Update README.md

README.md

@@ -3,25 +3,21 @@ title: OrthoTimes QuickCephTool - Ceph Landmark Detection
 emoji: 🦷
 colorFrom: blue
 colorTo: green
-sdk: gradio
-sdk_version: 4.44.0
-app_file: app.py
+sdk: docker
+app_port: 7860
 pinned: false
 license: mit
 ---
 
 # OrthoTimes QuickCephTool — HRNet Landmark Detection
 
-Automatic cephalometric landmark detection using **HRNet-W32** pretrained on lateral cephalometric radiographs.
+Automatic cephalometric landmark detection using HRNet-W32.
 
-## Model
-- Source: [cwlachap/hrnet-cephalometric-landmark-detection](https://huggingface.co/cwlachap/hrnet-cephalometric-landmark-detection)
-- MRE: ~1.2–1.6 mm
-- SDR@2mm: ~80–85%
-- Detects 19 landmarks
+## API
 
-## Landmarks detected
-S, N, Or, Po, ANS, PNS, A, U1tip, L1tip, B, Pog, Me, Gn, Go, Co, L1ap, U1ap, U6, L6
+POST /detect
+```json
+{"image_b64": "<base64 encoded image>"}
+```
 
-## API
-Used by OrthoTimes QuickCephTool HTML application for automatic landmark placement on upload.
+Returns normalised (0-1) coordinates for 19 landmarks.