Update README.md
Browse files
README.md
CHANGED
@@ -11,7 +11,18 @@ size_categories:
|
|
11 |
# Dataset Card for BTX24/tekno21-brain-stroke-dataset-binary
|
12 |
|
13 |
## Dataset Description
|
14 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
|
16 |
## Dataset Creation
|
17 |
Contains stroke images from TEKNOFEST 2021
|
|
|
11 |
# Dataset Card for BTX24/tekno21-brain-stroke-dataset-binary
|
12 |
|
13 |
## Dataset Description
|
14 |
+
|
15 |
+
📌 **EN:**
|
16 |
+
The **TEKNO21 Brain Stroke Detection Model** is a deep learning system trained on the **TEKNOFEST-2021 Brain Stroke Dataset**, containing **7,369 brain CT images** labeled by expert radiologists. Using **YOLOv8, U-Net, and DeepLabv3**, it classifies CT scans as **Stroke Present** or **Stroke Absent** with a **0.5 confidence threshold**. Developed for research, not diagnosis, it assists radiologists in stroke detection. The dataset was curated under **TÜSEB and the Turkish Ministry of Health**, with access requiring confirmation.
|
17 |
+
|
18 |
+
🔗 **Dataset:** [TEKNOFEST-2021 Stroke Dataset](https://acikveri.saglik.gov.tr/Home/DataSetDetail/1)
|
19 |
+
|
20 |
+
📌 **TR:**
|
21 |
+
**TEKNO21 Beyin İnme Algılama Modeli**, **TEKNOFEST-2021 Beyin İnme Veri Seti** ile eğitilmiş, **7.369 beyin BT görüntüsüne** dayalı derin öğrenme modelidir. **YOLOv8, U-Net ve DeepLabv3** kullanarak BT kesitlerini **İnme Var / İnme Yok** olarak sınıflandırır (**0.5 güven skoru**). Tanı koyma amacı taşımaz, radyologlara destek sağlar. **TÜSEB ve Sağlık Bakanlığı** tarafından hazırlanan veri setine erişim için yetkili onay gereklidir.
|
22 |
+
|
23 |
+
🔗 **Veri Seti:** [TEKNOFEST-2021 İnme Veri Seti](https://acikveri.saglik.gov.tr/Home/DataSetDetail/1)
|
24 |
+
|
25 |
+
|
26 |
|
27 |
## Dataset Creation
|
28 |
Contains stroke images from TEKNOFEST 2021
|