ClarusC64's picture
Update README.md
ef05fed verified
metadata
license: mit
language:
  - en
pretty_name: ABX-HT-007 Allergic Sensitization Pattern
tags:
  - antibiotics
  - toxicity
  - allergy
  - hypersensitivity
  - immune
  - eosinophils
  - tryptase
  - tabular
task_categories:
  - tabular-classification
size_categories:
  - n<1k

ABX-HT-007 Allergic Sensitization Pattern

Purpose

Detect early allergic sensitization when immune response markers stop tracking treatment duration before hypersensitivity events.

Core pattern

  • host_stress_index high
  • duration_exposure_index high
  • immune_coherence_index drops
  • eos_rise_vs_expected or tryptase_rise_vs_expected stays high
  • later_hypersensitivity_flag appears

Files

  • data/train.csv
  • data/test.csv
  • scorer.py

Schema

Each row is one timepoint in a within series treatment course.

Required columns

  • row_id
  • series_id
  • treatment_day
  • host_model
  • drug
  • duration_exposure_index
  • total_dose_mg
  • expected_total_dose_mg
  • dose_deviation
  • eos_percent
  • expected_eos_percent
  • eos_rise_vs_expected
  • tryptase_ng_mL
  • expected_tryptase_ng_mL
  • tryptase_rise_vs_expected
  • immune_coherence_index
  • host_stress_index
  • later_hypersensitivity_flag
  • assay_method
  • source_type
  • allergic_sensitization_signal
  • earliest_allergic_sensitization

Labels

  • allergic_sensitization_signal

    • 1 for rows at or after first confirmed sensitization onset
  • earliest_allergic_sensitization

    • 1 only for the first onset row in that series

Scorer logic in v1

  • exclude series with high baseline eosinophils
  • candidate onset point
    • host_stress_index at least 0.80
    • duration_exposure_index at least 0.80
    • immune_coherence_index at most 0.40
    • eos_rise_vs_expected at least 3.0 or tryptase_rise_vs_expected at least 5.0
    • for two consecutive points
    • ignore one point eosinophil spike then recovery artifacts
  • confirmation
    • later_hypersensitivity_flag equals 1 later in series

Evaluation

Run

  • python scorer.py --path data/test.csv