XTRACTIS FOR PREDICTIVE MEDECINE

ANATOMOPATHOLOGICAL DIAGNOSIS OF BREAST CANCER

Benchmark vs. Random Forests, Boosted Trees & Neural Networks

How to make an automated —yet transparent— medical diagnosis of breast cancer from microscopic images of patient mammary cells?

Goals & benefits

Identify the cellular characteristics involved in cancer and enhance medical knowledge by helping pathologists and oncologists understand the causal relationships between specific cell features, their combination, and the presence of cancer.

Help the medical profession to make earlier and more personalized decisions through rapid, systematic, and explainable diagnoses.

Contribute to improving patient care (pain, survival, duration of treatment) and extend access to high-level diagnoses even in medical deserts

XTRACTIS RESULTS

Predictive Model that is:

Intelligible.

A Decision System composed of 7 unchained gradual rules, each using some of the 13 variables that XTRACTIS identified as significant including weak signals.

Robust.

Excellent performance on External Test.

Efficient & Operational.

Running in real-time up to 70,000 predictions per second (i7, 8 physical cores, 2.5GHz), offline or online (API).

Summary of Use Case 04/2022 (v1.2)

Results by
XTRACTIS® GENERATE 12.1.40851 (02/2022)

DOCUMENT CONTENTS

  1. Problem Definition
  2. Xtractis Solution
  3. Top-Model Induction
  4. Predictions for 2 cases
  5. Top-Models Benchmark