--- a
+++ b/README.md
@@ -0,0 +1,88 @@
+<h1>Norwegian Endurance Athlete ECG Database</h1>
+
+<p><strong>Bjørn-Jostein Singstad</strong></p>
+<p><strong>Published:</strong> March 23, 2022. Version: 1.0.0</p>
+
+<h2>Citation</h2>
+<p>Singstad, B. (2022). Norwegian Endurance Athlete ECG Database (version 1.0.0). PhysioNet. <a href="https://doi.org/10.13026/qpjf-gk87" target="_blank">https://doi.org/10.13026/qpjf-gk87</a>.</p>
+<p>Please include the standard citation for PhysioNet: Goldberger, A., et al. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Circulation, 101(23), pp. e215–e220.</p>
+
+<h2>Abstract</h2>
+<p>
+The Norwegian Endurance Athlete ECG Database contains 12-lead ECG recordings from 28 elite athletes from various sports in Norway. All recordings are 10 seconds resting ECGs recorded with a GE MAC VUE 360 electrocardiograph and interpreted using both the Marquette SL12 algorithm and a trained cardiologist. Data was collected at the University of Oslo in early 2020.
+</p>
+
+<h2>Background</h2>
+<p>
+Athletic training induces cardiac remodeling, such as increased ventricular wall thickness and chamber size. These changes are healthy adaptations but may mimic pathologies associated with sudden cardiac death (SCD). ECG is a valuable tool in detecting such abnormalities, but accurate interpretation in athletes is challenging. This dataset aims to support research into improving athlete ECG diagnostics.
+</p>
+
+<h2>Methods</h2>
+<p>
+The study involved 28 healthy athletes (19 men, 9 women, aged 20–43, mean 25 years). Most were rowers (86%), with a few kayakers and cyclists. Training hours averaged around 800 annually from 2017–2019. Standard 10-second resting ECGs were recorded using GE MAC VUE 360. ECGs were interpreted using the GE Marquette SL12 algorithm and a cardiologist following international athlete ECG criteria.
+</p>
+<p>
+The study was approved by relevant ethics committees and all participants gave informed consent, including permission to share data publicly.
+</p>
+
+<h2>Data Description</h2>
+<p>
+Each of the 28 ECG waveform files includes 12 arrays (one per lead), stored in <code>.dat</code> and <code>.hea</code> formats. These files are compatible with Python's WFDB package. All signals are sampled at 500Hz for 10 seconds (5000 samples).
+</p>
+
+<pre><code>
+ath_001 12 500 5000
+ath_001.dat 16 50000/mV ... I
+ath_001.dat 16 50000/mV ... II
+...
+#SL12: sinus bradycardia with marked sinus arrhythmia, Right Axis Deviation, Borderline ECG
+#C: Sinus arrhythmia, Normal ECG
+</code></pre>
+
+<h2>Usage Notes</h2>
+<p>
+This dataset is intended to support development of ECG analysis algorithms, especially for athletes. It is unique in that ECGs are annotated by both a cardiologist and a commercial algorithm.
+</p>
+
+<p>Example Python code to load ECGs using WFDB:</p>
+
+<pre><code>import wfdb
+import numpy as np
+import os
+
+directory = "./your/directory/"
+ECGs = []
+for ecgfilename in sorted(os.listdir(directory)):
+    if ecgfilename.endswith(".dat"):
+        ecg = wfdb.rdsamp(directory + ecgfilename.split(".")[0])
+        ECGs.append(ecg)
+ECGs = np.asarray(ECGs)
+</code></pre>
+
+<p>
+Note: Although subjects are elite athletes, no echocardiographic examinations were performed to confirm structural heart remodeling.
+</p>
+
+<h2>Release Notes</h2>
+<ul>
+  <li>1.0.0 – Initial release of the dataset.</li>
+</ul>
+
+<h2>Ethics</h2>
+<p>The authors declare no ethics concerns.</p>
+
+<h2>Acknowledgements</h2>
+<p>
+Thanks to Professor Emeritus Knut Gjessdal for medical expertise and ECG interpretation. The work was conducted at the University of Oslo with support from Professor Ørjan Grrøttem Martinsen.
+</p>
+
+<h2>Conflicts of Interest</h2>
+<p>No conflicts of interest declared.</p>
+
+<h2>References</h2>
+<ol>
+  <li>Fagard R. (2003). Athlete's heart. Heart, 89(12), 1455–1461. <a href="https://doi.org/10.1136/heart.89.12.1455" target="_blank">Link</a></li>
+  <li>Sharma, S. et al. (2015). Exercise and the heart. Eur Heart J, 36(23), 1445–1453. <a href="https://doi.org/10.1093/eurheartj/ehv090" target="_blank">Link</a></li>
+  <li>Corrado, D. et al. (2006). Trends in sudden cardiovascular death. JAMA, 296(13), 1593–1601. <a href="https://doi.org/10.1001/jama.296.13.1593" target="_blank">Link</a></li>
+  <li>Marijon, E. et al. (2011). Sports-related sudden death. Circulation, 124(6), 672–681. <a href="https://doi.org/10.1161/CIRCULATIONAHA.110.008979" target="_blank">Link</a></li>
+</ol>