docs: homogenize README

Now comming with only 4 sections: source, environment, references and license.

Author: ktsiknos-brainchip

docs/images/sample_0.png

@@ -1,13 +1,9 @@
 # ECG Anomaly Detection with Quantized Autoencoder
-This project implements an anomaly detector for ECG data using a convolutional autoencoder, evaluated in both float and quantized (8-bit) form and converted to SNN.
 
 ## Source
-This model uses [**LeNet**](https://huggingface.co/mindspore-ai/LeNet).
+This project implements an anomaly detector for ECG data using a convolutional autoencoder, evaluated in both float and quantized (8-bit) form and converted to SNN.
 
----
 ## Environment
-The following dependencies are required to build and run this model:
-
 ```
 matplotlib==3.8.4
 seaborn==0.13.2
@@ -18,3 +14,9 @@ tensorflow==2.15.0
 cnn2snn==2.13.0
 quantizeml==0.16.0
 ```
+
+## References
+- This model uses [**LeNet**](https://huggingface.co/mindspore-ai/LeNet).
+
+## License
+Apache 2.0 License

docs/images/sample_1.png

@@ -1,19 +1,11 @@
 
 # ECG Arrhythmia Classification with 2D CNN and Quantization
 
+## Source
 This project performs ECG heartbeat classification (4 classes) using a compact 2D CNN on the [MIT-BIH Arrhythmia Database](https://physionet.org/content/mitdb/1.0.0/). The model is then post-training quantized and converted to an Akida-compatible spiking neural network (SNN), followed by final evaluation of the SNN model.
 Classifying irregular heartbeats is critical for identifying their origin and assessing the potential severity of arrhythmias, which directly impacts diagnosis and treatment. Key discriminative features lie in **QRS morphology and timing** of the beats, making **CNNs**  well-suited for this task due to their ability to efficiently capture local temporal patterns.
 
-
-## Source
-This model uses [**LeNet**](https://huggingface.co/mindspore-ai/LeNet).
-
-
----
-
 ## Environment
-The following dependencies are required to build and run this model:
-
 ```
 matplotlib==3.8.4
 numpy==1.26.4
@@ -25,3 +17,9 @@ wfdb==4.3.0
 cnn2snn==2.13.0
 quantizeml==0.16.0
 ```
+
+## References
+- This model uses [**LeNet**](https://huggingface.co/mindspore-ai/LeNet).
+
+## License
+Apache 2.0 License