json.

Training route

• Approach: local_teacher_ensemble_knowledge_distillation
• Base model: alexyalunin/RuBioBERT
• Direct validation hit@3: 0.835
• No-distillation threshold: 0.9
• Teacher models (fallback KD only): ['alexyalunin/RuBioRoBERTa', 'ai-forever/ruBert-base', 'DeepPavlov/rubert-base-cased']
• Selected KD config (fallback only): temperature=2.0, hard_loss_weight=0.5

Metrics (test split)

metric	final specialist	teacher ensemble / fallback
macro_f1	0.6122	0.6805
micro_f1	0.5606	0.6541
weighted_f1	0.5830	0.6549
subset_accuracy	0.2965	0.4824
hit@1	0.6482	0.6784
hit@3	0.8241	0.8442
recall@3	0.8147	0.8382
mrr	0.7516	0.7735

Full per-label breakdown is available in metrics.json.

Inference

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

repo = "Dmitry43243242/icd10-ru-subgroup-e"
tok = AutoTokenizer.from_pretrained(repo)
mdl = AutoModelForSequenceClassification.from_pretrained(repo)
mdl.eval()

text = "жалобы пациента..."
inp = tok(text, return_tensors="pt", truncation=True, max_length=512)
with torch.no_grad():
    probs = torch.sigmoid(mdl(**inp).logits)[0]
preds = [mdl.config.id2label[i] for i, p in enumerate(probs.tolist()) if p >= 0.5]
top5 = sorted(
    [(mdl.config.id2label[i], p) for i, p in enumerate(probs.tolist())],
    key=lambda x: -x[1],
)[:5]
print(preds, top5)