MammoBiradsScore_encoderfile

🩺 Mammo BIRADS Score Classifier (.encoderfile)

This repository contains a standalone, zero-dependency executable binary (.encoderfile) built with Mozilla AI's Encoderfile specification. It is fine-tuned from ahmetcangunay/MammoBiradsScore and designed for fast, CPU-optimized Turkish mammography report text classification into BIRADS (1-5) scores.

🏆 TEKNOFEST Achievement: Developed as part of the project that achieved 5th Place & Finalist status in the TEKNOFEST 2024 Sağlıkta Yapay Zekâ Yarışması (Üniversite ve Üzeri Seviyesi - Bilgisayarlı Görüyle Hastalık Tespiti Kategorisi).

---

🚀 Quick Start / Instant CLI Usage

You can download and run this executable directly on any x86_64 Linux machine without installing Python, PyTorch, Transformers, or SpaCy.

1. Download & Prepare Binary

# Download the executable via Hugging Face CLI
hf download ahmetcangunay/MammoBiradsScore_encoderfile mammo-birads-classifier.x86_64-unknown-linux-gnu.encoderfile --local-dir .

# Make it executable
chmod +x mammo-birads-classifier.x86_64-unknown-linux-gnu.encoderfile

2. Direct CLI Inference (infer)

./mammo-birads-classifier.x86_64-unknown-linux-gnu.encoderfile infer "BILATERAL MAMOGRAFI INCELEMESI: Sağ meme üst dış kadranda yaklaşık 1 cm çapında düzgün sınırlı nodüler lezyon izlendi."

Example Response:

{
  "results": [
    {
      "logits": [
        -1.8540025,
        -0.78142524,
        2.4493582,
        0.46748734,
        -0.26670128
      ],
      "scores": [
        -1.8540025,
        -0.78142524,
        2.4493582,
        0.46748734,
        -0.26670128
      ],
      "predicted_index": 2,
      "predicted_label": "BIRADS3"
    }
  ],
  "model_id": "mammo-birads-classifier",
  "metadata": {}
}

🌐 Serving Options

REST API Server (serve)

Start a lightweight local REST server:

./mammo-birads-classifier.x86_64-unknown-linux-gnu.encoderfile serve --http-port 8080

Anthropic MCP Server (mcp)

Start as a Model Context Protocol (MCP) server for integration with LLM agents:

./mammo-birads-classifier.x86_64-unknown-linux-gnu.encoderfile mcp

📊 Model Performance & Metrics

Evaluated on an independent test dataset of 200 clinical mammography report snippets:

• Overall Accuracy: 94.00%

• Weighted F1 Score: 0.9377

Classification Report

Medical Safety Note: Critical categories (BIRADS 4 & 5) exhibit a 100% Recall rate, minimizing False Negatives for high-risk clinical findings.

⚙️ Technical Details

• Base Architecture: ahmetcangunay/MammoBiradsScore (Fine-tuned from savasy/bert-turkish-text-classification)

• Runtime: CPU-bound PyTorch C++ / Standalone Rust Bindings

• Output Classes: BIRADS1, BIRADS2, BIRADS3, BIRADS4, BIRADS5

• Target OS: x86_64-unknown-linux-gnu