feat: filtrage des DAS parasites (artefacts OCR trackare)

Nouveau module das_filter.py avec 7 règles de rejet (trop court, chiffres,
lettre+chiffres OCR, mots concaténés/répétés, fragments non-médicaux) +
nettoyage newlines/ponctuation. Filtrage appliqué aux 3 sources de DAS :
trackare, regex et edsnlp. 31 tests unitaires.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
dom
2026-02-11 17:48:25 +01:00
parent 96ccb4850f
commit 31c29078a1
4 changed files with 172 additions and 3 deletions

106
tests/test_das_filter.py Normal file
View File

@@ -0,0 +1,106 @@
"""Tests unitaires pour le filtre de DAS parasites."""
import pytest
from src.medical.das_filter import clean_diagnostic_text, is_valid_diagnostic_text
class TestCleanDiagnosticText:
def test_removes_trailing_punctuation(self):
assert clean_diagnostic_text("Thrombopénie,") == "Thrombopénie"
def test_removes_trailing_semicolon(self):
assert clean_diagnostic_text("HTA;") == "HTA"
def test_replaces_newlines(self):
assert clean_diagnostic_text("Insuffisance rénale\naigue") == "Insuffisance rénale aigue"
def test_strips_whitespace(self):
assert clean_diagnostic_text(" HTA ") == "HTA"
def test_collapses_multiple_spaces(self):
assert clean_diagnostic_text("Insuffisance rénale aigue") == "Insuffisance rénale aigue"
def test_combined_cleanup(self):
assert clean_diagnostic_text(" Anticoagulant\nanticoagulant, ") == "Anticoagulant anticoagulant"
class TestIsValidDiagnosticText:
# --- Rejets ---
def test_reject_empty(self):
assert not is_valid_diagnostic_text("")
def test_reject_too_short(self):
assert not is_valid_diagnostic_text("Ab")
def test_reject_digits_only(self):
assert not is_valid_diagnostic_text("666666666666664")
def test_reject_mostly_digits(self):
assert not is_valid_diagnostic_text("12345abc")
def test_reject_letter_space_digits(self):
assert not is_valid_diagnostic_text("H 51")
def test_reject_letter_space_digits_a_accent(self):
assert not is_valid_diagnostic_text("À 08")
def test_reject_letter_newline_digits(self):
# Après clean, "H\n10" devient "H 10"
assert not is_valid_diagnostic_text("H 10")
def test_reject_concatenated_words(self):
assert not is_valid_diagnostic_text("Ventilationventilation")
def test_reject_concatenated_words_long(self):
assert not is_valid_diagnostic_text("ventilationventilationventilation")
def test_reject_repeated_words(self):
assert not is_valid_diagnostic_text("Spontanée spontanée spontanée spontanée")
def test_reject_repeated_words_three(self):
assert not is_valid_diagnostic_text("oui oui oui")
def test_reject_fragment_de(self):
assert not is_valid_diagnostic_text("De laboratoire")
def test_reject_fragment_du(self):
assert not is_valid_diagnostic_text("Du sang")
def test_reject_fragment_des(self):
assert not is_valid_diagnostic_text("Des résultats")
def test_reject_bracket_fragment(self):
assert not is_valid_diagnostic_text("] de laboratoire")
def test_reject_isolement(self):
assert not is_valid_diagnostic_text("Isolement")
def test_reject_pp_marker(self):
assert not is_valid_diagnostic_text("Pp 500")
# --- Acceptations ---
def test_accept_hta(self):
assert is_valid_diagnostic_text("HTA")
def test_accept_cholecystite(self):
assert is_valid_diagnostic_text("Cholécystite aiguë")
def test_accept_lithiase(self):
assert is_valid_diagnostic_text("Lithiase vésiculaire")
def test_accept_insuffisance_renale(self):
assert is_valid_diagnostic_text("Insuffisance rénale aigue")
def test_accept_obesite_with_imc(self):
assert is_valid_diagnostic_text("Obésité (IMC 35.251)")
def test_accept_short_valid(self):
# 3 chars = seuil exact, doit passer
assert is_valid_diagnostic_text("HTA")
def test_accept_diabete(self):
assert is_valid_diagnostic_text("Diabète de type 2")
def test_accept_sepsis(self):
assert is_valid_diagnostic_text("Sepsis sévère")